ERCBflare User Guide for version 2.00

ERCBflare User Guide for version 2.00
ERCBflare
FT
A Model for Temporary Flaring Permits,
Non-Routine Flaring and Routine
Flaring Air Dispersion Modelling for
Sour Gas Facilities
D
R
A
User Guide, ERCBflare Version 2.00
ERCBflare
FT
A Model for Temporary Flaring Permits,
Non-Routine Flaring and Routine
Flaring Air Dispersion Modelling for
Sour Gas Facilities
A
User Guide, ERCBflare Version 2.00
A Report Prepared for
R
Energy Resources Conservation Board (ERCB)
D
Funded in Part by
Petroleum Technology Alliance of Canada (PTAC)
Alberta Upstream Petroleum Research Fund (AUPRF)
Energy Resources Conservation Board (ERCB)
8 May 2013
ERCBflare was developed for
Energy Resources Conservation Board (ERCB) of Alberta, Canada
by:
Brian W. Zelt, Ph.D., P.Eng.
Zelt Professional Services Inc.
Please Reference this document as:
FT
Michael J. Zelensky, M.Sc., P.Eng.
Energy Resources Conservation Board
Energy Resources Conservation Board (ERCB). 2012.
ERCBflare User Guide: A Model for Temporary Flaring Permits, Non-Routine Flaring and Routine
Flaring Air Dispersion Modelling for Sour Gas Facilities, Version 2
Distributed by
A
8 May 2013
D
R
Energy Resources Conservation Board
Centennial Place
Suite 1000, 250 – 5 Street SW
Calgary, Alberta T2P 0R4
Telephone: 403-297-2625
Fax: 403-297-3520
Web site: http://www.ercb.ca/regulations-and-directives/directives/directive060
Liability Disclaimer
These models, software and documentation were prepared by the ERCB (Energy Resources
Conservation Board), to the specifications set by ERCB. ERCB warrants that the SOFTWARE
will perform substantially in accordance with the provided documentation. ERCB, nor any person
acting on their behalf, makes any warranty, guarantee, or representation, expressed or implied, that
the software and related materials, without limitation, are free of errors, are consistent with any
standard of merchantability or will meet user's requirements for a particular application, that any
calculations performed by the software are correct or accurate, that the software is compatible with
particular computer systems, computer peripherals and other software packages, or that the
software will run uninterrupted.
FT
EXCEPT AS PROVIDED ABOVE, THIS DOCUMENTATION AND SOFTWARE IS LICENSED "AS IS"
WITHOUT WARRANTY AS TO ITS PERFORMANCE. NO OTHER WARRANTIES APPLY. ERCB
DISCLAIMS ALL OTHER WARRANTIES, EITHER EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED
TO IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.
THIS LIMITED WARRANTY GIVES YOU SPECIFIC LEGAL RIGHTS. YOU MAY HAVE OTHERS, WHICH
VARY FROM PROVINCE TO PROVINCE.
D
R
A
IN NO EVENT SHALL ERCB, OR THEIR SUPPLIERS BE LIABLE FOR ANY DAMAGES WHATSOEVER
INCLUDING, WITHOUT LIMITATION, DAMAGES FOR LOSS OF BUSINESS PROFITS, BUSINESS
INTERRUPTION, LOSS OF BUSINESS INFORMATION, CONSEQUENTIAL DAMAGES, OR OTHER
PECUNIARY LOSS, HOWSOEVER ARISING OUT OF THE USE, ATTEMPTED USE OF OR INABILITY TO
USE THIS PRODUCT, EVEN IF ERCB HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.
Minimum System Requirements for ERCBflare
ERCBflare is a Windows® based software
application requiring Microsoft® Excel as a user
interface. The minimum system requirements* are:
•
•
FT
•
•
•
•
•
•
Windows®8; Windows®7; Vista; Windows XP
Microsoft® Excel. (Program testing completed on
Office 2007, 2010 and 2013)
x64 or x86
1024x768 minimum resolution recommended
2 GHz processor minimum recommended
4 GB RAM minimum recommended
Windows® - compatible mouse
Optional: colour or b/w printer (300 dpi or better
recommended)
D
R
A
** Users may require administrative rights in order to
run the install package.
TABLE OF CONTENTS
1.
Introduction .......................................................... 1
Model Changes ............................................................................................................. 2
Version 1.x to 2.00
2
User Qualifications ....................................................................................................... 6
Suitable Technical Background
6
How this Document is Organized ............................................................................... 7
About this Guide ........................................................................................................... 8
Where to Go for Help .................................................................................................... 8
What is ERCBflare all about? ...................................................................................... 8
INSTALLATION AND SETUP................................. 11
FT
2.
Overview of the Installation Process ........................................................................ 11
SETUP .......................................................................................................................... 13
3.
ERCBflare INPUTS .............................................. 17
Introduction ................................................................................................................. 17
Graphical User-Interface ............................................................................................ 18
Overview – The ERCBflare GUI in Excel
18
Stand-Alone
Central Database
18
19
A
The ERCBflare Excel Pages
20
iSTART ......................................................................................................................... 22
Fundamentals
22
Permit or Evaluation Mode
Flaring Classification
Hide Technical Pages
Hide iBATCH Page
Hide iBIN page
22
23
25
25
25
R
Advanced Switches
D
MRUNBAT
MPAUSE
MKEEP
MKEEPOUT
MRELLOC
MEXPREC
MSCREEN
MELEV
MDEMGET
MLCCGET
25
26
26
26
27
27
27
27
28
28
28
Non-Default Settings
Ambient Temperature
Average Ambient Wind Speed
PG Stability Class
Qmin Fraction of Qmax
SO2 1h Air Quality Objective
H2S 1h Air Quality Objective
Receptor Resolution for Maximum Concentration
Raw, Fuel and Lift Gas Temperature Before Combustion
Run Flags
User Period Selection (Annual or Month)
Transient Blowdown distribution of mass option
Blowdown User Entry of Qmax
Blowdown User Entry of Qtotal
Land Use around the Well
28
29
30
30
30
31
31
31
31
32
33
33
33
33
34
iFACILITY ..................................................................................................................... 34
Administrative-Operator
Administrative-Consultant
TABLE OF CONTENTS
34
35
i
Well or Facility Details
35
35
35
35
36
36
37
37
37
37
38
38
39
39
43
43
44
44
44
44
44
FT
Well Name
Licence Number
Unique Well Identifier (UWI)
Oil or Gas
Critical Well Classification
Formation(s) and Zone(s) to be Tested
Number of Zones to be Tested
Lahee Classification
Surface Location
Mapping Projection
Surface Coordinates of Flare
Flare Base Elevation
Land-use Characterization
Operations to be Conducted
Total Volume of Raw Gas to be Flared
Total Estimated Days with Flaring for ALL Zones
Well tied into Production Facilities
Information on the Feasibility of In-Line Well Testing Attached
Previous Flaring/Incineration Permit
Requested Approval Start Date/End Date
Emergency Response Details
H2S Release Rate
Emergency Response Planning Zone Distance
Is an ERP Required?
ERP Reference Number
Fluid Production Details
Anticipated Fluid Production
Gas to Fluid Ratio
45
45
45
45
46
46
46
46
iFLARING ..................................................................................................................... 47
Flaring Details
47
A
Subject Zone/Source
Scenario Name
Time of Year to Model the Flare
Flare Stack Tip Exit Height
Flare Stack Tip Exit Diameter
Requested Maximum Raw Gas H2S Concentration
48
48
48
49
49
49
Continuous/Steady or Transient
Continuous or Short-Term Steady
R
Maximum Raw Gas Flow Rate
Average Flow Rate Parameters
49
50
50
50
Transient Source
Fuel Gas
Lift Gas
Flare Assist
D
Steam Assist
Air Assist
Flare Assist Flammability Check
51
51
52
52
53
54
54
Gas Compositions
Gas Composition Total
Gas Analysis Reference
54
55
56
iTERRAIN ..................................................................................................................... 56
Topographical Map Details and Maximum Terrain
56
1:50,000 Topographical Map(s)
57
Contour Interval Units of Map
57
Contour Interval of Map With Maximum Elevation
57
Is Maximum Elevation A Surveyed Peak?
57
Maximum Contour Elevation or Surveyed Peak Elevation
58
Map Distance from Flare to Centre of Maximum Contour or Maximum Surveyed Peak
Elevation
58
Worst Case Terrain and Complex Terrain Criteria Parameters
Complex Terrain Summary
58
60
Get DEM Data – BUTTON
60
Load XYZ DEM Data – BUTTON
61
iNOTES......................................................................................................................... 61
TABLE OF CONTENTS
ii
4.
AIR DISPERSION MODELLING AND OUTPUT ......... 63
cmax
xmax
ws
PG
xobj1
xobj2
xobj3
Concentration CLIMIT
Concentration RBC
FT
PSEUDO-SOURCE PARAMETERS ............................................................................ 64
CALCULATION BUTTONS ......................................................................................... 67
1.AERSCREEN-MAX
70
2.AERSCREEN-(RBC) (User Met)
70
3.AERMOD-RBC (User Terrain & User Met)
71
4.Create AERMOD files (Average Wind Speed)
71
5.AERSCREEN-MAX (HbH)
71
6.AERSCREEN-RBC (User Met)
71
7.AERMOD-RBC (HbH User Terrain & User Met)
72
8.Create AERMOD files (HbH User Met)
72
Example Run-Times
72
NON-ROUTINE FLARE AIR DISPERSION MODELLING .......................................... 73
ROUTINE FLARE AIR DISPERSION MODELLING ................................................... 74
AERSCREEN INPUTS ................................................................................................. 75
Hour by Hour
76
AIR DISPERSION MODELLING PREDICTIONS ........................................................ 78
oAERSCREEN Output Summary
78
79
79
79
79
80
80
80
80
80
R
A
oMODELLING Output Summary
81
ONE-HOUR AVERAGES FROM SUB-HOURLY EMISSIONS ................................... 81
ERCB D060 Permit Conditions .................................................................................. 82
OUTPUT SUMMARY OF SOURCE ............................................................................. 84
OUTPUT SUMMARY OF AIR DISPERSION MODELLING ........................................ 86
OUTPUT FIGURE 1 ..................................................................................................... 86
OUTPUT FIGURE 2 ..................................................................................................... 88
5.
TRANSIENT BLOWDOWN FLARING ...................... 90
TRANSIENT BLOWDOWN INPUTS ........................................................................... 90
Transient Source
91
D
Expected Maximum Initial Pressure, PRESS0
92
Expected Minimum Initial Gas Temperature, TEMP0
92
Expected Minimum Final Pressure, PRESS1
92
Pipeline/Vessel Inside Diameter, VESSELDIA
93
Pipeline/Vessel Length, VESSELLEN
93
Minimum Orifice Diameter, ORIFICE_DIA
93
Discharge Coefficient, DCOEFF
93
Select the way the blowdown curve is converted from a continuous curve to discrete
steps, MDIST
94
Raw Gas User Initial maximum flow rate, QMAX
94
Raw Gas User Total volume within vessels/pipes, QTOTAL
95
User selected # of puffs, NPUFFS
95
User selected puff duration, PUFDUR
95
TRANSIENT BLOWDOWN AIR DISPERSION MODELLING .................................... 96
TRANSIENT BLOWDOWN OUTPUT .......................................................................... 97
6.
ADVANCED TECHNIQUES ................................... 98
SITE SPECIFIC METEOROLOGY ............................................................................... 98
Step 1
99
Step 2
100
Step 3
101
TABLE OF CONTENTS
iii
Step 4 (Optional)
101
SITE SPECIFIC TERRAIN ......................................................................................... 102
Step 1
102
Step 2
103
REFINED MODELLING ............................................................................................. 104
Non-Routine Flare Air Dispersion Modelling
104
Routine Flare Air Dispersion Modelling
105
POST-PROCESSING EXTERNAL REFINED MODELLING OUTPUT ..................... 106
7.
BATCH OPERATIONS ....................................... 107
Batch File Step-by-Step ........................................................................................... 107
iBATCH PAGE AS A DATABASE ............................................................................ 108
EXAMPLE iBATCH SENSITIVITY SETUP................................................................ 109
References ...................................................... 112
FT
8.
Appendix A .............................................................. 114
A
FUNCTION calcmaxcr(…)......................................................................................... 114
FUNCTION wsstkht(…) ............................................................................................. 118
FUNCTION fneq90(…)............................................................................................... 119
FUNCTION fneq91(…)............................................................................................... 120
FUNCTION NR_Texhaust(…) ................................................................................... 121
Function NR_XL(…) .................................................................................................. 122
FUNCTION p_sat(…) ................................................................................................. 123
FUNCTION t_sat(…) .................................................................................................. 124
FUNCTION FNz(…).................................................................................................... 125
Credits........................................................................................................................ 127
Mapping Code
127
Nearest Neighbour Code
127
R
LIST OF TABLES
Land Cover Classification .................................................................... 41
Screening Land Cover Classification ................................................... 42
Screening Meteorological Variations Used for AERSCREENMAKEMET.exe .................................................................................... 43
Representative Pipeline Quality Natural Gas (Hubbard 2009) ............ 55
D
Table 1:
Table 2:
Table 3:
Table 5:
LIST OF FIGURES
Figure 1:
Figure 2:
Figure 3:
Figure 4:
Figure 5:
Figure 6:
TABLE OF CONTENTS
ERCBflare Page Name and Typical Linkage....................................... 19
High Resolution Land Cover for Alberta .............................................. 40
Screening Land Cover for Western Canada........................................ 42
Worst Case Complex Terrain Graphic Showing Terrain Elevations as a
Function of Distance from the Source Versus Stable Atmosphere
Plume Rise........................................................................................... 59
SO2 Flare Model for Source Parameters ............................................. 65
H2S Stripping Model for Source Parameters ....................................... 66
iv
Figure 7:
Figure 8:
Calculation Flowchart .......................................................................... 68
Illustration of the Range of Plume Rise (centrelines) for An Hour By
Hour Source Model .............................................................................. 78
Example oFIGURE 1 showing the Maximum Concentration at the
Receptor Location in Comparison to the Terrain Elevation ................. 87
Example oFIGURE 1 showing the Wind Speed that Causes the
Maximum Concentration at the Receptor Location ............................. 88
Example Emissions Chart from oFIGURE 2 for Hour-By-Hour
Assessment ......................................................................................... 89
Example Efficiency Chart from oFIGURE 2 for Hour-By-Hour
Assessment ......................................................................................... 89
Coefficent of Discharge for Gas Flow Through an Orifice (Mannan
2005) .................................................................................................... 94
Example Continuous Exponential Blowdown Curve shown ERCBflare
modelled Discrete Stepped Sequence using Equal Mass Steps or
Equal Duration Steps ........................................................................... 96
Figure 9:
Figure 10:
Figure 11:
Figure 12:
Figure 13:
D
R
A
FT
Figure 14:
TABLE OF CONTENTS
v
1. Introduction
FT
ERCBflare, A Model for Temporary Flaring
Permits, Non-Routine Flaring and Routine
Flaring Air Dispersion Modelling for Sour
Gas Facilities
The Energy Resources Conservation Board (ERCB) of Alberta has developed
ERCBflare for evaluating non-routine and routine flaring for Sour Gas Facilities.
This User Guide is for the ERCBflare modelling spreadsheet and associated
modules. The User Guide outlines how to install, use the model and provides
examples on the use of the model.
A
The ERCBflare model was designed to perform state of the art source estimates
for flares and perform screening level air dispersion modelling using state of the
art techniques using existing dispersion models. The user must have a good
understanding of the following documents:
R
1. ERCB Directive 060 (requirements for temporary flaring permits, nonroutine and routine flaring air dispersion modelling expectations, ERCB
2013),
2. ESRD Air Quality Modelling Guideline (2013), ESRD Non-Routine
Flaring Management: Modelling Guidance (2013),
D
3. AERSCREEN and AERMOD (U.S. EPA 2011)
The ERCBflare model documentation is comprised of three components that are
described in the following table.
1 Introduction
1
Description
ERCBflare.xlsm
a Microsoft®-Excel (Office 2013) application software for
Windows® containing macros and a user interface to
external calculation modules
Screening Meteorological Data Files
A set of sixteen data files created using the
AERSCREEN modelling system component MAKEMET
for the eight basic land use groups described in the
AERMOD user guide. For each land use type, there is a
surface and a profile meteorological file. Each file is an
AERMOD-ready meteorology file containing a full range
of meteorological conditions using three month seasons
and the range of conditions found in Alberta.
User Guide Version 1.00.pdf
this User Guide
FT
Module
All of these documents and programs are available as a single download from the
on the ERCB website. For the latest updates visit the ERCB website:
http://www.ercb.ca/regulations-and-directives/directives/directive060
A
Also required is the AERMOD air dispersion model. The AERMOD source code,
documentation and executable files are available at the U.S. EPA Technology
Transfer Network website:
R
http://www.epa.gov/ttn/scram/dispersion_prefrec.htm#aermod
D
Model Changes
Version 1.x to 2.00
•
Added annual and monthly air dispersion modelling predictions. The
monthly predictions are based upon the month of the proposed flaring plus
the month before and after. Using a 5-year data set, this allows for
N>8760 and thus the Risk Based Criteria can be applied.
•
Added inefficiency by-products stream for air dispersion modelling
impacts of products of inefficiency. These products are modelled as raw
H2S whereas combustion products are modelled SO2. The H2S source
differs from the SO2 source; the H2S has a lower heat component based
1 Introduction
2
upon convective stripping of energy from the flame and momentum is
calculated from the mass flow not combusted.
Added lift gas stream effects to combustion calculations.
•
Added flare assist streams for air and/or steam flaring assist. The flare
assist streams impact the combustion calculations by adding momentum,
energy and reducing the combustion efficiency of the flare.
•
Added conversion of sub-hourly emissions predicted concentrations to
hourly average concentration.
•
Added transient blowdown calculations. The spreadsheet prompts for
volume, pressure and temperature of vessel to blowdown and calculates
the exponential blowdown curve for the inputs. The curve is divided into
three steps for modelling. The modelling predictions determine the
maximum hourly concentration from the curve based upon the duration of
each step and the maximum predicted concentrations for each step.
•
Added oFIGURE 1 page to display in a graphical format the predictions
as a function of distance from the facility. The graphic shows max
concentration, wind speed producing the maximum concentration, PG
atmospheric stability leading to maximum concentration as a function of
distance.
•
Added oFIGURE 2 page to display in a graphical format the statistical
summary of emissions and combustion efficiency for hour-by-hour flaring
analysis
R
A
FT
•
Added a DEFINITIONS reference page as per D060
•
Added a LAHEE reference page. Linked the Lahee reference page to the
flaring inputs for the determination of the maximum flaring volume
allowance as per D060.
D
•
•
Changed the summary page to reflect the numerous changes below. Also
added a check-list style table at the top of the page to summarize how the
inputs compare to D060 requirements for approvals.
•
Changed the ATTACHMENTS page to iNOTES page. iNOTES page
has specific prompts for questions that ERCB approval reviewers may
consider in the review of the application.
•
Added a iBATCH and oBATCH pages for batch operation of
spreadsheet. Users can save the input page to the batch page. Inputs and
outputs are saved.
1 Introduction
3
Added NON-DEFAULT flag for all output pages when a non-default
setting is selected.
•
Added start-page for the selection of type of assessment, advanced
program operations and non-default settings
•
Non-routine flaring uses the hour-by-hour variation in source parameters.
This is implemented in AERMOD using a time varying emissions file and
a co-located source. Three sources are defined based upon an estimate of
the final rise of the hourly variation.
•
Routine flaring uses the average meteorological wind speed and
temperature.
•
Added the prediction of concentrations based upon the non-routine flaring
Risk Based Criteria.
•
Added distinction between non-routine flaring and routine flaring. Both
are modelled at 3-emission rates. Non-routine flaring results are
compared to risk based criteria and routine flaring results are compared to
ESRD established modelling objectives.
•
Added AERSCREEN/AERMOD air dispersion modelling. Associated with
this change are the following additions:
A
FT
•
R
o added 8-screening meteorological data sets corresponding to the 8land use types in the Alberta Air Quality Modelling Guideline. The
meteorological data sets represent 100% land cover for the
respective land cover type
D
o added a mapping of Alberta, British Columbia and Saskatchewan
for land cover reduced to the 8-land use types. The spreadsheet
prompts for a location and the nearest land cover cell value is used
to represent the air dispersion modelling.
o allowance for coordinates in geographical coordinates, UTM zones
8,9,10,11,12,13 and 10TM
o air dispersion modelling is performed for the parallel air-flow and
elevated terrain from 100m to 10km. The spreadsheet prompts for
terrain elevations from the base elevation to the maximum terrain
elevation.
o Screening assessment using the AERSCREEN approach with the 8screening meteorological data sets is intended for rapid
assessment.
1 Introduction
4
Added AERMOD air dispersion modelling for refined analysis. Following
a step-wise progression from screening to refined analysis, ERCBflare can
create a refined dispersion modelling input files for AERMOD, run
AERMOD, and post-process the results for non-routine, routine and
blowdown sources.
•
Removed SCREEN3; removed the 99% percentile concentration
prediction based upon ISCST3 air dispersion modelling; fuel gas log; and
minimum fuel calculation based upon the 99% concentration.
•
Added digital terrain processing for inputs to the iTERRAIN page.
Digital elevation data (DEM) is downloaded from the internet site
automatically or pulled from a user’s local library. The terrain processing
extracts worst case terrain contour elevations as would be done if
performed manually; therefore, manual entry of worst case terrain remains
an applicable option.
•
Added iUSERMET page to develop a site specific meteorological data set
suitable for refined dispersion modelling using AERMOD. The
meteorological data set is composed of data from the ESRD MMEU
Meteorological Processor extraction of MM5 data for the province of
Alberta. The iUSERMET page processing also downloads land use
classification code (LCC) files for Canada from the internet or the user’s
local library. The LCC files are processed using AERSURFACE
methodology to determine an average Bowen Ratio, albedo and surface
roughness for the user’s assessment site. The AERMET processor is
subsequently used to create a site specific meteorological file.
R
A
FT
•
Added iUSERTER page to develop a site specific receptor grid suitable
for refined dispersion modelling using AERMOD. Digital elevation data
(DEM) is downloaded from the internet site automatically or pulled from a
user’s local library. The user can create an ESRD standard assessment
grid or modify the receptor spacing. Terrain and hill scale factors are
determined from the DEM data and output to an AERMOD formatted
insert file.
D
•
•
Added oPOSTPROCESS page to post process external AERMOD
output for the determination of ERCB D060 risk based criteria. Similar to
the post-processing provided by the ERCBflare spreadsheet automated
assessment, the oPOSTPROCESS page allows a user to process output
created external to the ERCBflare interface. The oPOSTPROCESS
page allows for advanced statistics for graphical presentation or in-depth
analysis.
1 Introduction
5
User Qualifications
FT
ERCBflare is freely distributed to assist in D060 temporary flaring permitting,
non-routine and routine flaring air dispersion modelling within Alberta.
ERCBflare performs both screening level calculations (uses only a few user inputs
to create a realistic and conservative estimate of flare emissions and
concentrations) and also refined level calculations. Although relatively few
inputs are required through the interface, it is a complex tool. ERCBflare requires
inputs that may require sound engineering judgement or other technical expertise.
It uses site-specific thermodynamics, fluid dynamics, and air dispersion
modelling. Flare dispersion assessment is a multidisciplinary and iterative task
with many assumptions and judgments.
A
The ERCBflare model was created so that a minimal amount of technical
background is required to run the models. However, there remains some technical
knowledge required to supply suitable inputs and the ability to understand
whether the output is appropriate for the inputs and meets the needs of
stakeholders. The user must recognise that the models are technical in nature and
the correct interpretation of the result may require technical expertise that
proceeds from consequences of the inputs. In any modelling assessment, high
quality input data is very important.
R
The model has been created with a professional commitment to environmental
protection and safeguarding the well-being of the public. It is the responsibility of
the software user to accept and continue this commitment in their application of
the software. The software is supplied as a tool to assist the user to comply with
applicable statutes, regulations and bylaws. Neither the software nor application
of the software is intended to replace statutes, regulations or bylaws.
D
Suitable Technical Background
Environmental issues are interdisciplinary in nature. The practice of
environmental science requires the integration of diverse disciplines and
philosophies; many projects will require a team of appropriate specialists to
address complex environmental issues. Persons doing the assessment should
undertake only that aspect of environmental work that they are competent to
perform by virtue of training and experience. Thus they should seek out and use
appropriate Environmental Specialists to provide expert advice on certain
environmental issues.
The basis of the models is technical with expertise required in chemistry,
thermodynamics, atmospheric physics, meteorological processes, industrial
processes and regulatory affairs. While the full technical background is not a
1 Introduction
6
requirement to execute the models, the user of the software is required to have a
general engineering and environmental science background; a general knowledge
of the emission sources: wells, pipelines, and pipeline networks; and a working
knowledge of the most current version of:
•
•
•
ERCB Directive 060 –
Alberta ESRD Air Quality Modelling Guideline and Non-Routine Flaring
Management: Modelling Guidance.
AERMOD/AERMET/AERSCREEN/AERSURFACE user guides
FT
There are many technical inputs required by ERCBflare to perform
the calculations. Some have been prescribed by the ERCB as
mandatory default entries. Others are input by the user for the
specific flaring scenario.
Use of ERCBflare and understanding whether the predictions are
appropriate for the user inputs still, however, requires some specific
technical understanding.
A
How this Document is Organized
Capter
Description
1 Introduction
Backgrounder for ERCBflare
2 Installation and Setup
How to install the ERCBflare program
3 ERCBflare Inputs
Description of the program prompts and user entries
4 Air Dispersion Modelling and
Output
Description of the program output calculations and
graphics
5 Transient Blowdown Flaring
Description of inputs and modelling for flaring from the
depressurization of vessels and pipelines
6 Advanced Techniques
A description of the modelling that can be performed
using ERCBflare going beyond screening air dispersion
modelling
7 ERCBflare Batch Operation
Description of the operation of ERCBflare for multiscenario operation
D
R
This User Guide is organized as both an instructional guide and a tutorial. This
guide is divided into the following chapters:
1 Introduction
7
About this Guide
The following symbols and conventions are used in this guide
Used for menu, command, and keyboard selections you
Bold make and screens you will see.
Italics Used for emphasis and to identify new terms
text User typed responses or entries
FT
Helpful information about a particular topic.
Important information to prevent problems and ensure that
you are successful in using the software.
A hyperlink to a section within the User Guide, an internet
web site or email address
A
link
R
Where to Go for Help
D
ERCB welcomes your feedback. Comments on the current version, suggestions
for features in future versions, or bug reports in the User Guide or ERCBflare
software can be submitted to the email below. Please provide information on the
version of Windows you are using, version of Excel, and enough information to
duplicate the issue. Thank you.
[email protected]
What is ERCBflare all about?
The ERCBflare spreadsheet tool was created to provide consistency in the
calculation of flare-type source parameters for use with standard air quality
dispersion models that are based upon chimney-type sources parameters. The
basic premise of ERCBflare is to determine the energy released during a flaring
event; this energy is released to the atmosphere giving buoyancy and momentum
1 Introduction
8
plume rise for the flare-type source. ERCBflare then uses the buoyancy and
momentum energy and back-calculates chimney-type equivalent source
parameters that can be used as inputs in typical air dispersion models. These
parameters being the height, diameter, velocity and temperature of the source. The
equivalent source parameters are frequently called pseudo-parameters. The
parameters are called pseudo-parameters because they are not real-world
dimensions, but only calculated inputs that represent the flare-type source so that
the correct plume height is used in the standard dispersion model. Standard
dispersion models determine the energy available for plume rise based upon the
calculation of buoyancy and momentum energy from the entered chimney source
parameters (height, diameter, velocity and temperature.)
A
FT
Flaring can be loosely divided into three types: continuous, short-term steady and
transient. Continuous flaring is flaring when the emissions occur hour after hour
for long periods of time (weeks, months or years). The flared gas for continuous
emissions can vary from zero flow rates, to low flow rates and to high flow rates
over the course the flare duration. Short-term steady flaring is similar to
continuous flaring but flaring occurs only for prescribed period of time. Transient
flaring is similar to short-term steady flaring where the flare is operated for a
prescribed period of time, but the flow rate to the flare originates from a high
pressure blow down resulting in a scientifically well-defined change in flow rates
from very high to low over a specific period of time. In each of the flare types, it
is important to examine the range of low flow rates to high flow rates and the
pollutants released during these periods. Low flow rates are associated with low
plume rise and high flow rates are associated with higher plume rise.
D
R
From a regulatory standpoint, flaring is divided into two broad categories:
continuous (including short-term steady) flaring for routine operations; and, nonroutine flaring (including short-term steady and transient flaring). Non-routine
flaring has special regulations because it is associated with specific disposal of
large amounts of gas and typically high pollutant emissions. Non-routine flaring
is divided into two categories: planned flaring (including well test flaring,
maintenance and turn-around operations); and, unplanned flaring (including
emergency flaring). The distinction between routine, non-routine (planned) and
non-routine (unplanned) is the frequency of the flaring events and the inherent
risk to environmental harm (consequence). Planned non-routine flaring from
temporary flares requires a permit from the ERCB.
ERCBflare version 1.x was originally created to specifically address the concerns
related to the applications to ERCB for non-routine flaring associated with well
test flaring. Because non-routine flaring events are often associated high energy
and high levels of pollutant emissions, it was necessary to update the ERCBflare
calculation tool to incorporate the more complex source conditions demanded by
industry users, operations and high impact but low risk predictions. ERCBflare
version 2.x was created to specifically address these complex questions and
therefore ERCBflare is inherently more complex than its predecessors (ERCBflare
version 1.x and ESRD flare tool). An important change also included in the
1 Introduction
9
ERCBflare 2.x is the updated air quality dispersion model used to predict
environmental harm (AERMOD), and it too is more complex than its relatively
simplistic predecessor (SCREEN3).
D
R
A
FT
The purpose of ERCBflare version 2.x is to provide a relatively easy-to-use user
interface tool to assist the user in the complex analysis of routine and non-routine
flaring. The goal of ERCBflare is to predict a rapid (within minutes),
conservative estimate of flaring impacts. It also provides next-step analyses
methods to bridge the gap between rapid screening analysis and full-refined air
quality dispersion modelling.
1 Introduction
10
2.
INSTALLATION AND SETUP
FT
This chapter describes how to install the ERCBflare software. The ERCBflare
modelling package includes a user interface (a Microsoft®-Excel Office 2013
spreadsheet) that prompts the user for inputs and controls operation of the air
dispersion modelling stand-alone programs (specifically AERMOD).
Microsoft Excel must be installed on your computer in
order to use the ERCBflare tool.
A
Overview of the Installation
Process
R
If you have not already done so, download the installation software from the
ERCB website (see above).
D
The single install program setup.exe file contains all of the software required for
the ERCBflare program to run within the users existing Microsoft Office
(including Excel) environment.
STEP 1: The ERCBflare tool is packaged as a downloadable ZIP file which the
user can unpack to form a folder tree and access directly. Download the file
ERCBflare package from:
http://www.ercb.ca/regulations-and-directives/directives/directive060
There are three ZIP files provided
File
Intent
Recommended Download Destination
Folder
ERCBflare_vxxxxxx.zip
Required
c:\myfiles\ercbmodels\ercbflare\
ERCBflare_UserGuide_vxxxxxx.pdf
Required
c:\myfiles\ercbmodels\ercbflare\docs\
2 Installation & Setup
11
MAKEMET.zip
Optional
c:\myfiles\ercbmodels\ercbflare\makemet\
The ERCBflare_vxxxxx.zip file contains the necessary user-interface and
meteorological data files used in the assessment flaring. It is recommended that
the users download and read the user guide for ERCBflare.
FT
The MAKEMET.zip file contains a modified version of the U.S. EPA
MAKEMET program. The U.S. EPA version of MAKEMET is command line
driven, requiring the user to enter data and several prompts. The modified version
of MAKEMET uses input files, and therefore is more readily useable for batch
operations, and is less entry-error prone. MAKEMET is an optional processing
that is required only if the user wishes to create a site-specific screening
meteorological data file.
The files may be installed to alternate folder locations during
the installation.
A
STEP 2: User’s will also require files that are updated and distributed by the U.S.
EPA: AERMOD and AERMET. The AERMOD source code, documentation and
executable files are available at the U.S. EPA Technology Transfer Network
website:
http://www.epa.gov/ttn/scram/dispersion_prefrec.htm#aermod
R
The AERMET source code, documentation and executable files are available at
the U.S. EPA Technology Transfer Network website:
D
http://www.epa.gov/ttn/scram/metobsdata_procaccprogs.htm#aermet
The files may be installed to alternate folder locations during
the installation.
STEP 3: Using the Windows explorer (Windows Key + e) locate the
ERCBflare_vxxxx.zip file that you downloaded STEP 1. Right clip on the file
and select ‘unzip to here’ (if you have the WINZIP utility) or ‘unpack’ (if you are
using the embedded unzip functionality within Windows). The following folders
and files will be created (where [mydir] is the folder used in STEP 1:
2 Installation & Setup
12
[mydir]ERCBflare_vxxxxx.xlsm
[mydir]metfiles\(conif.x;cult.x;
decid.x;desert.x;grass.x,swamp.x;
urban.x;water.x)
FT
[mydir]metfiles\test_ercbflare\(aermod.x)
The ERCBflare spreadsheet tool
The screening meteorological data files
created using MAKEMET. These 8meteorological files are used to determine
screening level concentrations using
ERCBflare in the screening modes. For each
meteorological file name there is a *.SFC and
*.PFL file representing an AERMOD ready
surface profile meteorology file, respectively.
This folder contains an example site specific
meteorological file. There is a *.SFC and *.PFL
file representing an AERMOD ready surface
profile meteorology file, respectively
The test_ercbflare example site specific meteorological
data file is provided as an example for use in the tutorials
provided in this User Guide. User’s should not use the
example site specific meteorological data for their air
quality evaluations or permit applications.
A
STEP 4: If you haven’t already done so, download and install the AERMOD air
dispersion model and AERMET meteorological model from the U.S. EPA
website, previously listed.
D
R
AERMOD must be installed on your machine in order to
use all of the features of ERCBflare for air dispersion
modelling screening.
SETUP
Now that you are finished installing the ERCBflare spreadsheet tool,
meteorological files and AERMOD program, you can complete the
installation by loading the ERCBflare spreadsheet and following the steps
below.
1.
From the select Windows START->Excel, this will launch Excel. Then
using the Excel menu option File->Open or use the Windows Explorer to
navigate to the installation folder for ERCBflare and [mydir] and double-
2 Installation & Setup
13
click on the ERCBflare_vxxxxx.xlsm file. This will load the ERCBflare
tool into Excel.
The ERCBflare spreadsheet is an XLSM file that contains typical
spreadsheet-like calculations as well as program macros. This is the “M”
part of the filename extension. When the file is opened, your computer and
office security settings may display a warning. Select ‘Trust this document’
or ‘Enable Macros for this document’ if prompted when the file is opened.
Failing to do so will prevent ERCBflare from operating.
3.
Click on the iSTART page. Select “Show” for the Hide iBIN Page option.
4.
Click on the iBIN page. For the entry for AERMOD, type the full path and
file name of the AERMOD.exe file (previously installed). The Browse
button can be used to navigate your file folder tree using common Windows
Explorer methods.
5.
For the entry for METFILES, type the full path and file name of the folder
location where the screening meteorological files were stored. By default
this will be:
FT
2.
A
[mydir]\metfiles
The files can be placed anywhere on your network. ERCBflare requires
read access to the files only.
If the BROWSE button was used to select the files or the SAVE button
pressed, then ERCBflare stores the entry for your computer setup. If you
load an uninitialized ERCBflare spreadsheet or a spreadsheet initialized to
another users folder locations, ERCBflare synchronizes the settings to your
computer setup. You have the option of saving the file when you have
completed the calculations. You can force ERCBflare to synchronize by
pressing the SYNC button.
D
R
6.
7.
Click on the iSTART page. Select “Hide” for the Hide iBIN Page option.
Typically, you shouldn’t have to revisit the iBIN page. However, use the
iSTART page to show the iBIN page to make any changes to your system
as required.
Select ‘Trust this document’ or ‘Enable Macros for this
document’ if prompted when the file is opened. Failing to
do so will prevent ERCBflare from operating.
ERCBflare spreadsheets must be saved as an XLSM file
type to preserve the macro (Visual Basic for Applications)
functionality.
2 Installation & Setup
14
This completes the mandatory components initialization of the ERCBflare. The
user may wish to also initialize several optional components of ERCBflare.
AERMET: On the iSTART page, for the row entry for AERMET, enter the
full path for the location of the AERMET.exe file or use the BROWSE
button to navigate to the installation folder and click on the AERMET.exe.
FT
MAKEMET: On the iSTART page, for the row entry for MAKEMET, enter
the full path for the location of the MAKEMET.exe file or use the
BROWSE button to navigate to the installation folder and click on the
MAKEMET.exe.
DEMLIB: On the iSTART page, for the row entry for DEMLIB, enter the
full path for the location of where digital elevation files should be stored
when downloaded from the internet. DEMLIB represents a library or cache
to save time on subsequent analyses to avoid download times. Also, the
DEMLIB provides a library of files for repeatability of ERCBflare
predictions.
A
LCCLIB: On the iSTART page, for the row entry for LCCLIB, enter the
full path for the location of where land-use classification code (LCC) files
should be stored when downloaded from the internet. LCCLIB represents a
library or cache to save time on subsequent analyses to avoid download
times. Also, the LCCLIB provides a library of files for repeatability of
ERCBflare predictions.
R
DEMURL: on the iSTART page lists the internet URL where DEM files are
downloaded from. This entry is not editable.
D
LCCURL: on the iSTART page lists the internet URL where LCC files are
downloaded from. This entry is not editable.
To test your system you can follow the example on the Chapter 7: Batch
Operations to load the provided example or follow these basic steps.
1.
Click on the iSTART page. Select “Show” for the Hide iBATCH Page
option.
2.
Click on the first data row (row 8) below the titles. This selects this row as
the active Batch operation row.
3.
Click the Load Current Row button at the top of the page. Clicking the
button copies the inputs from the iBATCH page to the respective inputs on
2 Installation & Setup
15
the iFACILITY, iFLARING, iTERRAIN, iNOTES, and mode selections
on the iSTART page.
Click on iFLARING to confirm that the iBATCH entries were copied to
the iFLARING and other pages.
5.
Click on the Recalculate button at the top of the iFLARING page. This
will re-direct you to the oMODELLING page and the location where the
modelling option buttons are placed. The modelling options area shows
two streams of buttons. On the left are Non-Routine flaring modelling
options and on the right are Routine flaring modelling options. These are
discussed in more detail in a later chapter of this guide. Above the
modelling buttons are the summarized source inputs to be used in the
modelling based upon average meteorological wind and temperature
conditions. Below the buttons area, the results of the air dispersion
modelling will be summarized once complete.
6.
For this example, it doesn’t matter if the Non-Routine mode or Routine
mode was selected. Click on the 5. AERSCREEN-MAX button. This will
create input files based upon the screening meteorological files and will
execute the AERMOD program. If you properly installed both of these and
pointed the iBIN page to their proper location, then the ERCBflare program
should loop through 6-scenarios of calculations. The screen may flash and
update during the calculation sequence. A status of the calculations is
shown at the bottom of the screen. It is recommended that you DO NOT
continue to use your computer for other concurrent Windows applications
because this can cause interference and instability within the calculations.
R
A
FT
4.
D
It is recommended that you DO NOT continue to use your
computer for other concurrent Windows applications
because this can cause interference and instability within
the calculations.
2 Installation & Setup
16
3.
ERCBflare INPUTS
This chapter provides the following information about the general operation of the
ERCBflare program:
what inputs are and how they work together
•
what the buttons/menu items do
•
overview of the calculation processes
FT
•
Introduction
R
A
The ERCBflare program uses the familiar Microsoft Excel as host for the
calculations. ERCBflare.xlsm contains the user-interface for the calculations,
allowing the user to input information and view calculation results. The
ERCBflare application consists of the ERCBflare.xlsm macro enhanced
spreadsheet. The spreadsheet uses external files (read only meteorological files
and the AERMOD.exe executable air dispersion model).
D
The ERCBflare spreadsheet is a stand-alone spreadsheet
containing all of the necessary macros and programming to
load and process input and output from the modules
3 ERCBflare Inputs
ERCBflare.xlsm is an Excel spreadsheet file (.xlsm)
containing macros (programming) that acts as a user
interface for input files and output from the processing
modules
All inputs and outputs are stored within the spreadsheet.
Temporary files are created in the parent folder for the
spreadsheet during air dispersion modelling calculations.
The user can preserve intermediate modelling files or create
modelling files to the folder of choice when those options are
selected.
17
Graphical User-Interface
Overview – The ERCBflare GUI in Excel
FT
All user input and output are controlled using an ERCBflare.xlsm file for
Microsoft Excel. The Graphical User-Interface (GUI) file (.xlsm) contains
several Excel worksheets (pages). The pages are shown in Figure 1 and shows
the typical linkage of sheets. Typical operations are linked with bold lines; the
user may wish to view operations linked with solid lines; and technical or optional
operations are linked with dashed lines. Typical operations include:
Selection of Permit (Non-Routine Planned Flaring) mode or Routine (Air
Quality Evaluation of Flaring) mode
•
Inputs are entered on several pages
•
Air dispersion modelling options are available for low-level screening
(maximum predictions are compared to objectives), screening (percentile
concentration predictions are compared to objectives or risk based criteria)
and refined air dispersion modelling (recommended methods to be used if
screening doesn’t pass).
•
Outputs are provided in a summary table and graphics are provided that
may be useful in reporting
R
A
•
•
Technical pages display the detailed source calculations and intermediate
calculation steps
D
The ERCBflare.xlsm spreadsheet can be used as a stand-alone input/output
calculation for each project/scenario or it may be used as a central database of
inputs. Both systems have advantages and disadvantages.
Stand-Alone
When the user uses ERCBflare in a stand-alone framework, the user would load a
copy of the spreadsheet into EXCEL; user inputs are entered on the input pages;
and following the completion of the air dispersion modelling the outputs are
displayed on the output pages. The user would save the ERCBflare spreadsheet in
a sub-folder of the project or scenario being assessed. In this manner, the exact
inputs and outputs are maintained for later verification or reference.
3 ERCBflare Inputs
18
Central Database
iBIN
iBATCH
oBATCH
iNOTES
oSUMMARY
iFACILITY
oBLOWDOWN
oFIGURE1
A
iSTART
FT
When the user uses the ERCBflare in a central database framework, a single copy
of the ERCBflare spreadsheet is used to input, perform calculations and same
summary output. Detailed output information for the inputs are lost, when new
information is input, but the inputs can be restored from the database and the
outputs readily re-created. This framework is advantageous to perform sensitivity
testing or design scenarios where specific output results are required and not all of
the details. The central database framework corresponds closely with the batch
operation of ERCBflare (see Section 6: Batch Operations)
iFLARING
oMODELLING
R
iTERRAIN
iUSERMET
AERSURFACE
oMATRIX
oCALCULATIONS
iUSERTER
D
oFIGURE2
PROPERTIES
oAERSCREEN
ABOUT
oPOSTPROCESS
LAHEE
DEFINITIONS
Figure 1:
3 ERCBflare Inputs
Modelling Output
Created External to
ERCBflare
ERCBflare Page Name and Typical Linkage
19
The ERCBflare Excel Pages
ERCBflare has 24 GUI Excel pages representing informational, input, output and
technical output. Pages for input have tabs that are green, output and information
pages are pink, and technical output pages are yellow (coloured tabs on pages are
visible for Office 2002 and newer versions of Excel only).
FT
Tabs for the page types are colour coded for Office 2002
and newer versions of Excel (earlier versions do not display
the page colours).
Input pages are green; output pages are pink; and technical
pages are yellow.
Input pages start with “i”
Output pages start with “o”
A
All calculations are performed in the Excel pages for complete transparency of all
calculations. The programming and macros imbedded in the ERCBflare tool are
security locked to protect the integrity of the user interface. The programming
and macros do not perform any of the calculations with the exception of: equation
root solvers; thermodynamic equations of state solvers; and thermodynamic
property from gas composition matrix operations. These calculations are
described in the Appendix B.
R
A complete list of user-interface pages are described in the following table:
Purpose
ABOUT
User
Information
General information and instructions for use of the spreadsheet. Use this
sheet to link to the latest information regarding ERCBflare.
LAHEE
User
Information
A list of the LAHEE well classifications. The LAHEE system has been
adopted by ERCB to classify well operations. The LAHEE classification is
linked to ERCB D060 non-routine planned flaring maximum flare rates.
DEFINITIONS
User
Information
A list of ERCB D060 terminology related to routine and non-routine flaring
iSTART
Input
iBIN
Input
iBATCH
Input
oBATCH
Output
iFACILITY
Input
D
Page Name
3 ERCBflare Inputs
Description
All flare calculations must start at this page to specify the mode of the
screening calculations. Also on this page are several non-default operation
selections and inputs.
The ERCBflare.xlsm GUI needs to know where the user has stored
meteorological files and the location of the AERMOD executable.
The iBATCH page is used as a database of inputs and summary outputs.
The user can use the iBATCH to run sensitivity tests or a running list of
wells for a field of operations
The oBATCH page contains the output for matching row numbers on the
iBATCH page
The iFACILITY page is used to enter facility level inputs such as owner and
operator, location of the flare and emergency planning. Many of these
inputs are important inputs for the non-routine planned flaring temporary
permit process. If the routine flaring mode is selected, then many of
iFACILITY entries are greyed-out and are not required entries.
20
iFLARING
Input
oBLOWDOWN
Input
(Normally Hidden)
iTERRAIN
Input
iUSERMET
Input
iUSERTER
Input
iNOTES
Input
oSUMMARY
Output
oFIGURE 1
The iFLARING page is used to enter flare level inputs such as source flow
rates and gas compositions. The iFLARE page is also used to select
between a steady release and a transient blowdown release.
The oBLOWDOWN page is normally hidden unless the selection on the
iFLARING page for MBLOWDOWN=1. In this case, the transient blowdown
calculations are invoked. The oBLOWDOWN page contains the
calculations for the exponential blowdown approximation for source flow
rate and conservation of mass.
The iTERRAIN page is used to enter the worse terrain elevations
surrounding the well location. A graphic is provided on the page to illustrate
the entries. The graphic is a good way to confirm the user entries are
correct.
The iUSERMET page is a step-by-step entry and tutorial page for creating a
site specific (user-created) refined meteorological data set. The output will
be 5-year data set with site specific land use characterization. An
AERMOD ready surface and profile files are created.
The iUSERTER page is a step-by-step entry and tutorial page for create a
site specific (user receptor) refined receptor grid. The output will be an
AERMOD formatted file that can be inserted into an AERMOD ready inputs
control file.
The iNOTES page has two purposes. First, the ERCB approval reviewers
specifically request a descriptive response for the first four questions to
assist in their understanding of the application and approval decision.
Secondly, the page allows the user keep track of assumptions or
information regarding the assessment.
The oSUMMARY page provides a summary check list of the flaring inputs
against the limits and requires outlined in ERCB D060. The oSUMMARY
page also summarizes the important source and air dispersion modelling
predictions the reviewer will consider in approval/application process. Many
of the fields are not applicable for routine flare air dispersion modelling but
the page can be a useful summary of the air dispersion modelling results.
The oFIGURE 1 page displays the air dispersion modelling as a function of
distance from the source. The graphic can provide useful information for
flaring management.
The oFIGURE 2 page display the source emissions and combustion
efficiency calculation results when using the hour-by-hour source model.
The oMATRIX page displays a screening matrix summary of the
concentrations and predicted distances as a function of wind speed and
atmospheric stability class.
The oMODELLING page is a technical output page that displays the
calculated pseudo-parameters for the flare source that are suitable for
refined air dispersion modelling outside of the ERCBflare model based upon
average meteorological temperature and wind speeds. It also presents air
dispersion modelling options depending upon the flare Non-Routine or
Routine flaring mode selected on the iSTART page. The results of the air
dispersion modelling are summarized for each scenario at the bottom of the
page.
The oAERSCREEN page is used to configure the AERMOD air dispersion
model to perform AERSCREEN style screening air dispersion modelling.
The detailed output of the air dispersion modelling are listed on this page as
well as the oFIGURE 1 page. The user can use this page to determine
seasonal or annual exceedance of the objectives for each model scenario.
The oPOSTPROCESS page can be used to post-process air dispersion
modelling result files (POSTFILE) to create tabulated and graphical
statistical summaries. This page is included because the risk based criteria
cannot be calculated using the standard AERMOD output.
The oCALCULATIONS page is a technical listing of the combustion
calculations. It also lists the source mass and energy balance intermediate
calculations. The results of these calculations are summarized on the
oMODELLING page.
The PROPERTIES page displays a technical reference listing of the
chemical and thermodynamic properties used in the calculations.
Output
Output
R
oFIGURE 2
Description
FT
Purpose
A
Page Name
Output
oMODELLING
Summary of Technical
Output and Location of
Modelling Calculation
Option Buttons
D
oMATRIX
oAERSCREEN
Technical Output of Air
dispersion modelling
Intermediate Results
oPOSTPROCESS
Technical Output of Air
dispersion modelling
EXTERNAL Results
oCALCULATIONS
Technical Output of
Intermediate Source
Calculations
PROPERTIES
Reference
3 ERCBflare Inputs
21
Page Name
Purpose
AERSURFACE
Reference
Description
The AERSURFACE page displays a technical reference listing of the
Bowen ratio, albedo and surface roughness used by AERSURFACE and
used in the land-use classification processing by ERCBflare to create the
site specific meteorological data file on the iUSERMET page
Fundamentals
FT
iSTART
The iSTART page is the starting point for all ERCBflare modelling.
D
R
A
The iSTART page fundamental entries are illustrated in the figure below. These
options specify cosmetic or convenience modes (i.e., hide or don’t hide technical
information) but also control how the screening calculations are to be performed
(i.e., Permit or Evaluation of air quality). If ERCBflare is operated in the nonroutine flaring mode, then the ERCB risk-based criteria will be used to evaluate
air dispersion modelling predictions. Otherwise, routine flaring air dispersion
modelling mode uses the establish ESRD risk based criteria to evaluate air
dispersion modelling.
Permit or Evaluation Mode
This selection is used to toggle the visibility of input cells, technical pages and
operations based upon whether the intent of the calculations is to be used for an
ERCB temporary flaring permit or any other purpose for evaluating air quality
concentrations. If the intent is for a permit, then specific information related to
the permitting process is gathered from the user as well as restricting the
calculations to the ERCB mandated flaring assessment calculations using hourby-hour variability. An ERCB temporary flaring permit is by definition a non-
3 ERCBflare Inputs
22
routine planned flaring event, therefore the flaring classification is locked to this
setting when PERMIT is selected.
When the ERCBflare calculation mode is set to EVALUATION, many of the
inputs and emissions restrictions are grayed-out and are not required to be filled
in. The EVALUATION mode can be used for what-if analyses, ESRD facility
approval analyses or other flaring investigations. In EVALUATION, routine,
non-routine planned or non-routine emergency flaring classifications can be
selected.
Flaring Classification
FT
The flaring classification setting is an important selection since it prescribes the
modelling methods to be used as well as the objectives to be used for evaluation
of the results of the modelling.
R
A
A Non-Routine flaring assessment is intended to accompany an ERCB D060 nonroutine flaring approval application. Non-Routine flaring has strict guidelines on
the intended purpose of the flaring, timing of the flare event, and the ability to
predict emissions and design the stack. Non-Routine flaring is typically
associated with higher than normal flare volumes with rare intermittent
occurrence or very-rare occurrence. Non-Routine flaring is divided in two types
of non-routine flaring. Because of the variable nature of non-routine flare events,
non-routine flare emissions and air dispersion modelling are analyzed using a
mandatory maximum, average and low flaring rate. All three flow rates must be
analyzed because of the non-linearity between the energy released and resulting
plume rise, flaring emissions and terrain impacts. It can occur that worst case
impacts result from low flaring rates.
D
A routine flaring assessment is intended to analyze steady or continuous
emissions from a flare stack. Routine flaring assessments are performed using a
single flaring rate that is representative of the maximum emissions rate. If the
flare is used for multiple scenarios or multiple emission rate events, the user
should assess each flaring event and flow rate independently to verify that flaring
under all scenarios will meet the ambient air quality objectives.
In the example shown in the figure below, the user has previously selected the
PERMIT Mode on the iSTART page. Only the Non-Routine—Planned Flaring is
available.
3 ERCBflare Inputs
23
FT
(see also to the right, an extension of the table with ppm equivalents)
R
A
Only the Non-Routine flaring assessments can make used of the risk based criteria
(RBC). The RBC used for Non-Routine planned and emergency/upset flaring
classifications are listed to the right of each option. For planned flaring, the
modelled 99% 1h average SO2 concentration at each receptor must be less than or
equal 450 µg/m³ (or 0.172 ppm). The 99.9% 1h average SO2 concentration at
each receptor must be less than or equal to 900 µg/m³ (or 0.344 ppm). The upper
limit of SO2 is not risk based. The same RBC is applied for predictions of H2S,
the modelled 99% 1h average H2S concentration at each receptor must be less
than or equal 14 µg/m³ (or 0.01 ppm). The 99.9% 1h average H2S concentration
at each receptor must be less than or equal to 13931 µg/m³ (or 10 ppm). The
upper limit of the H2S is based upon the evacuation criteria for H2S used in the
ERCBH2S emergency response planning.
D
For emergency/upset flaring, air dispersion modelling uses the 90% RBC 1h
average SO2 concentration at each receptor must be less than or equal to
450 µg/m³ (0.172 ppm). The emergency/upset flaring maximum concentration
(100%) 1h average SO2 concentration at each receptor must be less than or equal
to 9923 µg/m³ (3.789 ppm). The maximum SO2 represents the evacuation criteria
for SO2 used in the ERCBH2S emergency response planning. For H2S, the 90%
RBC 1h average concentration at each receptor must be less than or equal to
14 µg/m³ (0.01 ppm) and the 100% maximum 1h average H2S concentration must
be less than or equal to 13931 µg/m³ (10 ppm) representing the evacuation criteria
for H2S.
Although ERCBflare uses RBC criteria for modelling and flare stack design
purposes, the ESRD requires that any direct measurement of air quality outside of
the facility fenceline (i.e., public area) must be less than or equal to the ambient
air quality objectives which is protective of human and environmental health.
3 ERCBflare Inputs
24
These objectives do not prescribe that a human receptor must be present, but are
protective to human and environment that could be or have the right to be present.
Hide Technical Pages
FT
The hide technical pages option is cosmetic for ERCBflare operations. Hide
technical toggles the visibility of oCALCULATIONS and oAERSCREEN
pages. Although, oMODELLING is a technical page, it must remain visible
because it displays the summarized source parameters, is home to the air
dispersion modelling buttons, and displays the summary of the equivalent 1h
average concentrations.
Hide iBATCH Page
The hide iBATCH page option is cosmetic for ERCBflare operations. Hide
iBATCH page toggles the visibility of the iBATCH page. If the user is not using
the batch mode features, then this page can be hidden from view.
A
Hide iBIN page
R
The hide iBIN page option is cosmetic for ERCBflare options. Hide iBIN is used
to toggle the visibility of the iBIN page. Once the ERCBflare spreadsheet has
been configured, typically operations will not require the page (see Chapter 2:
Installation and Setup) and it can be hidden from view.
Advanced Switches
D
The advanced switches are useful for advanced users for debugging and for other
automation links. The advanced switch block is shown in the figure below. The
defaults are shown and the user can over-ride the default by entering a value in
the INPUTS column. Entries different than the default are highlighted for
convenience.
3 ERCBflare Inputs
25
FT
MRUNBAT
The ERCBflare model creates a small batch file that is used to call the AERMOD
program and link it to the temporary filed used as input to AERMOD. To save
time, ERCBflare doesn’t recreate the RUN.BAT file each time. The user may set
the mrunbat=1 to force run.bat to be created each run or 0 to prevent the file from
being recreated. This option can be useful if the wizard user wants to add pre- or
post-processing commands into RUN.BAT file.
A
MPAUSE
D
R
ERCBflare creates a synchronous secondary process using the command prompt.
By default ERCBflare minimizes the command prompt to the task bar as a
reference during the air dispersion modelling. The command prompt window is
closed automatically after execution of AERMOD and control is returned to
ERCBflare calculation process. For debugging, a PAUSE statement can be added
to the RUN.bat file to force the user to look at the command prompt window and
enter an acknowledgment to proceed. Select mpause=1 to include the PAUSE
statement. The mrunbat=1 option should also be set to force the RUN.BAT file to
be re-written.
MKEEP
The mkeep option prevents ERCBflare from deleting the temporary input files
used to control and get results from individual AERMOD scenarios. The
mkeep=1, the temporary files are not deleted. Note that this may create a large
number of files. The files have a temporary name prefix created by the system.
The mkeep option is useful for creating input files for further analysis external to
ERCBflare or for debugging.
3 ERCBflare Inputs
26
MKEEPOUT
The mkeepout option prevents ERCBflare from deleting the temporary output
files used to control and get results from individual AERMOD scenarios. The
mkeepout=1, the temporary files are not deleted. Note that this may create a large
number of files. The files have a temporary name prefix created by the system.
The mkeepout option is useful for debugging or for post-processing the advanced
oPOSTPROCESS page.
MRELLOC
A
FT
The mrelloc specifies whether the output files for the user specified AERMOD
input files contain relative location coordinates or the absolute coordinates
specified on the iFACILITY page. By default, ERCBflare performs its
calculations using relative location coordinates with the flare located at 0,0.
When the user selects CREATE AERMOD INPUT files from the
oMODELLING page, the user may be merging this input file with site specific
terrain. To assist in this automation, the user can select mrelloc=1 to force the
flare location to the X,Y location rather than edit the created file. If a site specific
receptor file is used, then this option has no effect.
MEXPREC
D
R
The mexprec switch is similar to the mrelloc switch. Mexprec switch is used to
prevent the export of the receptor locations used in the ERCBflare screening. By
default, ECBflare creates a list of discrete receptor locations and terrain
elevations. When mexprec=1, ERCBflare only outputs the start header and end
header for the AERMOD input file and the user automation can be insert the sitespecific receptor grid and terrain elevations. If a site specific receptor file is used,
then this option has no effect.
MSCREEN
The mscreen switch is used to force the ERCBflare modelling runs using
AERMOD to use the SCREEN control option. This setting forces all wind
directions to go to all receptors. This setting is a worst-case option and will lead
to conservative predictions. This setting is useful to perform screening
assessments using user-meteorological files, such as a user created screening file
created using MAKEMET. By default, modelling performed using calculation
buttons AERSCREEN-XXX use the SCREEN control setting and modelling
performed using calculation buttons AERMOD-XXX do not use the SCREEN
setting. Therefore, the mscreen setting only affects the AERMOD-XXX button
calculations.
3 ERCBflare Inputs
27
MELEV
MDEMGET
FT
The melev setting is used in connection with the Get DEM Data buttong on the
iTERRAIN page. On the iFACILITY page, the user is prompted to enter the
location of the flare and its elevation. On the iTERRAIN page, DEM data is used
to determine the surrounding terrain elevations and interpolates to determine the
flare base elevation according to the DEM data. Due to grading or DEM
variation, the iFACILITY elevation may be different than the DEM derived
elevation creating an effective pedestal or pit effect. In either case, the flare stack
height could be arbitrarily increased or decreased leading to incorrect ground
level concentrations. If the difference between the two elevations is greater than
2 m, a warning is displayed and the user is required to justify the difference. By
default, the melev switch is used to force the iFACILITY elevation entry to the
DEM derived elevation. This makes the flare height consistent with the DEM
data set.
R
A
The mdemget setting is used to force ERCBflare to download digital terrain data
without first looking within the saved library. This switch can be used to ensure
that the most up-to-date data is being used in the analysis at the expense of
repeated down load time. It is recommended that this setting is set to “0” to not
automatically download. For repeatability of the assessment it is recommended
that a Library be created. Updates, when available from the download site, can be
included by starting a new library periodically.
MLCCGET
D
Similar to the mdemget option, the mlccget setting is used to force ERCBflare to
download land classification data without first looking within the saved library.
This switch can be used to ensure that the most up-to-date data is being used in
the analysis at the expense of repeated down load time. It is recommended that
this setting is set to “0” to not automatically download. For repeatability of the
assessment it is recommended that a Library be created. Updates, when available
from the download site, can be included by starting a new library periodically.
Non-Default Settings
The non-default settings area is a listing of options the user can select to perform
calculations outside of the ERCB (Alberta) regulatory requirements. When
settings are entered different than the default setting, a warning message is
displayed on all output pages indicating that the modelling is not appropriate for
3 ERCBflare Inputs
28
FT
ERCB approval/applications. Some of the settings are explicit (e.g., average wind
speed) and others are the result of selections on other input pages (example land
cover). The non-default settings block is shown in the figure below.
R
A
When an entry is made in the non-default inputs, the entry is highlighted to notify
the user that the value is different than the default and a warning comment is
displayed adjacent to the entry. In addition, all output pages (see below) are
branded with a warning message that a non-default setting was used. For
regulatory use in Alberta, all settings must have their default value.
D
Ambient Temperature
The average annual temperature for Alberta was determined to be 5 °C. The
average temperature is used for routine flare modelling for determining
combustion chemical properties and energy balance. The raw gas and fuel gas
may originate (either in a pipeline, facility or below ground) at higher or lower
temperatures. However, ERCBflare assumes there is sufficient piping length
between the stream source and the flare tip that the flared gas temperature will be
equal to the ambient temperature.
For non-routine flaring, the air dispersion modelling and combustion calculations
make use of hourly meteorology. The ambient temperature is set to the ambient
temperature in the meteorological data set on an hour-by-hour basis.
3 ERCBflare Inputs
29
Average Ambient Wind Speed
The average ambient wind speed (m/s) for Alberta was determined to be 3.5 m/s.
The wind speed is used for routine flare air dispersion modelling for determining
the combustion efficiency based upon the ratio of stack velocity to ambient wind
speed velocity. For routine flaring, air dispersion modelling is performed using
the average meteorological conditions. Flare stacks should be designed to achieve
good combustion efficiency under the full range of expected wind speeds to
ensure that ambient air quality measurements would not exceed objectives.
PG Stability Class
FT
For non-routine flaring, the air dispersion modelling and combustion calculations
make use of hourly meteorology. The ambient wind speed is set to the ambient
wind speed in the meteorological data set on an hour-by-hour basis.
A
The PG (Pasquill-Gifford) atmospheric stability class is used in conjunction with
the ambient wind speed. The ambient wind speed is typically measured at the
10 m anemometer height and is converted to a flare stack tip height using wind
profile exponent powerlaw. The average meteorological condition is assumed to
be neutral stability. The PG stability class entered here is not used for air
dispersion modelling.
R
For non-routine flaring, the air dispersion modelling and combustion calculations
make use of hourly meteorology. The PG stability is set to the ambient wind
speed exponent powerlaw in the meteorological data set on an hour-by-hour basis.
Class
1
2
3
4
5
6
Class A
Class B
Class C
Class D
Class E
Class F
D
PG
Stability
Description
Extremely unstable conditions
Moderately unstable conditions
Slightly unstable conditions
Neutral conditions
Slightly stable conditions
Moderately stable conditions
Wind Speed
Exponent
0.07
0.07
0.10
0.15
0.35
0.55
Ref: US EPA Industrial Source Complex Model (ISCST) rural wind speed exponents from Irwin
(???)
Qmin Fraction of Qmax
The flaring assessment provided within ERCBflare uses a three-flowrate analysis
as opposed to the maximum design rate flowrate analysis. While the maximum
flowrate is an important consideration to determine environmental consequence
since it is frequently associated with maximum mass of pollutant emissions rates,
3 ERCBflare Inputs
30
typically flaring operations are well below the maximum rate. The high energy
associated with the maximum design rate results in a high plume rise and can is
not necessarily the worst case consequence. Average flowrates or low flowrates
from the same diameter flare tip may result in lower plume rise and downwash
conditions producing high groundlevel concentrations. Typical operation of flares
is about 1/8th of the maximum design flowrate. ERCBflare uses 0.125 of Qmax as
the Qmin flow rate by default. For non-default assessments, a different value can
be entered for the Qmin flowrate.
SO2 1h Air Quality Objective
FT
The one hour average ambient air quality objective for SO2 in Alberta is
450 µg/m³ (ESRD 2011). For other jurisdictions, the ambient objective can be
changed. This changes the linkages to references the objective in the RBC and
graphics.
H2S 1h Air Quality Objective
A
The one hour average ambient air quality objective for H2S in Alberta is 14 µg/m³
(ESRD 2011). For other jurisdictions, the ambient objective can be changed.
This changes the linkages to references the objective in the RBC and graphics.
Receptor Resolution for Maximum Concentration
D
R
The ERCBflare analysis uses a logarithm distribution of 100-receptor points
between 100 m and 10,000 m from the source. Near the source, the receptor to
receptor distance is about 5 m whereas far from the receptor the receptor distance
is about 500 m. ERCBflare will perform multiple iterations to determine the
maximum concentration until a minimum receptor resolution has been achieved.
The default resolution is set to 20 m. Therefore, if the predicted maximum
concentration is near the source ERCBflare does not iterate, but if the predicted
maximum concentration is far from source, ERCBflare may require at second
iteration. A resolution of 1 m may require three iterations and will therefore
increase run-times by a factor of three.
Raw, Fuel and Lift Gas Temperature Before
Combustion
The ERCBflare program makes the assumption that the Raw gas, Fuel gas and
Lift gas temperature before combustion is at ambient temperature. This
assumption is based upon the acknowledgement that there is a measurable length
of piping between process operations and the flare tip at which the gases will start
3 ERCBflare Inputs
31
to achieve an equilibrium temperature; and, it is a reasonable conservative
assumption for the starting point of the energy balance. The starting temperature
will have a small effect on the energy balance of the energy contributing to plume
rise. When the initial gas temperature is equal to the ambient temperature, more
energy may be required to create the same plume rise at different times of the year
(winter vs summer). Alternatively, if the initial gas temperature is a fixed value
and not a function of ambient temperature, then the temperature difference
between the initial gas temperature and the ambient temperature results in a net
increase in plume rise.
FT
In ERCBflare, the initial gas temperature (TGinit) is set to be equal to the ambient
temperature (Ta). TGinit can be selected to a non-default starting condition on the
iSTART page. Setting TGinit to a temperature different than the Ta temperature
setting, forces the initial temperature for the Fuel gas, Raw gas and Lift gas to be
a constant temperature, even when the ambient temperature changes through a
typical range of -40 °C to +30 °C.
Run Flags
R
A
The run flags are used to limit the number of scenarios run by ERCBflare when
the air dispersion modelling buttons are click on the oMODELLING page. The
run flags are listed at the top of the table on the oAERSCREEN page (row 3).
The run flags are set internally according to the Non-Routine flare air dispersion
modelling (3 scenarios of SO2 and 3 scenarios of H2S are performed) or Routine
flare air dispersion modelling (1 scenario of SO2 and 1 scenario of H2S are
performed). Under all scenarios both a parallel and a terrain case scenario are
always performed. Using the table below, run flags can be forced on or off postreading of the run flags on the oAERSCREEN page. The entry for the run flags
on the iSTART page is a bit-wise comparison.
D
To run all six scenarios, enter MYRUN=63. To run only scenario 1 and 4, enter
the MYRUN=9. The default setting is MYRUN=0 which forces ERCBflare to
use the settings on the oAERSCREEN page.
Scenario/Case
Typical
MYRUN setting
1
2
3
4
5
6
SO2 Max
SO2 Average
SO2 Low (Qmax/8)
H2S Max
H2S Average
H2S Low (Qmax/8)
1
2
4
8
16
32
A small switch block is provided to the right of the Non-Default Settings entry
area that can be used to simplify the entry of MYRUN values. Using the mouse
to click on the tiles below each of the 1 though 6 cases, selects the case to be run.
3 ERCBflare Inputs
32
If all of the cases are blank, then all of the cases are run, default setting. In the
example below, Case 1 and Case 4 are selected to be run.
User Period Selection (Annual or Month)
FT
The mflmon flag is automatically raised based upon the selection of the modelling
period on the iFLARING page. If the user over-rides the default, option, then the
result is displayed as a warning on the iSTART page.
Transient Blowdown distribution of mass option
A
The mdist flag is based upon the selection of the transient blow down exponential
curve discretization option available on the iFLARING page. The default is to
divide the exponential curve into steps of equal mass. However, the user may
choose to use steps of equal duration. When this non-default selection is used, the
warning is summarized on the iSTART page.
Blowdown User Entry of Qmax
D
R
The ERCBflare transient source model is based upon the initial volume and
conditions of the gas. The source model calculates the maximum release rate and
the total volume to be released. Advanced users may use external source model
that predict these two variables. The QMAX (source maximum mass flowrate)
and QTOTAL (source total volume) can be entered on the iFLARING page and
over-riding the ERCBflare source model. This non-default entry is summarized
as a warning flag on the iSTART page.
Blowdown User Entry of Qtotal
The ERCBflare transient source model is based upon the initial volume and
conditions of the gas. The source model calculates the maximum release rate and
the total volume to be released. Advanced users may use external source model
that predict these two variables. The QMAX (source maximum mass flowrate)
and QTOTAL (source total volume) can be entered on the iFLARING page and
over-riding the ERCBflare source model. This non-default entry is summarized
as a warning flag on the iSTART page.
3 ERCBflare Inputs
33
Land Use around the Well
iFACILITY
FT
The ERCBflare screening model requires the land use surrounding the
well/facility to be categorized into one of 8-screening categories according to
ERSD Air Dispersion Modelling Guideline (ESRD 2013). The land use is
determined based upon the entry of the flare stack coordinates and the determined
nearest associated land use point in the database. An advanced user may select a
different land use category for assessment. However, this non-default selection is
flagged on the iSTART page.
The iFACILITY page is used to specify well-site or facility level inputs.
Administrative-Operator
D
R
A
In the Administrative input section, shown below, input is required to identify the
licensee and contact information.
For routine flaring assessments, the modelling is not presented for an ERCB
approval process and therefore the Administrative section is not strictly required.
However, the completion of these entries forms a good engineering practice of
documentation.
3 ERCBflare Inputs
34
Administrative-Consultant
The consultant or company personnel responsible for entering information in the
ERCBflare is entered in the Consultant section. The ERCB permit review
personnel may contact the consultant should a point of clarification be required
during the review whereas any substantial deviation or information gap will be
directed through the Operator contact who submitted the flaring approval.
Well or Facility Details
Well Name
FT
The Well/Facility details section is required for a non-routine flaring application
to the ERCB. The required details are described below. These details ensure that
the flare Operator and ERCB are referring to the same well for the permit.
A
Existing wells have a licenced name. The licence name is often a combination of
the reservoir, zone or other name followed by the surface or down-hole location.
The exact name as it appears on the licence should be entered, since this name and
licence number need to agree in the ERCB well database. New wells may have
been given a temporary name by the ERCB.
R
Licence Number
D
The ERCB licence number of the well should be entered. This is typically a 5digit numeric value. New wells may not have a licence number yet, in which case
a suitable entry like “new well” may be entered.
Unique Well Identifier (UWI)
The unique well identifier (UWI) is the standard well identification that was developed
for the petroleum industry by the Geoscience Data Committee of the Canadian Petroleum
Association (CPA) and has been adopted by the oil and gas regulatory agencies of the
four western provinces and federal areas. It consists of 16 characters, which make up four
basic components:
i. legal survey location
ii. survey system code
iii. location exception code
iv. event sequence code
Together these define the approximate geographical location of the bottom of a drill hole
and a specific drilling or producing event at the drill hole.
3 ERCBflare Inputs
35
The unique well identifier, although based on the legal survey position of a well, is
primarily for identification rather than location. The location component describes the
bottomhole location of the well, not the surface position of the well.
Oil or Gas
The Oil or Gas flag is used by ERCBflare to check inputs for soundness. If the
Gas is selected then the reservoir engineer would provide the ERCBflare
Consultant with only expected gas flowrates and volumes. These are entered on
the iFLARING page.
A
FT
If the Oil well is selected, then the well will typically produce oil, solution gas
and produced gas in various amounts. The reservoir and/or production engineer
would provide the ERCBflare consultant with oil production rate and maximum
GOR (gas to oil ratio taking into account both produced gas and solution gas).
The Oil selection is used by ERCBflare to check that the entered maximum
flaring rate on the iFLARING page is not less than product of oil rate and GOR
(these are entered on the iFACILITY page). Additionally, ERCB D060 has
volume allowances for Oil versus Gas wells; these checks are made on the
oSUMMARY page in the D060 Permit Conditions checklist group.
Critical Well Classification
R
A critical sour well is a sour gas well that could potentially release large quantities
of hydrogen sulphide (H2S), causing significant harm to nearby people.
All applications to the ERCB to drill oil or gas wells must take into account the
possibility of encountering sour gas. If the ERCB’s first evaluation shows that
there may be H2S, then the application is examined further.
D
The ERCB uses two major criteria to determine if a sour well is to be classified as
critical:
• how close the well is to an urban centre or public facility, such as a major
recreational facility, and
• the potential H2S release rate during the drilling stage.
The potential H2S release rate is determined by both the percentage of H2S in the
gas and the rate at which gas that can be delivered to the surface. This is measured
in cubic metres per second at standard pressure and temperature.
For example, a well may have both a weak flow of gas with only 1 per cent H2S
content but still be critical if it is very close to a town. But a gas well with 10 per
cent H2S content located in a remote location without people nearby might not be
classified as critical.
3 ERCBflare Inputs
36
The Critical Well selection in ERCBflare is an important flag for both the
Operator and the ERCB approval review process. However, the selection does
not impact the calculations performed by ERCBflare and is therefore cosmetic.
Formation(s) and Zone(s) to be Tested
FT
A well being drilled will have a target formation and zone. Wells being tested or
enhanced may have multiple formations and zones. The ERCB database tracks
the activities and gas compositions of the various zones and formations. This
entry is used by ERCB to confirm the Operator activities correspond to
ERCBflare assessment and to verify/validate basic information used in the
assessment.
Number of Zones to be Tested
R
A
The ERCBflare prompts the user for the number of zones being tested. The
number of zones tested is descriptive of the operations planned by the Operator
for the well and flaring activities. The ERCBflare spreadsheet must be completed
for each zone tested. The flaring activities associated with the well will be limited
by total volume flared according to the number of zones to be tested. The total
volume flared for each of the completed ERCBflare spreadsheets on the
individual zone test must be less than or equal to the total allowable for all zones
volume listed on the oSUMMARY page in the ERCB D060 Permit Conditions
group.
Lahee Classification
D
The volume allowance for an individual zone is a function of the Lahee
Classification for the well. A listing of the descriptions of the Lahee
classifications are provided on the LAHEE page. Adjacent to each description is
the Tier number corresponding to the ERCB D060 volume allowance. The
oSUMMARY page in the ERCB D060 Permit Conditions group compares the
volume flared for this zone to the Tier allowance.
Surface Location
The surface location of the well is entered as a legal land survey description
according to the Dominion Land Survey System. The format of the surface
location should be:
(LSD-SEC-TWP-RGE-W?M)
Corresponding to:
3 ERCBflare Inputs
37
Legal subdivision
SEC
Section
TWP
Township
RGE
Range
W?M
Meridian
Legal subdivisions are numbered from
1 to 16. A legal subdivision measures 402 by 402 metres
(1320 by 1320 feet).
Sections are numbered 1-36. A section measures 1.609
by 1.609 km (1 mile by 1 mile).
Townships are numbered 001-126. A township
measures 9.7 km (6 miles) north to south
Ranges are numbered from 1 to 30. Note that West of
the sixth meridian contains a maximum of 14 ranges
only. A range measures 9.7 km (6 miles) east to west.
For the purposes of survey locations, the province of
Alberta is subdivided into three areas defined by the
meridians of longitude. For unique well identifier
purposes, these are referred to as west (W) of the fourth,
fifth, and sixth meridians.
Acceptable values: W4, W5, W6
FT
LSD
Mapping Projection
A
Surface coordinates are entered for the flare. For well tests, the flare location is
nominally assigned to the well location unless its specific location is known. In
the latter case, the specific location should be used for the surface coordinates of
the flare. The coordinates entered are associated with a mapping projection.
ERCBflare accepts the mapping projections (datum) in the table below. The
project is used to convert the user entered coordinates to geographic (latitude and
longitude) for internal use in determining the closest land cover data in the
ERCBflare database.
D
R
Geographic (WGS84)
10TM (NAD83)
UTM Zone 8 (NAD83)
UTM Zone 9 (NAD83)
UTM Zone 10 (NAD83)
UTM Zone 11 (NAD83)
UTM Zone 12 (NAD83)
UTM Zone 13 (NAD83)
Surface Coordinates of Flare
The (X,Y) pair of surface coordinates are entered for the flare location. If a well
test flare is being assessed and the exact flare location is not known, then the well
surface flare location can be entered. The coordinates must correspond to the
mapping projection entered in the selection above.
3 ERCBflare Inputs
38
Flare Base Elevation
The base elevation of the well is the terrain elevation at the well surface (or flare
location). The base elevation is an important variable since it is used to determine
the atmospheric pressure for the combustion calculations. The terrain elevations
vary dramatically across the province and can result in non-negligible changes in
the emissions or source parameters. The flare base elevation is also an important
entry relative to the terrain elevations used. Improper entries can result in
artificial increase in effective stack height or lowering of stack height.
FT
Land-use Characterization
The Land Cover is based upon available data for Canada (www.geobase.ca; circa
2000; based upon Canadian Forestry datasets and Canada Agriculture datasets).
Land cover has been summarized for Alberta, B.C. and Saskatchewan based upon
a gridded screening approach.
A
Land cover (see Figure 2) was sampled within a cell size of 10km × 10km at nine
locations. Land cover was reduced to the ESRD air dispersion modelling
screening land cover sets (see Table 1 and Table 2). The most frequent land cover
for the cell is represented at the location of the cell centre. This methodology
creates a data set that is small enough to be easily contained within the
spreadsheet while still well representing the dominant land features. The coarse
data is essentially pre-averaged therefore ERCBflare can just select the closest
grid point as a representative LCC (see Figure 3).
D
R
ERCBflare uses the flare location coordinates to perform a search for the closest
land cover sample in the database (see Figure 3). If the flare location falls outside
of the database land cover range or if the user wishes to perform sensitivity testing
on the flaring predictions using a different land cover, then the user can select a
non-default land cover. For approvals using the ERCBflare spreadsheet, the
default land cover is required.
3 ERCBflare Inputs
39
FT
A
R
D
Figure 2:
3 ERCBflare Inputs
High Resolution Land Cover for Alberta
40
Land Cover Classification
D
R
A
FT
Table 1:
3 ERCBflare Inputs
41
FT
Screening Land Cover Classification
D
R
Table 2:
Screening Land Cover for Western Canada
A
Figure 3:
The land cover is used to define the appropriate AERSCREEN meteorological data
set to use in the air dispersion modelling. The AERSCREEN utility program
MAKEMET.exe was used to prepare seasonal screening data sets according to
Table 3. The set of screening meteorological data sets are included in the install
package for ERCBflare. Each file contains approximately 2500 variations of
3 ERCBflare Inputs
42
meteorology. Both AERMOD profile and surface files are provided so that they
can be used for other assessments or sensitivity analysis.
Table 3:
Screening Meteorological Variations Used for
AERSCREEN- MAKEMET.exe
Variable
decid
culiv
grass
urban
swamp
water
desert
0.5
10
0.5
10
0.5
10
0.5
10
0.5
10
0.5
10
0.5
10
0.5
10
SPRING -- Min Temp. (K)
Max Temp. (K)
239.4
239.4
239.4
239.4
239.4
239.4
239.4
239.4
306.1
0.12
306.1
0.12
306.1
0.14
306.1
0.18
306.1
0.14
306.1
0.12
306.1
0.12
306.1
0.3
0.7
1.3
0.7
1
0.3
0.03
0.4
0.05
1
1
0.1
0.2
0.1
0.0001
3
0.3
Albedo
Bowen Ratio
Zo, roughness (m)
FT
conif
Minimum wind speed (m/s)
Anemometer height (m)
SUMMER --.Min Temp. (K)
Max Temp. (K)
271.9
310.5
271.9
310.5
271.9
310.5
271.9
310.5
271.9
310.5
271.9
310.5
271.9
310.5
271.9
310.5
Albedo
Bowen Ratio
0.12
0.3
0.12
0.3
0.2
0.5
0.18
0.8
0.16
2
0.12
0.1
0.1
0.1
0.28
4
1.3
Zo, roughness (m)
Max Temp. (K)
Albedo
Bowen Ratio
Zo, roughness (m)
1.3
0.2
0.1
1
0.2
0.0001
0.3
238.9
306.5
238.9
306.5
238.9
306.5
238.9
306.5
238.9
306.5
238.9
306.5
238.9
306.5
238.9
306.5
0.12
0.8
0.12
1
0.18
0.7
0.2
1
0.18
2
0.12
0.1
0.14
0.1
0.28
6
A
FALL -- .Min Temp. (K)
0.8
0.05
0.01
1
0.2
0.0001
0.3
226.5
226.5
226.5
226.5
226.5
226.5
226.5
226.5
Max Temp. (K)
Albedo
290
0.35
290
0.5
290
0.6
290
0.6
290
0.35
290
0.35
290
0.2
290
0.45
1.5
0.01
1.5
0.001
1.5
1
1.5
0.05
1.5
0.0001
6
0.15
R
1.3
WINTER -- Min Temp. (K)
1.5
1.3
1.5
0.5
D
Bowen Ratio
Zo, roughness (m)
Operations to be Conducted
The ERCB approval review requires a description of the well test operations
(cleanup/workover/testing/etc) to be performed. This is brief description of the
work that will be completed and may include more than one zone.
Total Volume of Raw Gas to be Flared
During the operations to be conducted (see above description) the estimated total
volume of raw gas to be flared during clean-up and testing of all zones is entered.
The total volume is compared to the ERCB D060 Permit Conditions. This entry
provides a soundness check between the expectations of the Operator for the well
3 ERCBflare Inputs
43
clean up and the limits of prescribed by D060 for reducing flaring and conserving
gas.
Total Estimated Days with Flaring for ALL Zones
The total work over of the well and cleanup is entered. The entry is compared to
the D060 Permit Conditions limiting the duration of activities and flaring.
Well tied into Production Facilities
FT
This entry is used by ERCB for consideration of future activities at the well, such
as adding pipeline to tie the well or feasibility of performing in-line well testing.
Additionally, this information is valuable in consideration of conservation of gas.
Information on the Feasibility of In-Line Well
Testing Attached
A
If the well is tied into existing facilities, it may be possible to perform an in-line
well test rather than a flared well test. An engineered feasibility report should be
attached to the application for well with a tie-in.
R
Previous Flaring/Incineration Permit
D
If the well is an existing well, then previous operations may provide a historical
perspective on the necessity of the work over, flaring and emissions. Provide the
previous permit number and date if the approval was acquired within the past
twelve months.
Requested Approval Start Date/End Date
Scheduling non-routine well test flaring is often a moving target. ERCBflare
requires a start date and end date when the flaring is forecast to occur. The
duration may not exceed 60 days.
The start and end date are important conditions for the well test flare permit.
3 ERCBflare Inputs
44
Emergency Response Details
In addition to consideration of energy conservation, flaring reduction and
ensuring environmental objectives will be met, the ERCB requires that the
Operator also plan sour gas emergency response plan (ERP) details. One step in
the emergency response planning is the determination of the emergency planning
zone. The ERCBH2S model is expected to be run to determine the emergency
planning zone (EPZ) size.
H2S Release Rate
FT
In the event of an accident at the well, gas can be released from the formation
under its own pressure. Depending upon the operations, gas can reach the surface
through the casing (casing diameter source) or through tubing (tubing diameter
source). The release rate of H2S is the product of the gas composition and the
estimate absoluted open flow (AOF) for the well. Because there is no
backpressure, the AOF is generally several times larger than the flaring rate for
the well test.
A
The AOF is entered in ERCBflare and is compared to the maximum flaring rate as
a check of proper inputs.
Emergency Response Planning Zone Distance
R
The emergency response planning zone distance is entered in ERCBflare to
ensure that the Operator has considered ERP in advance of the planned
operations.
D
Is an ERP Required?
The licensee or operator must meet emergency response plan requirements for
sour wells. The plan must incorporate provisions for the temporary equipment as
appropriate. (See ERCB D071: Emergency Preparedness and Response
Requirements.)
The licensee must submit a sour well site-specific drilling and/or completion ERP
to the ERCB for approval in accordance with Direction 071 or for any other
situation in which the ERCB determines that a plan is required.
3 ERCBflare Inputs
45
Surface development within the EPZ influences whether an ERP is required.
Surface development includes residences that are required to egress through the
EPZ and residences immediately adjacent to the EPZ.
In the event that a licence application requires a public hearing, the licensee is
expected to develop and provide an ERP to the ERCB that has been deemed
technically complete prior to issuing a notice of public hearing.
A sour well site-specific drilling and/or completion ERP may be used for testing,
workover, or well servicing operations on that well for a period of up to one year
after ERCB approval provided that those operations
FT
ERP Reference Number
If an ERP is required, it has been completed and approved by the ERCB, then it
will have been assigned an ERP reference number. Enter the reference number
for the ERP.
A
Fluid Production Details
D
R
If Oil well was selected on the iFACILITY Well Details inputs, then the well will
typically produce oil, solution gas and produced gas in various amounts. The
reservoir and/or production engineer would provide the ERCBflare consultant
with oil production rate and maximum GOR (gas to oil ratio taking into account
both produced gas and solution gas). The Oil selection is used by ERCBflare to
check that the entered maximum flaring rate on the iFLARING page is not less
than product of oil rate and GOR (these are entered on the iFACILITY page).
Additionally, ERCB D060 has volume allowances for Oil versus Gas wells; these
checks are made on the oSUMMARY page in the D060 Permit Conditions
checklist group.
Anticipated Fluid Production
Enter the anticipated fluid production rate.
Gas to Fluid Ratio
The gas to fluid ratio for oil wells is used to estimate the total gas available to be
flared using
[Gas Rate] = [Fluid Produced] × [GOR]
3 ERCBflare Inputs
46
The maximum flaring rate used for the ERCBflare assessment should not be
greater than [Gas Rate].
iFLARING
FT
Careful attention should be applied to the estimate of maximum fluid production
and maximum GOR. Since both values are maximums and inherently rare events,
the product of the two is a very conservative estimate and a very rare event. The
assessment of flaring based upon the extreme maximum gas rate may lead to an
over-sized flare (a flare that has a large diameter to accommodate the worst case
flow but is operated a much lower flowrates.) Oversized flares may lead to
flaring combustion in-efficiency resulting in an exceedance of the ambient air
quality objectives.
A
The iFLARING page is the page where the ERCBflare source parameters are
entered for the specific flaring event being assessed. Whereas, the iFACILITY
page contained global considerations of the overall work-over operations, the
focus on the iFLARING page is on the conditions of an individual flaring
scenario.
R
Flaring Details
D
The Flaring Details group is used by both Non-Routine and Routine flaring
assessments.
3 ERCBflare Inputs
47
Subject Zone/Source
The well work over may involve several zones. On the iFLARING page enter the
single zone that will be considered for the flare event. For sources other than
wells, enter a brief description.
Scenario Name
FT
The scenario name is descriptive name used by the Consultant to reference the
flare event. It is also used to identify flaring events in the iBATCH database. In
this case, it is helpful if the Scenario Name is unique, but it is not a requirement.
Time of Year to Model the Flare
The default assessment for ERCBflare is to consider the entire meteorological
period so that a sufficient range of meteorological variations is considered in the
screening assessment.
D
R
A
The ERCBflare assessment can be performed using a screening meteorological
data set or a 5-year site-specific meteorological data set that may have been
developed. If an annual assessment is selected, then the either the screening or
the site specific meteorological data set can be used in the modelling.
Alternatively, to possibly take advantage of favourable meteorological conditions,
the flaring assessment can be performed by limiting the meteorology to only the
month of the expected flare event (the mid-point of the start and end dates for the
flaring entered on the iFACILITY page). In this latter case, the assessment must
be performed using the site specific meteorology data set, since the screening data
set is not monthly but only seasonal and it does not contain sufficient number of
variations. Whereas the 5-year site specific meteorological data can be reduced to
five 3-month set of variations and provide more than 8760 variations. The
number of variations tested is displayed on the oAERSCREEN output page.
ERCBflare blocks output for monthly assessments and risk based output when the
number of variations is less than 8760.
The default centre-month for the assessment is determined by the mid-point of the
start and end dates for the flaring entered on the iFACILITY page. Sensitivity
testing can be performed by modelling alternative months, but these months will
be flagged as non-default.
For non-routine flaring, the AERSCREEN-MAX(HbH) modelling button (see
oMODELLING) will be disabled when the monthly assessment option is used.
The AERSCREEN-MAX is used for preliminary and screening assessments,
therefore it remains active. Note that monthly assessment using the provided
screening meteorological data sets will produce erroneous results since the format
3 ERCBflare Inputs
48
of the AERSCREEN date and time do not match calendar dates. The option is
available for future or advanced users.
Seasonal predictions are provided using the screening meteorological data sets
when the Annual option is used. Although, the risk based criteria are displayed in
this case when N<8760, the screening meteorological dataset analysis uses the
maximum concentrations only.
Flare Stack Tip Exit Height
The flare stack height above grade is entered.
FT
The meteorological conditions at the flare stack height are used for combustion
and source modelling with a minimum of 10m. The site specific meteorological
data sets should be extracted at the flare stack exit height for the best estimate of
the meteorology at that height when 3D wind fields are used. ERCBflare always
corrects for the difference between the meteorological anemometer height and the
stack. Therefore, if the site specific data was extracted at the stack height, then
the correction in the wind speed will be zero.
A
Flare Stack Tip Exit Diameter
R
The inside diameter of the flare stack at the exit height are entered. If the flare
stack exit is not circular, then an equivalent diameter can be entered based upon
the exit area:
𝐷𝐷 = �𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴 × 4/𝜋𝜋
This assumption will maintain the proper momentum of the flare gas at the exit.
D
Requested Maximum Raw Gas H2S Concentration
For the Subject Zone, the maximum H2S concentration of the gas composition
(entered in percent, %). The maximum H2S concentration does not have to match
the reference composition provided in the gas composition (lower portion of
iFLARING page).
Continuous/Steady or Transient
The ERCBflare tool can be used to assess steady/continuous emissions or
transient emissions resulting from a vessel or pipeline blowdown. The selection
3 ERCBflare Inputs
49
is made in the Continuous/Steady or Transient data group. Transient flare inputs
and modelling are discussed in Chapter 6.
FT
Continuous or Short-Term Steady
A
For non-routine flaring assessments three emissions modelling scenarios are
considered: maximum (Qmax), average (Qave) and a low rate (Qmax/8).
Therefore two sets of inputs are required to define Qmax and Qave. For routine
flaring assessment only the Qmax scenario is required.
R
Maximum Raw Gas Flow Rate
D
The maximum raw gas flow rate is the peak gas flow rate expected from the nonroutine well test or upset/emergency scenario. For routine flare air dispersion
modelling the maximum gas flow rate represents the scenario flow rate. Routine
flare air dispersion modelling may require multiple assessments to ensure that the
complete range of flare flow rates results in adequate combustion efficiency to
avoid ground level exceedance of the ambient air quality objectives. Assessment
of only the maximum flow rate, may result in an over-sized flare (a flare that has
a large diameter to accommodate the worst case flow but is operated a much
lower flowrates). Oversized flares may lead to flaring combustion in-efficiency
or flame instability and downwash resulting in an exceedance of the ambient air
quality objectives.
Average Flow Rate Parameters
The average flow rate is determined by conservation of mass and entering any
two of: the total volume flared, the average flow rate, or flaring duration. Using
3 ERCBflare Inputs
50
conservation of mass, the missing variable is readily calculated. Specifying all
three can lead to errors if not entered correctly.
𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉 = 𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹 × 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷
Volume
The volume is the total volume to be flared during the flaring of this zone.
Flow Rate
The flow rate is the average flow rate during flaring of this zone. An estimate of
the average flow rate is Qmax/2.
Duration
FT
The duration of actual flaring of this zone needs to be entered. Don’t enter the
time elapsed since the start of operations. For instance, in flaring occurs during
daylight hours only (12-hours) and flaring occurred on two days of operations,
then the duration is 24-hours.
A
Transient Source
See Chapter 6: Transient Blowdown Sources
R
Fuel Gas
D
The fuel gas input group is greyed out if no fuel gas is added. When fuel gas is
added, enter the fuel gas to raw gas ratio for each of the Qmax, Qave and Qmax/8
scenarios.
Below the input locations for each fuel gas ratio, an information or warning
message appears. The message may contain important information such as the
required fuel gas ratio to meet D060 minimum heating values. The messages may
be influenced by other iFLARING entries such as gas composition, flow rates, lift
gas and/or flare assist.
3 ERCBflare Inputs
51
Lift Gas
The Lift Gas input group is greyed out if no lift gas is used to supplement the flow
to the surface for the well. It is assumed that lift gas is homogenously mixed with
the raw gas brought to the surface.
A
Flare Assist
FT
In addition to the amount of lift gas entered in this group, the lift gas gascomposition must be entered on the lower part of the iFLARING page in the Gas
Composition input group. Lift gas may be inert or hydrocarbon. If the lift gas is
inert, then additional fuel gas may be required. The warning messages below the
fuel gas ratio inputs can provide important information regarding extra fuel gas
requirements.
D
R
Flare assist is not commonly used for non-routine flaring but may be an important
aspect of routine flaring at a facility. ERCBflare incorporates the effects of either
or both steam assist and air assist. Although commonly used to influence the
visual impacts of flaring (i.e., reduce particulates) through the addition of
momentum at the source and some latent heat, flare assist has been found to
reduce flare efficiency by reducing of the net heating value of the flared gas.
Flare assist may reduce particulate formation, however, the formation of
particulates is not an accurate measure of flaring efficiency. Because both air and
steam introduce composition components beyond theoretical stoichiometric
conditions, flare assist efficiency studies indicate a reduction of flare combustion
efficiency. While small rates of flare assist don’t significantly impact combustion
efficiency, and the added momentum may assist the flare, steam to raw gas rates
above 2 can reduce combustion efficiency below ideal levels. ERCBflare
combines the flare assist streams into a single flared gas stream while conserving
momentum and energy. ERCBflare uses a combustion efficiency model based
upon the net heating value of the flared gas stream in relation to ambient wind
speed. Since flare assist increases momentum but reduces heating value, a
competing effect results especially at high assist rates.
3 ERCBflare Inputs
52
Steam Assist
Flare Assist: Number of Ports
FT
Steam flare assist may consist of many injector ports. It is assumed that all of the
ports are will increase upward momentum. Ports located around the flare exit
diameter as well as ports in the centre of co-flowing injectors are included.
Flare Assist: Port Diameter of a Single Injector
All of the injection ports are assumed to be the same diameter. If injectors vary in
diameter then an equivalent diameter can be entered based upon the total area of
all ports.
𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇_𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴 4
×
𝑁𝑁𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝
𝜋𝜋
A
𝐷𝐷 = �
Entry Mass rate or Volume Rate?
R
Typically steam assist is specified as a mass flow rate. ERCBflare allows for
either mass or volume flow.
Steam Assist Pressure
The quality of the steam is required. ERCBflare assumes saturated steam and will
calculate the critical pressure. If the pressure is known, it should be entered.
D
Steam Assist Temperature
The quality of the steam is required. Typical low pressure steam is about 150 °C.
ERCBflare assumes saturated steam and will calculate the critical temperature.
Steam Assist
ERCBflare allows steam assist flow rates to vary with flaring scenario rates
(Qmax, Qave and Qmax/8). Because of the significant impact flare assist rates
have on the flame combustion and flame instability (i.e., blowout or flammability
limits) flare assist rates must be tuned to the raw gas flow rates.
3 ERCBflare Inputs
53
Air Assist
Air Assist: Number of Ports
Air flare assist may consist of many injector ports. It is assumed that all of the
ports are will increase upward momentum. Ports located around the flare exit
diameter as well as ports in the centre of co-flowing injectors are included.
Air Assist: Port Diameter of a Single Injector
All of the injection ports are assumed to be the same diameter. If injectors vary in
diameter then an equivalent diameter can be entered based upon the total area of
all ports.
Air Assist Rate
𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇_𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴 4
×
𝑁𝑁𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝
𝜋𝜋
FT
𝐷𝐷 = �
A
The air assist flow rates are entered as a volume flow rate at reference conditions
(×10³m³/d, 15°C and 101.325 kPa). ERCBflare allows air assist flow rates to vary
with flaring scenario rates (Qmax, Qave and Qmax/8). Because of the significant
impact flare assist rates have on the flame combustion and flame instability (i.e.,
blowout or flammability limits) flare assist rates must be tuned to the raw gas
flow rates.
R
Flare Assist Flammability Check
D
Below the Air Assist flow rate entries a warning message may appear for each of
the flaring scenario rates (Qmax, Qave and Qmax/8) indicating whether the flare
is considered over-assisted. The flare is considered over-assisted when the flaring
efficiency is predicted to drop below 98%.
Gas Compositions
The gas composition input group provides entries for the raw gas, lift gas and fuel
gas streams. The flare assist streams of steam and air have known gas
composition and don’t require specific entry. The user should provide a
referenced raw gas stream composition that may or may not have the same H2S
concentration as the flare modelled amount. The reason for this is to allow for
regulatory H2S limit or fluctuating range of H2S. The ERCBflare model will renormalize the raw gas composition using requested H2S in place of the H2S
content specified in the Raw Gas composition.
3 ERCBflare Inputs
54
The fuel gas composition is similarly entered. Typically, fuel gas will be 100%
propane if the flaring is performed at remote locations. Otherwise, fuel gas may
be supplied by pipeline spec natural gas which contains typically >90% methane.
Representative pipeline quality natural gas composition is listed in Hubbard
(2009) and TransCanada (2012) are listed in Table 4.
Table 4:
Representative Pipeline Quality Natural Gas (Hubbard
2009)
Major & Minor
Components (Mole%)
Maximum
Alberta
Maximums
75
-------
-10
5
2
0.5
3-4
3-4
Not specified
Not specified
Not specified
Not specified
Not specified
Not specified
2% by Volume
0.25-1.0
gr/100scf
0.25-1.0
gr/100scf
5-20 gr/100scf
7.0 lb/mmscf
0.2-1.0 ppmv
23 mg/m³
FT
Methane
Ethane
Propane
Butanes
Pentanes plus
Nitrogen & other inerts
Carbon dioxide
Minimum
Trace Components
Hydrogen Sulphide
--
A
Mercaptan Sulphur
--
----
115 mg/m³
65 mg/m³
0.4% by volume
R
Total Sulphur
Water Vapour
Oxygen
Lift gas composition may be inert (e.g., nitrogen, N2) or may be a hydrocarbon.
The lift gas composition can be entered similarly to raw and fuel gas. The lift gas
composition should be referenced.
D
The gas composition should sum to 1.0000. If the entries do not sum to 1.0000, a
warning is displayed as shown below. This feature assists in preventing
typographical errors.
Gas Composition Total
The gas composition total is calculated below each stream composition provides a
check that the gas composition was entered correctly. It is assumed that the
inputs are normalized to 1.0000 and therefore, the Gas Composition Total should
display 1.0000 when the entries are completed. Blank cells are assumed to be
zero.
3 ERCBflare Inputs
55
Gas Analysis Reference
iTERRAIN
FT
The Gas Analysis Reference should be entered in the field at the bottom of gas
stream. The reference should include the following:
• location of where the gas was sampled
• date of the sample
• well name or id
The iTERRAIN page is used to input mapping and terrain elevations. The inputs
are split into two inputs group. The first input group is used to enter mapping
details the maximum terrain elevation in the study area (within 10 km of the flare
location). The second input group is used to enter the terrain elevations from the
flare location to the maximum elevation point.
D
R
A
The iTERRAIN page is designed to force the user to read a map rather than just
entry of digital terrain data. Careful consideration for the map details assists in
the understanding of the location of sensitive receptors and validation of receptor
terrain heights. At the same time, it is recognized that users may be using digital
versions of 1:50,000 map sheets and digital terrain elevations are also available.
If digital elevations are used, digital contours must be demonstrated represent
1:50,000 scale map sheet contours and maps must show the location of closest
distance to contour elevation selections.
Topographical Map Details and Maximum
Terrain
The map details input group is used to document the 1:50,000 topographical mapset used to for determining map elevations, contour intervals and maximum
elevation.
3 ERCBflare Inputs
56
FT
1:50,000 Topographical Map(s)
The list of 1:50,000 scale topographical map sheets used in the review of the
elevation contours.
Contour Interval Units of Map
A
Specify whether the map has units of feet or metres.
Contour Interval of Map With Maximum Elevation
R
Specify the contour interval on the map sheet. The goal is to determine the
maximum contour elevation within 10km of the flare location. 10km is considered
the maximum reasonable applicability of a plume model in elevated terrain for
flaring assessment.
D
Is Maximum Elevation A Surveyed Peak?
1:50,000 topographic maps mark surveyed peaks on hills but not necessarily on
minor hills or sub-peaks on hill complexes. The maximum location is a surveyed
peak then that elevation is used as the last point in the terrain list. If digital
elevation data is used to determine the maximum elevation, that elevation is
considered a surveyed peak. Otherwise, if a map sheet is used, it is not possible
to determine the peak between contour intervals, and the surveyed peak is set to
“no” and the distance to centre of the maximum contour is entered.
3 ERCBflare Inputs
57
Maximum Contour Elevation or Surveyed Peak
Elevation
Specify determined maximum elevation as a surveyed peak (or digital elevation
model maximum) or the maximum contour level within 10 km (200 mm radius on
1:50,000 scale topographical map.) The units of the maximum elevation or
contour level should correspond to the entry above (Contour Interval Units.)
FT
Map Distance from Flare to Centre of Maximum
Contour or Maximum Surveyed Peak Elevation
Enter the distance (in user measurement, mm) from the flare location to the centre
of the maximum contour or to the maximum surveyed peak elevation. If the peak
is not a surveyed peak then the maximum terrain elevation is determined as:
𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸 = 𝑀𝑀𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎 𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶 𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿 +
𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶 𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼
2
A
Worst Case Terrain and Complex Terrain
Criteria Parameters
D
R
The non-routine flaring air dispersion modelling uses an hour-by-hour source
characterization with the effective height of the source change for each hour of
the meteorology. Therefore, the impacts of terrain height on the air dispersion
modelling are also varying. Additionally, because terrain heights regularly
exceed stack height, most air dispersion modelling assessments invoke the
model’s complex terrain processing algorithms. ERCBflare requires worst case
terrain heights to be input for screening by the air dispersion model. ERCBflare
evaluates both parallel and elevated terrain air dispersion modelling predictions.
Based upon the user inputs in the Topographical Map Details input group, the
Worst Case Terrain input group has created a skeleton table to be completed by
the user. Listed in the table are the contour levels and input prompts for the map
distance (in mm for a 1:50,000 scale topographical map) to the nearest contour
line. The table displays common complex terrain criteria at each location/contour
level pair.
3 ERCBflare Inputs
58
FT
Below the Worst Case Terrain input group is a graphic showing the entries of
terrain height and distance compared to a schematic of the stack height and
complex terrain criterion for the stable atmosphere. Gaussian plume air
dispersion models are most likely to predict maximum concentrations at the point
where the complex terrain criterion crosses the terrain elevations.
D
R
A
The graphic also shows the effective flame height for the average meteorological
wind speed and temperature.
Figure 4:
3 ERCBflare Inputs
Worst Case Complex Terrain Graphic Showing Terrain
Elevations as a Function of Distance from the Source
Versus Stable Atmosphere Plume Rise
59
Complex Terrain Summary
A
FT
The Complex Terrain Summary Group lists summary information about the
terrain elevations relative to the stack height and stable plume rise predictions.
R
The iTERRAIN graphic is a good way to confirm the user entries for
terrain elevations are representative.
Get DEM Data – BUTTON
D
The Get DEM Data button is used for automatic terrain processing. When the
button is pressed, ERCBflare uses the UTM coordinates or Latitude/Longitude
coordinates entered on the iFACILITY page to determine the NTS mapsheets
required for modelling domain around the flare location. ERCBflare first looks in
the DEMLIB area (see iBIN page) for the mapsheet. If the file is not available it
will be downloaded automatically and placed into the DEMLIB library location.
ERCBflare will then process all of the required mapsheets and fill in the
iTERRAIN page based upon the mapsheet data. The raw data is displayed to the
right of the iTERRAIN main entry area. The DEM mapsheet data is 25 m
resolution.
The elevation of the flare location is determined from the DEM data using
triangulation. The elevation at the flare location is displayed to the right of the
iTERRAIN main entry area. The user should verify that the DEM data elevation
is similar to the site survey elevation entered on the iFACILITY page.
Differences between the iFACILITY page elevation and the DEM map data
3 ERCBflare Inputs
60
elevation can result in an artificial increase effective stack height or an artificial
lowering of the stack height. The melev=1 setting on the iSTART page is used,
the elevation of the flare on the iFACILITY is reset to the elevation determined by
the DEM.
Load XYZ DEM Data – BUTTON
A
iNOTES
FT
User-defined DEM data can be used in place of the DEMLIB library. A text file
with comma or space delimited x,y,z data. Where x,y are coordinates in the same
projection used to specify the source location on the iFACILITY page; and, z is
the elevation in metres. The same processing is used as the DEM mapsheet data.
The user-defined DEM data should be of similar resolution (25 m) and data
quality as the DEM mapsheet data.
The iNOTES page has two purposes. First, the ERCB reviewers specifically
request a descriptive response for the first four questions to assist in their
understanding of non-routine flaring. Secondly, the page allows the user keep
track of assumptions or information regarding the assessment
D
R
Information is to be provided for the following prompts on the iNOTES page.
These prompts are the minimum information ERCB requires to gain an
understanding of the non-routine operations:
1) For permanent facilities describe the flaring event modelled including the
type of facility, pipeline segments or vessels to depressurize (if
applicable), PSV size, etc.
2) Estimate the frequency of the flaring event
a) number of events per year;
b) duration of each event;
c) total number of hours per year
3) Describe attempts to reduce or eliminate the flaring event(s) if applicable
4) State any engineering assumptions you’ve made in completing the entries
for ERCBflare
The remainder of the iNOTES page can be used to summarize references,
assumptions or other notes pertinent to the flaring scenario. The iNOTES area
can be used in Batch mode to store sensitivity and scenario notes.
3 ERCBflare Inputs
61
D
R
A
FT
For non-routine flares, the ERCB request a descriptive response
for the first four questions to assist in their understanding of the
flaring. You may provide additional information as attachments,
but the information entered in those responses should provide a
sufficient executive summary for the reviewer.
3 ERCBflare Inputs
62
4.
AIR DISPERSION MODELLING
AND OUTPUT
FT
Non-routine flaring events are typically of short duration, less than an hour, but
may extend over a day. The flow rate may be steady (constant) in time if there is
a supply of gas but may decrease in time if a vessel or pipeline is being blown
down. Regulatory air dispersion models such as AERMOD and CALPUFF
simulate steady releases with a continuous plume. CALPUFF is the only model
that can directly model short duration and transient releases.
A
An approach is needed to apply the existing air dispersion models to non-routine
flaring emissions in order to verify compliance with ambient air quality
objectives. The continuous plume models can be used with appropriate inputs and
adjustments to the output. The release is not continuous: the hour by hour
predictions can be used as this is equivalent to modelling the release occurring in
any hour of the year. The daily and annual average predictions cannot be used as
they are based on a continuous release.
R
For a steady release the inputs are based on the release rate. For example, inlet
gas or acid gas diverted to flare can be modelled as steady releases. The gas
released to the flare is being replaced by gas coming into the plant or process area
and the pressures within the piping remains relatively constant. For a steady
release of d-minutes duration, the predicted 1 hour average concentration outputs
from a continuous plume model are adjusted. If the release duration is less than
60 minutes, the hourly predictions are multiplied by d/60. If the release duration
d is greater than or equal to 60 minutes, the hourly predictions are not adjusted.
D
The oMODELLING page presents a summary of the flared gas streams flow
rates. Below the flow rates, are listed the heat released to the plume, calculated
excess air, estimated flame temperature, flared mixture net heating value and the
conversion efficiency based up on the average meteorological conditions. The
combustion efficiency and the conversion efficiency are assumed to be the same.
These are determined using:
𝜂𝜂𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 = 1 −
𝐴𝐴
𝐵𝐵 𝑈𝑈𝑎𝑎
𝑒𝑒𝑒𝑒𝑒𝑒 �
�
3
(𝑔𝑔𝑔𝑔𝑔𝑔𝑔𝑔𝑔𝑔)1/3
𝑁𝑁𝑁𝑁𝑁𝑁
Where A=133.3 [(MJ/kg)3]; B=0.317; Net Heating Value (NHV) [MJ/kg]; Ua is
ambient wind speed [m/s]; g is gravity [m/s2]; Us is source exit velocity [m/s]; Ds
is source diameter [m].
4 Dispersion Modelling and Output
63
FT
PSEUDO-SOURCE PARAMETERS
R
A
The ERCBflare calculations produce a set of pseudo-source parameters based
upon conservation of energy and conservation of momentum. Using an estimate
of the flame temperature and momentum ERCBflare determines the buoyancy
flux and the momentum flux parameters. These flux parameters are used by air
dispersion models to determine the final rise of the emissions plume. These flux
parameters can then be used to reverse engineer (back calculate) source
parameters that will result in the calculation of the flux parameters. These source
parameters are called pseudo-source parameters (see Figure 5) since they mimic a
source that will lead to the correct plume rise.
D
ERCBflare provides a summary of the pseudo-source parameters on the
oMODELLING page. The pseudo-source parameters can be used for refined air
dispersion modelling of the flare source in air dispersion models or ERCBflare.
Pseudo-source parameters are based upon the actual source parameters for the
flare scenario but do not necessarily have real physical relevance. That is, the
pseudo-source diameter is not a real diameter of the stack or flame width, but only
a calculated diameter to mimic a source so that the calculated plume rise is
correct. Caution should be used in using the pseudo-source parameters for
anything except their intended purpose.
4 Dispersion Modelling and Output
64
Ts
2.52"
Hs’
Us
Ds
Figure 5:
FT
Hs
SO2 Flare Model for Source Parameters
D
R
A
The pseudo-source parameters for the SO2 emissions for a sour gas flare are based
on the estimated flame temperature using the estimated combustion efficiency.
The combustion efficiency is a function of the net heating value of the flared gas
and of the ambient wind speed. The summary table shows the parameters for
each of the Qmax, Qave and Qlow flow rate scenarios. Note that the SO2 source
model assumes that 100% of the raw gas sulphur is converted to SO2. The
effective height of the SO2 source is estimated flame height based upon the
Brzustowski Flare Model.
Note that the SO2 source model assumes that 100% of
the raw gas sulphur is converted to SO2.
4 Dispersion Modelling and Output
65
The pseudo-source parameters for the H2S emissions for a sour gas flare are based
on flare studies (Kostiuk, Johnson and Thomas 2004) that show that flare
inefficiency in high winds is a result of fuel stripping from the combustion zone
of the flame. This is illustrated in Figure 6. The fuel stripping is un-combusted
fuel and therefore contains non-oxidized raw gas components or partially oxidized
raw gas components depending upon the kinetics of the component and the
residence time at higher temperatures. Some of the raw gases exit from the flame
tip (path A) and some raw gases exit through path B. Both paths have
approximate equal magnitude of emissions. ERCBflare assumes the gases exiting
path B remain as H2S, whereas the fraction that exists through path A will likely
have been heated sufficiently to oxidize the H2S.
FT
The source model for H2S is based upon a simplifying assumption that the
temperature of the gases (path B) is constant and is the average of the raw gas
temperature and lower flammability limit of the mixed raw gases (because the
gases are not combusted they must be less than the combustion temperature).
Therefore the energy available for plume rise of the path B gases is proportional
to the mass emission rate which is a function of wind speed. Pseudo-source
parameters are back calculated based upon the heat transfer to buoyancy and an
effective height equal to half of the flame height based upon the Brzustowski
Flare Model.
R
A
The SO2 source model energy is corrected for the small loss in energy lost by the
stripping model used in the H2S model for completeness. The inefficiency
fraction the raw gas carries accounts for a small amount of momentum which is
used to estimate the source exit velocity. The exit velocity is limited by the
ERCBflare source model minimum exit velocity to prevent number errors in
dispersion models determined through experience.
A
D
Wind flowing past the flame
strips raw gases ~ combustion
in-efficiency emissions
WIND
Figure 6:
B
Inefficiency results in A (raw gas exiting
warm) and B (raw gas exiting cool)
H2S Stripping Model for Source Parameters
4 Dispersion Modelling and Output
66
FT
The table of the H2S pseudo-source parameters can be used for refined air
dispersion modelling of the H2S source. The temperature for the H2S source
model is a result of the energy and momentum back calculation.
CALCULATION BUTTONS
D
R
A
Calculation buttons (see oMODELLING page) are used to perform air dispersion
modelling using AERMOD and to create AERMOD ready input files. The
generalized flowchart that represents the calculation process once a calculation
button is pressed is shown in Figure 7. User inputs are gathered from the three
principal ERCBflare input pages (iFACILITY, iFLARING and iTERRAIN). For
a flaring emissions scenario (Qmax, Qave or Qlow) a terrain assessment is
selected, either (parallel or elevated terrain). For the parallel scenario the terrain
is forced to be the same elevation as the flare base elevation. For the elevated
terrain scenario, the terrain entered on the iTERRAIN page is interpolated for a
receptor grid from 100m to 10km. Once the maximum terrain elevation is
reached, all points on the receptor grid remain at the maximum elevation.
AERMOD uses a slope based algorithm for determining hill height influences at
each receptor location which are calculated by ERCBflare.
4 Dispersion Modelling and Output
67
ERCBflare Spreadsheet
iFLARING
iFACILITIES
iTERRAIN
Qmax/Qave/Qlow
PARALLEL / ELEVATED
TERRAIN
Uw, Ta, PG
COMBUSTION
CALCULATIONS
FT
PROCESSING MET
RUN AERMOD
AERMOD input
Ds, Us, Ts, Hs, Es
POST.bin
A
EXTERNAL
CALCULATE ENDPOINTS
R
DISPLAY
Figure 7:
Calculation Flowchart
D
Either a screening or a user specified site specific AERMOD style meteorological
surface file is sequentially processed. For each hour of meteorology, the ambient
wind speed, temperature and atmospheric stability class (Uw, Ta and PG) are
determined and then inserted into the ERCBflare combustion calculation engine.
The output from the combustion and related calculations are a set of source
pseudo-parameters that conserve source energy and momentum. Source diameter,
velocity, temperature, height and emissions (Ds, Us, Ts, Hs and Es) are saved to an
AERMOD ready variable emissions source file.
Once all of the meteorology has been processed in this way, the AERMOD
program is executed to perform the air dispersion modelling calculations. The
hourly AERMOD output for each receptor is saved into a binary output file
(POST.BIN). This file is then processed by ERCBflare and the Risk Based
Criteria (RBC) are used to compare and against the respective statistics
4 Dispersion Modelling and Output
68
calculated. The AERMOD program, AERMOD.inp file and POST.bin file are
external to the ERCBflare, in a DOS COMMAND PROMPT window.
The process is repeated to determine the concentrations for both parallel and
elevated terrain. Both parallel and elevated terrain are assessed since this
represent the range of terrain elevations possible, thus bounding the range of
expected concentrations with distance. Parallel terrain results may be applicable
along or through a valley and the elevated terrain may be applicable when winds
blow over hills in the study area.
FT
The process is repeated to determine the concentrations for Qmax, Qave and
Qmin. Because there is a non-linear relationship between the energy released,
plume rise, combustion efficiency, downwash and air dispersion modelling
predictions, it is important to assess the range of likely flaring rates to ensure
compliance with ambient air quality objectives.
For Routine Mode flare air dispersion modelling, the process outlined in Figure 7
is simplified since only the Alberta average meteorological conditions are used.
A
There are eight calculation buttons available when the user selects the Recalculate
button and is redirected to the oMODELLING page (listed below) and discussed
in the following sections.
Flaring Assessment
Mode
Button
R
1.AERSCREEN-MAX
2.AERSCREEN-(RBC) (User Met)
3.AERMOD-(RBC Refined Modelling using
User Terrain and User Met)
4.Create AERMOD files (Modelled Wind
Speed)- (User Terrain and User Met)
Purpose
Routine
Routine
Screening Evaluation
Bridge to Refined Modelling
Routine
Refined Modelling
Routine
Create AERMOD ready files
for further assessment by the
user.
Non-Routine
6.AERSCREEN-RBC (User Met)
Non-Routine
7.AERMOD-RBC Refined Modelling using User
Terrain and User Met
Non-Routine
Refined Modelling for Permet
8.Create AERMOD files (HbH with User Terrain
and User Met)
Non-Routine
Create AERMOD ready files
for further assessment by the
user.
D
5.AERSCREEN-MAX (HbH)
Screening for Permit
Bridge to Refined Modelling
for Permit
It is recommended that you DO NOT continue to use your
computer for other concurrent Windows applications when
running the dispersion models, because this can cause
interference and instability within the calculations.
4 Dispersion Modelling and Output
69
1.AERSCREEN-MAX
FT
This calculation runs AERMOD using AERSCREEN methodology. Wind is from a
single wind direction blowing towards all receptors, always. Two assessments are
performed, one for parallel terrain (terrain heights equal to the flare base
elevation) and complex terrain (terrain heights as input on the iTERRAIN page)
allowing the AERMOD processing to determine the how terrain impacts the
ambient air quality predictions. The screening meteorological data sets are used to
account for land cover and provide a seasonal variation in meteorology.
Approximately 2500 variations in meteorology are assessed. Because the number
of variations is less than 8760, only the maximum concentration is used in the
interpretation of the air dispersion modelling results (although the seasonal
predictions are available for interpretation).
A
The AERSCREEN modelling runs include both the parallel
and complex terrain assessments. The parallel terrain
predictions are a typical worst case for wind directions not
towards terrain, and the complex terrain predictions are a
typical worst case for wind directions towards terrain. All
assessment must include terrain
R
2.AERSCREEN-(RBC) (User Met)
D
This calculation runs AERMOD in using AERSCREEN methodology. Wind is
from a single wind direction blowing towards all receptors always. Two
assessments are performed, one for parallel terrain (terrain heights equal to the
flare base elevation) and complex terrain (terrain heights as input on the
iTERRAIN page) allowing the AERMOD processing to determine the how terrain
impacts the ambient air quality predictions. A site specific 5-year meteorological
data sets created by the user is used to account for site-specific land cover and to
provide site-specific variations in meteorology. Approximately a minimum of
8760 hours of variations in meteorology should (according to ESRD Air Quality
Modelling Guidelines) be used but 5-years are recommended. The Risk Based
Criteria (RBC) can be used to determine whether the air quality dispersion
modelling results meets ambient air quality objectives. 5-years of meteorology
are required for a monthly assessment using Risk Based Criteria.
4 Dispersion Modelling and Output
70
3.AERMOD-RBC (User Terrain & User Met)
4.Create AERMOD files (Average Wind
Speed)
FT
This calculation is similar to #2AERSCREEN-RBC(User Met) except that only
the AERMOD input files are created. AERMOD is not run and no ambient air
quality predictions are created. This option is useful for advanced users to
perform refined air dispersion modelling external to the ERCBflare spreadsheet
calculations but using the ERCBflare model to create the source parameters. The
advanced user can add site specific receptor grid and terrain into the AERMOD
files created and run AERMOD independent of ERCBflare.
Post-processing of an external AERMOD output file can be performed using the
oPOSTPROCESS page.
A
5.AERSCREEN-MAX (HbH)
D
R
This calculation runs AERMOD in using AERSCREEN methodology. Wind is
from a single wind direction blowing towards all receptors always. Two
assessments are performed, one for parallel terrain (terrain heights equal to the
flare base elevation) and complex terrain (terrain heights as input on the
iTERRAIN page) allowing the AERMOD processing to determine the how terrain
impacts the ambient air quality predictions. The screening meteorological data
sets are used to account for land cover and provide a seasonal variation in
meteorology. Approximately 2500 variations in meteorology are assessed. The
source parameters for the flare are determined on an hour-by-hour (HbH) basis.
Therefore, flare efficiency and plume rise are affected by the hour by hour
variation in wind speed, temperature and stability. Because the number of
variations is less than 8760, only the maximum concentration is used in the
interpretation of the air dispersion modelling results (although the seasonal
predictions are available for interpretation).
6.AERSCREEN-RBC (User Met)
This calculation runs AERMOD in using AERSCREEN methodology. Wind is
from a single wind direction blowing towards all receptors always. Two
assessments are performed, one for parallel terrain (terrain heights equal to the
4 Dispersion Modelling and Output
71
FT
flare base elevation) and complex terrain (terrain heights as input on the
iTERRAIN page) allowing the AERMOD processing to determine the how terrain
impacts the ambient air quality predictions. A site specific 5-year meteorological
data set created by the user is used to account for site-specific land cover and to
provide site-specific variations in meteorology. Approximately a minimum of
8760 hours of variations in meteorology should (according to ESRD Air Quality
Modelling Guideline) be used but 5-years are recommended. The source
parameters for the flare are determined on an hour-by-hour (HbH) basis.
Therefore, flare efficiency and plume rise are affected by the hour by hour
variation in wind speed, temperature and stability. The Risk Based Criteria
(RBC) can be used to determine whether the ambient air quality modelling results
meets ambient objectives. 5-years of meteorology are required for a monthly
assessment using Risk Based Criteria.
7.AERMOD-RBC (HbH User Terrain & User
Met)
A
8.Create AERMOD files (HbH User Met)
R
This calculation is similar to #4AERSCREEN-RBC(HBH User Met) except that
only the AERMOD input files are created. AERMOD is not run and no ambient
air quality predictions are created. This option is useful for advanced users to
perform refined air dispersion modelling external to the ERCBflare spreadsheet
calculations but using the ERCBflare model to create the source parameters. The
advanced user can add site specific receptor grid and terrain into the AERMOD
files created and run AERMOD independent of ERCBflare.
D
Post-processing of an external AERMOD output file can be performed using the
oPOSTPROCESS page.
Example Run-Times
Example run times are listed in the table below. The test computer has the
following specifications:
• Intel i7 CPU, 950 MHz bus @ 3.07 GHz; 18 GB Ram
• Windows 8, 64bit OS, x64 processor
• 7.5 Windows Experience Index
4 Dispersion Modelling and Output
72
FT
Calculation Button
Non-Routine
Routine
1
00:00:54 (NS=12)
0:00:19 (NS=4)
2
-0:02:46 (NS=4)
3
00:01:38 (NS=12)
-4
00:26:00 (NS=12)
-5
-0:00:00
6
00:20:00 (NS=12)
-NS= Number of scenarios [4=(1×SO2 + 1×H2S)×(parallel+complex)]; [12=(3×SO2 +
3×H2S)×(parallel+complex)]
NON-ROUTINE FLARE AIR
DISPERSION MODELLING
R
A
Based upon the selecting of the flaring assessment mode on the iSTART page,
the Non-Routine Flaring assessment begins with a selection of: PERMIT or with
EVALUATION and a subsequent selection of Non-Routine Planned Flaring or
Non-Routine Unplanned Flaring.
D
When Non-Routine mode is used for ERCBflare, the air dispersion modelling
options are highlighted by the red arrows on the left-hand side. Assessments may
begin with a screening level calculations(s) using the #5. AERSCREEN-MAX
using maximum predicted concentrations produced using the AERSCREEN
screening meteorological datasets. For a permit for non-routine flaring, the
assessment must be performed using the #3. AERSCREEN-MAX(HbH) hour-byhour calculation option which is one step more refined than the #1 button.
4 Dispersion Modelling and Output
73
FT
R
A
The oMODELLING summary page displays a stamp of the model used for the
predictions and the time period used for the meteorological data.
D
ROUTINE FLARE AIR
DISPERSION MODELLING
Based upon the selecting of the flaring assessment mode on the iSTART page,
the Routine Flaring assessment begins with a selection of: Evaluation.
When Routine Mode is used for ERCBflare, the air dispersion modelling options
are highlighted by the red arrows as shown on the right-hand side. Assessments
may begin with a low-level screening calculations(s) using the #1. AERSCREEN-
4 Dispersion Modelling and Output
74
R
A
FT
MAX using maximum predicted concentrations produced using the
AERSCREEN screening meteorological datasets. For routine flaring, the
assessment can be performed using the either #1. AERSCREEN-MAX or
#2. AERSCREEN-RBC calculation options. Button #1 produces the most
conservative predictions.
AERSCREEN INPUTS
D
The source model inputs are calculated on the oCALCULATIONS combustion
modelling page and summarized on the oMODELLING page. The
oAERSCREEN page collects all of the inputs used for the AERMOD air
dispersion modelling runs and is presented in the AERMOD Source Parameters
group. The table lists the same pseudo-source parameters presented on the
oMODELLING page as well as the Brzustoski Flare Model effective source
location. Also listed are the average meteorological conditions efficiency and
emissions. At table is presented for the SO2 source and the H2S source.
For Routine flaring, the average meteorological conditions are used and the
pseudo-source parameters listed in the table are the source parameters used in the
air dispersion modelling.
4 Dispersion Modelling and Output
75
FT
Hour by Hour
R
A
The pseudo-source parameters for the Non-Routine flaring air dispersion
modelling vary hour by hour. In this case, height, diameter, temperature, velocity,
emissions and location are functions of the hourly meteorology. The AERMOD
Gaussian plume air dispersion model does not have the inherent ability vary all
these parameters hourly. The variable emissions source file allows only the
temperature, velocity and emission rate to vary hourly. It assumes that the
sources parameters such as height and diameter are physical (real) dimensions
that normally would not vary.
D
ERCBflare uses a co-located source(s) configuration to bypass this limitation of
the AERMOD model. Ideally, ERCBflare would configure a unique source for
each hour of meteorology where any given source only has an emission when that
hour of meteorology occurs.
4 Dispersion Modelling and Output
76
R
A
FT
That configuration would require enormous numerical resources. Instead, a fewer
number of sources are defined. ERCBflare estimates the final plume rise height
based upon the U.S. EPA ISCST model plume rise which uses simplistic PG and
wind speed as inputs. For the complete meteorological data set, ERCBflare
determines the distribution of final plume rise heights and divides the range in N
parts. It then summarizes the pseudo-source heights and diameters that lead those
plume rise predictions. The pseudo-source parameters for each of the N-sources
is determined based upon the average of the pseudo-source parameters for that
plume rise group. ERCBflare uses by default N=3. Sensitivity testing has proven
shown that the predictions are not strongly influenced by the selection of N>3,
and N=3 was chosen for numerical efficiency and sufficient for screening
purposes.
D
WIND
Maximum Plume Rise
Hs*3
66% Plume Rise
Hs*2
33% Plume Rise
Hs*1
Minimum Plume Rise
Hs
4 Dispersion Modelling and Output
77
Figure 8:
Illustration of the Range of Plume Rise (centrelines) for
An Hour By Hour Source Model
The oAERSCREEN page displays a summary of the N=3 virtual source plume
rise groups and average pseudo-source heights and diameters. The hour-by-hour
processing creates a variable emissions source file and AERMOD input file
containing N=3 virtual sources based upon the input meteorological data set.
FT
The hour-by-hour air dispersion modelling using ERCBflare is numerically more
intensive (slower) since it requires the spreadsheet to recalculate for each hour of
meteorology. Although the spreadsheet recalculation for a given wind speed,
temperature, and atmospheric stability is less than ∆t=0.005 sec, when a 5-year
meteorological data set is processed, this results in (5×8760 hours×∆t ≈3min) run
time. The source processing must be completed for each of the Qmax,, Qave and
Qlow emission rates and for each of SO2 and H2S sources (i.e., 6 times). For
faster run times, the stand-alone ABflare refined model can be used.
A
AIR DISPERSION MODELLING
PREDICTIONS
R
oAERSCREEN Output Summary
D
After the AERMOD air dispersion modelling is completed, ERCBflare loads the
output file and meteorological file, and post-processes the predictions. The
oAERSCREEN page is updated with the air dispersion modelling predictions.
The AERMOD predictions for assumed hourly emissions durations are listed in
the table for parallel terrain and elevate terrain. The table is updated for both SO2
and H2S source model predictions for Qmax, Qave and Qmin emissions scenarios.
4 Dispersion Modelling and Output
78
FT
A
cmax
The maximum concentration (cmax) representing the 100% (peak) concentration
assuming a 1h emission duration.
R
xmax
D
The distance (xmax) from the flare to the location of the maximum concentration
(cmax). The distance is the planar distance and not the distance as measured
along the surface. Typically, the difference is small.
ws
The wind speed (ws) that resulted in the maximum concentration (cmax).
PG
The atmospheric stability class (PG: Pasquill-Gifford stability class) that resulted
in the maximum concentration (cmax).
4 Dispersion Modelling and Output
79
xobj1
The distance (xobj1) where the predicted maximum concentrations drop below
the Objective 1 listed in the oAERSCREEN page Objectives and Limits group.
Objective 1 is ambient air quality objective. If the maximum concentration (cmax)
prediction does not exceed Objective 1, then xobj1 is set to -1.
xobj2
FT
The distance (xobj2) where the predicted maximum concentrations drop below
the Objective 2 listed in the oAERSCREEN page Objectives and Limits group.
Objective 2 is emergency evacuation criteria. If the maximum concentration
(cmax) prediction does not exceed Objective 2, then xobj2 is set to -1.
xobj3
A
The distance (xobj3) where the predicted maximum concentrations drop below
the Objective 3 listed in the oAERSCREEN page Objectives and Limits group.
Objective 3 is 1/10th of the ambient air quality objective and is representative of
the maximum extent where plume overlap with other sources is considered. Thus
xobj3 is representative of the required domain size for refined air dispersion
modelling for the flare. If the maximum concentration (cmax) prediction does not
exceed Objective 3, then xobj3 is set to -1.
R
Concentration CLIMIT
D
Concentration CLIMIT or CMAX are the concentrations predicted for the Flaring
Mode upper concentration limit (CLIMIT) as specified on the iSTART page.
When an ERCBflare screening meteorological data set is used, the results are
displayed for each season in the meteorological file. When a site specific
meteorological data set is used, the table shows results for each year of the
meteorological data. The column on the right of the table displays the total
number of hours of predictions represented in the statistic. When N<8760, the
statistic is considered to be not valid for regulatory air dispersion modelling
approvals and applications.
Concentrations assume a 1h emission duration.
Concentration RBC
Concentration RBC (Risk Based Criteria) are the concentrations predicted for the
Flaring Mode RBC criteria as specified on the iSTART page. When an
4 Dispersion Modelling and Output
80
ERCBflare screening meteorological data set is used, the results are displayed for
each season in the meteorological file. When a site specific meteorological data
set is used, the table shows results for each year of the meteorological data. The
column on the right of the table displays the total number of hours of predictions
represented in the statistic. When N<8760, the statistic is considered to be not
valid for regulatory modelling approvals and applications.
Concentrations assume a 1h emission duration.
FT
oMODELLING Output Summary
A
After the AERMOD air dispersion modelling is completed, ERCBflare loads the
output file and meteorological file, and post-processes the predictions. The
oAERSCREEN page is updated with the air dispersion modelling predictions
and the oMODELLING page presents the conclusions of the air dispersion
modelling corrected for flare emission duration and the maximum of the parallel
terrain and elevated terrain predictions. The table is updated for both SO2 and
H2S source model predictions for Qmax, Qave and Qlow emissions scenarios.
The table shows the maximum CLIMIT and maximum RBC for multi-season or
multi-year assessments.
D
R
The variables are defined and discussed in the oAERSCREEN Output Summary
section.
ONE-HOUR AVERAGES FROM
SUB-HOURLY EMISSIONS
The AERMOD predictions are based upon hourly meteorology and therefore the
predicted concentrations represent one-hour time averages. Flare durations may
be less than one-hour, so the AERMOD predictions are adjusted based upon the
4 Dispersion Modelling and Output
81
fraction of non-zero emissions during the hour. This methodology does not
account for along wind diffusion which a puff-style air dispersion model can
account for.
When flare durations are less than one-hour duration, then the hourly-average
concentration (C1h) is determined from the modelled concentration (Cmodelled)
based upon continuous emissions by:
𝐶𝐶1ℎ =
𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷
× 𝐶𝐶𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚
1ℎ
A
FT
For a transient blowdown release, modelled as a sequence of N steps of an
exponential curve, the hourly concentration is determined using the equation
above cumulatively for the first step and remain steps until 1h is reached or the
end of the duration is reached. It is assumed that a 1h hour period starts at the
beginning of a step, therefore the calculation is performed recursively for N steps
beginning with the Nth step. Because large amounts of energy are associated with
the first step (i.e., high energy results in efficiency and high plume rise so ground
level predictions are small), the maximum 1h average concentration can occur for
sequences starting with the second or higher steps.
R
ERCB D060 Permit Conditions
D
The oSUMMARY page provides a summary of the source and scenario inputs
compared to ERCB D060 permitting conditions. The ERCB D060 reference
number is provided for reference along with the D060 allowable limit. The
scenario value for each condition is listed under This Application and it is
compared to the allowable value. The Test column states whether the This
Application meets or does not meet the D060 permitting condition and the
Requirements column provides suggestions.
4 Dispersion Modelling and Output
82
FT
R
A
The oSUMMARY page also provides a similar check list compared to the
ERCB D060 Figure 4 flow chart for air dispersion modelling and approval
requirements for Flaring Approvals. The Test column shows the result of the
comparison of the This Application compared to the ERCB D060 Figure 4
allowed. The Requirements column provides suggestions on further actions
required.
D
The oSUMMARY page provides a summary table of the Approval Limits group.
This list is the essence of the flare test approval application and the values should
be carefully reviewed since the approval will limit the flare test program to the
values listed in the table.
4 Dispersion Modelling and Output
83
FT
OUTPUT SUMMARY OF SOURCE
A
The oSUMMARY page provides a summary of the flare source assessed for the
flare test application.
Valuable information is listed in the table for both the approval review and the
operator:
For example, the table lists the fuel gas total volume required based upon
the heating value requirements or fuel gas addition. The operator can use
this information to evaluate the feasibility of providing this fuel gas for the
flaring event.
R
•
For example, near the bottom of the table the minimum and maximum
diameter of the flare nozzle are recommended on the basis of sonic
velocities and flaring efficiency. In the example below, the table indicates
that the diameter of the source is not within the recommended range and
therefore is does not produce acceptable velocities. In this case, however,
the actual source is 102 mm whereas the recommended diameter limit is
100 mm. A review of the average efficiency at the top of the summary
group or on the oMODELLING page indicates that the efficiency is
adequate, therefore the actual size (i.e., nominal pipe diameter, 102 mm) is
acceptable but since it is at the high end of the recommended range, it
could possibly lead to a lazy plume (stack down wash) under certain
meteorological conditions.
D
•
4 Dispersion Modelling and Output
84
FT
A
R
D
4 Dispersion Modelling and Output
85
OUTPUT SUMMARY OF AIR
DISPERSION MODELLING
R
A
FT
The oSUMMARY page provides a high level summary of the ERCBflare air
dispersion modelling. The table is divided in to two sections: screening and
refined air dispersion modelling. The screening air dispersion modelling is based
upon the screening meteorological data sets and the maximum predicted
concentrations. The refined air dispersion modelling section is based upon site
specific meteorological predictions and RBC statistics. High concentrations are
flagged and require more refined air dispersion modelling.
D
OUTPUT FIGURE 1
The oFIGURE 1 page presents the detailed output of the oAERSCREEN
summary. oFIGURE 1 can be used to select SO2 or H2S air dispersion modelling
predictions for maximum concentration, wind speed producing the maximum
concentration, PG stability producing the maximum concentration, or mixing
height producing the maximum concentration. These figures and combinations
can be useful for designing flare management programs or illustrations in air
dispersion modelling reporting.
In Figure 9, an air dispersion modelling output example is shown. Terrain
elevation are shown in green using the right abscissa. The Qmax, Qave and Qlow
are shown for the parallel assessment using thick line styles, while thin line styles
are used for the elevated terrain assessment. The figure shows where maximum
4 Dispersion Modelling and Output
86
predictions occur relative to the terrain. Also shown on the figure are the RBC
and CLIMIT concentration objectives.
Figure 10 shows the wind speed that leads to the maximum concentration
predictions. Figure 10 corresponds to the concentration predictions in Figure 9.
Worse case meteorological conditions are a function of terrain elevation and
distance from the source.
RBC
SO2 QMax Parallel
SO2 Qmin Parallel
SO2 Qmid Elevated
Terrain Rise Above Stack Base
90
FT
1,600
80
1,400
50
800
40
600
30
400
200
0
A
Concentration (µg/m³)
60
1,000
1000
2000
3000
4000
5000
6000
7000
Terrain Rise (m)
70
1,200
0
Limit
SO2 Qmid Parallel
SO2 QMax Elevated
SO2 Qmin Elevated
20
10
8000
9000
10000
11000
0
R
Distance from Source (m)
Example oFIGURE 1 showing the Maximum
Concentration at the Receptor Location in Comparison
to the Terrain Elevation
D
Figure 9:
4 Dispersion Modelling and Output
87
RBC
SO2 QMax Parallel
SO2 Qmin Parallel
SO2 Qmid Elevated
Terrain Rise Above Stack Base
Limit
SO2 Qmid Parallel
SO2 QMax Elevated
SO2 Qmin Elevated
90
12
80
60
8
50
6
40
30
4
20
2
0
Terrain Rise (m)
70
FT
Wind Speed (m/s)
10
10
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
10000
11000
0
Distance from Source (m)
Example oFIGURE 1 showing the Wind Speed that
Causes the Maximum Concentration at the Receptor
Location
A
Figure 10:
R
OUTPUT FIGURE 2
D
The graphic provide on the oFIGURE 2 page provides a useful summary of the
hour-by-hour predicted SO2 emissions and H2S emissions based upon the
combustion efficiency and the hour-by-hour meteorology. Note that, the
ERCBflare assessment for SO2 requires the assessment based upon 100%
conversion (i.e., constant emissions). The SO2 emissions provided in oFIGURE 2
are for reference only.
Figure 11 shows a probability plot for SO2 emissions (left abscissa) and H2S
emissions (right abscissa). The actual SO2 emissions reflect the predicted
efficiency for 99% of the time as shown in Figure 12. The H2S emissions are
relatively low most of the time but are high for the 1% of the time, when the flare
is inefficient.
4 Dispersion Modelling and Output
88
SO2 Qmax
SO2 Qmid
SO2 Qlow
H2S Qmax
H2S Qmid
H2S Qlow
80
2
1.8
70
1.6
60
1.2
40
1
0.8
30
0.6
FT
20
10
0
0
0.1
Figure 11:
1
10
20
30
40 50 60
80
90
99
0.2
0
99.9
100
Example Emissions Chart from oFIGURE 2 for HourBy-Hour Assessment
Qmid
Qlow
A
Qmax
1.08
1.03
0.98
0.93
R
Combustion Efficiency
70
0.4
0.88
0.83
D
0.78
0.73
0.68
0.63
0
Figure 12:
0.1
1
10
20
30
40 50 60
70
80
90
99
99.9
Example Efficiency Chart from oFIGURE 2 for Hour-ByHour Assessment
4 Dispersion Modelling and Output
89
100
H2S Emissions (g/s)
SO2 Emissions (g/s)
1.4
50
5.
TRANSIENT BLOWDOWN
FLARING
FT
Transient blowdown air dispersion flaring modelling represents unique challenges
for source modelling and air dispersion modelling. Because of the combined nonlinearity, a single emission rate does not provide a robust weight of evidence for
the protection of human or environmental health. Also, air dispersion models are
not designed to accommodate facility modelling of random, time varying,
transient emissions sources. Therefore, simplifications are required so that a
robust and conservative estimate of ambient air quality can be attained using
existing air dispersion modelling and methods.
A
TRANSIENT BLOWDOWN INPUTS
R
Pipelines or process vessels blowing down are examples of transient releases.
The gas released to the flare is not being replaced by gas coming into the plant or
process vessels thus the pressure within the piping decreases. The release rate is
proportional to the pressure and thus it decreases. The exponential function is a
reasonable approximation of the blowdown of vessels through a constant area
orifice. The release rate in time can be described by knowing the maximum flow
rate and the total volume of gas released.
•
•
D
Q(t ) = Q MAX exp(
−t
τ
)
where
Q
τ = • total
Q MAX
The time constant (τ, tau) is the time it would take to release the total volume at
the maximum flow rate. Theoretically, as the release rate approaches zero, it takes
an infinite time to release the total volume of gas, so a limit applied to the
duration.
Transient releases are more difficult to model as a steady release rate has to be
assumed due to the model limitations. Dramatic differences in ambient air quality
modelling predictions result if the maximum rate is used compared to the average
rate over the duration.
5 Transient Blowdown Flaring
90
Exponential releases decay slowly and would take an infinite time for the flow
rate to reach zero and for all of the mass to be released. But we must stop
modelling at some time. The exponential blowdown is for sonic conditions at the
smallest area in the discharge piping. The pressure profile of the source would be
the same as the mass release rate profile as the flow rate is directly proportional to
the pressure. Below the critical pressure the exit velocity is no longer sonic and
the flow rate decreases as the square root of pressure. The critical pressure is
about twice atmospheric pressure. The pressure in the vessel cannot drop below
atmospheric pressure or the flow would reverse. Some mass will remain in the
piping.
FT
The initial mass in the system is based on the initial absolute pressure. The
fraction remaining at the atmospheric pressure can be determined and used to stop
the flow rate. For typical initial pressures of 10,000 kPa and atmospheric pressure
of 100 kPa, the fraction of the mass released is f = 1 - 100/10000 = 99%. This
assumes an isothermal blowdown and ideal gas. The exponential blowdown
equation can be easily solved to yield the release duration, as follows:
A
tduration = − τ ln (1 − f )
Transient Source
D
R
The Transient Source group on the iFLARING page is used to enter the required
inputs to describe a transient source. The continuous exponential transient source
blowdown curve is simplified into a three-step sequence of flaring scenarios that
are evaluated independently and the results are combined to simulate a sequential
event. The transient source inputs lead to parameters that create a flaring scenario
for Qmax, Qmid and Qlow which are modelled in place of the previously
discussed Qmax, Qave and Qmax/8 scenarios.
5 Transient Blowdown Flaring
91
FT
Expected Maximum Initial Pressure, PRESS0
A
The initial pressure of the vessel or pipeline (gauge pressure) is required. The
final pressure cannot go below site ambient pressure. The initial pressure may be
the maximum operating pressure of the vessel or typical operating pressure of the
vessel.
R
Expected Minimum Initial Gas Temperature, TEMP0
D
The initial temperature of the vessel is required to determine the physiochemical
properties of the gas. For pipelines, the initial temperature may be correlated to
seasonal changes in ambient temperature. ERCBflare does not provide a linkage
between meteorological temperature and source temperature, and therefore
separate analysis may be required using high, average and low initial vessel
temperatures in order to determine the net impact of the source conditions on the
predictions.
Expected Minimum Final Pressure, PRESS1
The final pressure of the blowdown may be a function of secondary systems or
the vessel may be allowed to blowdown to near atmospheric pressure. In real
systems, this may require a very long time, therefore, a non-zero gauge pressure
of approximately 1 atmosphere (101 kPa) is a realistic endpoint.
5 Transient Blowdown Flaring
92
Pipeline/Vessel Inside Diameter, VESSELDIA
The inside diameter and length of the vessel and pipeline are required to
determine the total volume of gas flared. If the vessel is relatively short compared
to the diameter, then it likely has rounded ends. Additionally, if the vessel is a
facility blowdown, then reverse engineering may be required to enter pseudodiameter and lengths so that the total volume is correct.
Pipeline/Vessel Length, VESSELLEN
FT
The inside diameter and length of the vessel and pipeline are required to
determine the total volume of gas flared. If the vessel is relatively short compared
to the diameter, then it likely has rounded ends. Additionally, if the vessel is a
facility blowdown, then reverse engineering may be required to enter pseudodiameter and lengths so that the total volume is correct.
Minimum Orifice Diameter, ORIFICE_DIA
A
The transient blowdown of a vessel may go through various piping fittings before
reaching the flaring nozzle. The high rate of gas flow through the system will
result in compressibility limits (chocked flow) and the flow through the system is
limited by the minimum diameter in the system. The minimum diameter may be
the flare nozzle or a metering orifice.
R
Discharge Coefficient, DCOEFF
D
The minimum orifice diameter will control the rate of gas through and out of the
system because of choked flow. The discharge coefficient is not readily
determined, since it may be impacted by back pressure through piping to the flare.
A limiting case for flow through an orifice (Figure 13, Mannan 2005) provides a
default value of 0.6 which allows for pipe friction from the orifice to the flare.
5 Transient Blowdown Flaring
93
FT
Figure 13:
Coefficent of Discharge for Gas Flow Through an
Orifice (Mannan 2005)
A
Select the way the blowdown curve is converted
from a continuous curve to discrete steps, MDIST
D
R
MDIST directs ERCBflare to divide the blowdown into segments of equal volume
or of equal mass. The recommended setting is to calculate segments of equal
mass. The blowdown emission curve is well represented by an exponential
decrease in emissions with time. The mass emissions are therefore exponentially
distributed in time. The ERCBflare calculated source parameters to represent the
plume rise are therefore naturally changing as a function of mass emissions. Equal
volume steps may be more convenient when trying to match results from other air
dispersion models or for simplifying the calculation of time averages. If the equal
volume steps method is used for an exponential blowdown, the source parameters
calculated by ERCBflare are based upon the step volume, and therefore at small
release times, more mass is released at lower effective plume heights (see
Figure 14.)
Raw Gas User Initial maximum flow rate, QMAX
The vessel initial and final conditions are used by ERCBflare to calculate the
maximum flow rate (QMAX, initial flowrate of the continuous exponential
blowdown curve) and the total volume (QTOTAL). An advanced user may
calculate these variables usual in-house methods. ERCBflare accommodates nondefault analysis for this input branch point by the direct entry of QMAX and
5 Transient Blowdown Flaring
94
QTOTAL into the calculation sequence. The air dispersion modelling predictions
are flagged as non-default.
Raw Gas User Total volume within vessels/pipes,
QTOTAL
See QMAX.
User selected # of puffs, NPUFFS
FT
The continuous exponential curve must be assessed as a sequence of a discrete
number of steps in emission rates. ERCBflare uses 3 steps to simulate the
continuous curve.
User selected puff duration, PUFDUR
D
R
A
When the MDIST option is Equal Duration, the continuous exponential curve is
divided in three steps over the calculated maximum duration or three steps of
PUFDUR (puff duration) as entered by the user. If the PUFDUR results in a
sequence shorter than total blowdown duration, then the final step is adjusted to
contain the remaining mass in the blowdown curve.
5 Transient Blowdown Flaring
95
FT
A
Example Continuous Exponential Blowdown Curve
shown ERCBflare modelled Discrete Stepped
Sequence using Equal Mass Steps or Equal Duration
Steps
R
Figure 14:
D
TRANSIENT BLOWDOWN AIR
DISPERSION MODELLING
The source model is calculated on the oBLOWDOWN page. The
oBLOWDOWN is a technical page that displays the results of the source
calculations, and summarizes the oAERSCREEN detailed output similar to the
oFIGURE 1.
An important calculation group on the oBLOWDOWN page is illustrated below.
It summarizes the source flowrates and durations for the equal mass and equal
duration source models. The durations are used to predict 1h maximum time
averages from the output concentrations. In this case, each of the blowdown steps
is longer than one hour, and therefore maximum 1h average concentration from
each step is a representative possible maximum concentration.
5 Transient Blowdown Flaring
96
FT
TRANSIENT BLOWDOWN
OUTPUT
D
R
A
The source model is calculated on the oBLOWDOWN page. The
oBLOWDOWN is a technical page that displays the results of the source
calculations, and summarizes the oAERSCREEN detailed output similar to the
oFIGURE 1. The output graphic on the oBLOWDOWN page shows maximum
concentrations corrected for flaring duration and are there true 1h time averages.
5 Transient Blowdown Flaring
97
6.
ADVANCED TECHNIQUES
FT
The ERCBflare screening air dispersion modelling results may indicate that the
air quality may exceed the ambient air quality objectives. The ERCBflare tool
can be used to determine the sensitivity of the source parameters (e.g., stack
height, nozzle diameter, etc…) to the ambient air quality predictions. It may be
further determined that refined air dispersion modelling is required to design and
test the flaring scenarios. ERCBflare provides several options for further analysis
including:
creation of a site specific meteorological data set for refined air quality
dispersion modelling,
•
creation of a site specific receptor grid incorporating terrain,
•
running ERCBflare in refined dispersion mode consistent with screening
inputs and post-processing consistent with screening analysis
•
a bridge to refined air dispersion modelling by producing AERMOD ready
input files and AERMOD variable source emission files that can be run
external to ERCBflare; and,
•
post-processing of external AERMOD results using advanced statistics and
Risk Based Criteria methodology.
R
A
•
D
These advanced modelling techniques are discussed in the following sections.
SITE SPECIFIC METEOROLOGY
The iUSERMET page is displayed when the Show Technical Pages option is
selected on the iSTART page. The step-by-step process is displayed on the
iUSERMET page. The page provides detailed instructions on how to create a
site-specific meteorological data file for use with refined dispersion modelling. A
site specific meteorological data set is the first step in refining the screening air
quality dispersion assessment
6 Advanced Techniques
98
Step 1
The first step in creating a site specific meteorological data set is to acquire a
surface and upper air data file. ERCBflare uses the AQMG recommended
meteorological data set and meteorology extraction utility MMEU available from
the ESRD web-site:
http://www.albertamm5data.com/
FT
The MMEU program creates a SAMSON.dat and RAOBS.dat (surface and upper
air raw data meteorological data files). These files are processed using the
AERMET program to create AERMOD ready meteorological surface and upper
air files.
D
R
A
The procedure for STEP 1 is provided on the iUSERMET page as shown below.
The MMEU requires the location of the flare source in UTM coordinates only.
Although ERCBflare can use other coordinates for its assessment, the REFRESH
button can be used to convert the flare input coordinates to UTM coordinates
acceptable by the MMEU program.
The extraction process can take several hours depending upon the
computer and network speed. It is therefore best to plan ahead to
ensure that this Step 1 is performed well in advance of the analysis.
The MMEU extraction process is time consuming due to the large number of files
and large size of files required to be processed. Therefore, depending on the
computing resources available, this step can take a significant amount of time.
The ERCBflare site specific meteorological processing allows for
users to make use of SAMSON.dat and RAOBS.dat files created
using alternative methods. Therefore, STEP 1 can be by-passed
if a user these two files from alternative sources
6 Advanced Techniques
99
Step 2
FT
Step two in the process of creating a site specific meteorological data file is
determine the land-use surrounding the flare location. This step uses the flare
coordinates entered in STEP 1 to download land-use classification codes (LCC)
for Canada and process the data into an AERMET ready format. The steps in this
process are simplified to, press the ‘Get LCC Data’ button. However, behind the
button are complex calculations.
When the button is pressed, ERCBflare determines the NTS map sheets required
for the LCC assessment. The map sheets are output to the Map Sheets field on
the page. ERCBflare then checks the LCCLIB local library if the LCC map sheet
data has already been down loaded (see also the iSTART page mlccget flag). If
the data does not exist locally, then it is automatically downloaded from the
internet.
D
R
A
Once download, the LCC data map sheets are processed according to the
AERSURFACE rules and algorithms. Surface roughness is determined based
upon the LCC within 1 km of the flare location and Bowen ratio and albedo are
determined based upon LCC data within 5 km of the flare location. The
assessment uses a resolution of 100 m to subsample the LCC codes with in the
area.
The ‘Get LCC Data’ uses the user entries on the iUSERMET
page for the location of the flare. A user can enter any valid UTM
coordinates on the iUSERMET page to analyze LCC codes for
that site entered. This allows for what-if analysis or further
evaluation for sensitivity.
The results of the LCC assessment are displayed in the table on the iUSERMET
page. The table shows a listing of the LCC codes, the count of LCC codes within
5 km radius and the fractional percentage of the LCC for the study area. Also
listed below the detailed LCC statistics, is a re-assessment of the screening LCC
codes. This latter table can be used to compare to the land use file selected by the
screening assessment. The screening assessment uses a coarse nearest-neighbour
approach for the rapid assessment, and therefore there is expected to be
differences compared to a site specific assessment provided on this page.
6 Advanced Techniques
100
Step 3
A
FT
The third step in creating a site specific meteorological data file to combine the
surface and upper air data with the local land use statistics. The user is prompted
for the SAMSON.dat and RAOBS.dat file created in STEP 1 with the addition of
the time zone entry. The ‘Create UserMet’ button is then used to create
AERMOD ready surface and upper air files (AERMOD.sfc and AERMOD.pfl)
which are created in the same folder as the SAMSON.dat file. The AERMET.exe
program (see iBIN page) is used to create the output files.
R
The result of this step is a 5-year meteorological data set that is suitable for
refined dispersion modelling.
D
The folder containing the SAMSON.dat will contain any error
messages produced while running AERMET.exe.
Step 4 (Optional)
An optional step is to create a site specific screening meteorological file. This file
can be used similar to the ERCBflare provided screening meteorological data sets
for rapid user what-if analysis or further evaluations. The user is prompted for an
output folder. The ‘Create ScreenMet’ button is pressed to create the screening
meteorological data files using the MAKEMET.exe program (see iBIN page).
6 Advanced Techniques
101
FT
SITE SPECIFIC TERRAIN
Step 1
A
The iUSERTER page is displayed when the Show Technical Pages option is
selected on the iSTART page. The step-by-step processed displayed on the
iUSERTER page provides detailed instructions on how to create a site-specific
receptor file for use with refined dispersion modelling.
D
R
The procedure for STEP 1 is provided on the iUSERTER page as shown below.
The process requires the location of the flare source in UTM or 10TM coordinates
only. The REFRESH button can be used to convert the flare input coordinates to
UTM coordinates.
6 Advanced Techniques
102
The REFRESH button can be by-passed and the user can enter
valid UTM coordinates in the table. The ‘Create Receptor’ button
in the STEP 2 then processes the entered location.
Step 2
A
FT
The receptor grid spacing table is provided to create a default ESRD AQMG
receptor grid. The default values can be changed to create user-specific grids.
D
R
In Step 2, the name for the output receptor file is prompted for then the ‘Create
Receptors’ button is pressed to create the receptors. When the button is pressed,
ERCBflare determines the map sheets required for digital terrain information
(digital elevation model, DEM). The map sheets used in the assessment are listed
in the Map Sheets field. The assessment requires map sheets for the domain area
(10 km square radius) plus an additional 5 km radius to determine hill scale
factors for each receptor location. The hill scale factors are determined following
the algorithm used in AERMAP which determines the worst case slope positive
slope (greater than 10 degrees) from any hill in the study area at each receptor
location.
ERCBflare first looks in the DEMLIB local library (see iBIN page) according to
the mgetdem flag (see iSTART page). If the map sheet is not found within the
library, it is automatically downloaded from the internet and stored within the
DEMLIB location. The receptor grid defined in the table in STEP 1 is analyzed
and is output for reference at the bottom of the iUSERTER page. It is also stored
in the file location specified on iUSERTER. The file created is an AERMOD
ready receptor insert file that can be used to insert receptor locations into an
AERMOD ready input file using automation.
6 Advanced Techniques
103
FT
REFINED MODELLING
A
Refined dispersion modelling is available through ERCBflare using the
oMODELLING calculation buttons or using the create AERMOD ready input
files options.
R
Non-Routine Flare Air Dispersion Modelling
On the oMODELLING page:
Click on button 7. AERMOD (HbH User Terr/User Met) or 8.Create
AERMOD Input File(s) (HbH User Ter/User Met)
D
1.
2.
You will be prompted for a file name and folder in which to place the files.
Locate the folder where you want the files to be created and enter a name:
AERMOD.inp. Click OK to continue
3.
You will then be prompted for the filename of the site specific
meteorological file (see iUSERMET page) to be used to create the hour by
hour emissions source file. Locate the file and select it. Click OK to
continue.
4.
You will then be prompted for a receptor insert file to be used (see
iUSERTER page). If the user presses cancel at this step, then a screening
receptor configuration is used.
6 Advanced Techniques
104
If the user presses cancel at this step, then a screening receptor
configuration is used.
5.
ERCBflare will then look through the sources and meteorology to create
AERMOD.inp for each of the scenarios.
6.
If the #7 was pressed the analysis continues by running AERMOD and post
processing the results. If #8 was pressed, the AERMOD ready files are
created by not processed.
FT
Once complete, the folder will contain two files for each scenario to be
assessed, where: n is the case number corresponding to the number on
oAERSCREEN; and <AERMOD> is the filename entered on step 1 of this
sequence of instructions.
<AERMOD>_n.inp
A
<AERMOD>_n_emissions.dat
an AERMOD ready input file
an AERMOD ready variable source file
linked to AERMOD_n.inp
Routine Flare Air Dispersion Modelling
On the oMODELLING page:
Click on button 3.AERMOD-RBC (User Ter/User Met) or 4.Create
AERMOD files (Average Wind Speed)
2.
You will be prompted for a file name and folder in which to place the files.
Locate the folder where you want the files to be created and enter a name:
AERMOD.inp. Click OK to continue
D
R
1.
3.
You will then be prompted for the filename of the site specific
meteorological file (see iUSERMET page) to be used to create the hour by
hour emissions source file. Locate the file and select it. Click OK to
continue.
4.
You will then be prompted for a receptor insert file to be used (see
iUSERTER page). If the user presses cancel at this step, then a screening
receptor configuration is used.
If the user presses cancel at this step, then a screening receptor
configuration is used.
6 Advanced Techniques
105
5.
ERCBflare will then look through the sources to create AERMOD.inp for
each of the scenarios.
6.
If the #3 was pressed the analysis continues by running AERMOD and post
processing the results. If #4 was pressed, the AERMOD ready files are
created by not processed.
Once complete, the folder will contain two files for each scenario to be
assessed, where: n is the case number corresponding to the number on
oAERSCREEN; and <AERMOD> is the filename entered on step 1 of this
sequence of instructions.
an AERMOD ready input file
FT
<AERMOD>_n.inp
POST-PROCESSING EXTERNAL
REFINED MODELLING OUTPUT
A
A step-by-step process for the assessment of externally generated AERMOD
results is provided on the oPOSTPROCESS page.
The oPOSTPROCESS page requires that the following AERMOD output
control options where used:
R
OU PLOTFILE 1 ALL FIRST myplot.dat
OU POSTFILE 1 ALL UNFORM mypost.bin
D
The myplot.dat file is required to provide a listing of the receptor data points used
included in the assessment. The mypost.bin file contains the hour by hour
predictions of AERMOD for each receptor.
6 Advanced Techniques
106
7.
BATCH OPERATIONS
FT
The ERCBflare spreadsheet tool provides a rapid screening tool assessment for
many types of flaring assessments with a user-interface principally intended for
individual scenario assessment. Also included with ERCBflare is ability to
consecutively, process a list of scenarios for sensitivity analysis, what-if analysis
or exist as a simple database. A sensitivity analysis is an assessment where the
majority of the input variables are the same, but selected variables are modified
using a range of input values (example, stack height) in order to determine a
robust solution. For some operators with many flares, once all the base data has
been collected and entered, the ERCBflare tool can be run for the entire list of
flares and the process repeated if a common element (say gas composition or
licenced SO2 mole fraction) changes.
A
Operating ERCBflare on one or many scenarios makes use of the batch mode
capabilities. In general, input data is stored on the iBATCH page and following
an analysis, the output is saved on the oBATCH page.
R
Batch File Step-by-Step
The best way to start a batch mode configuration is to enter all of the input fields
in ERCBflare on the iFACILITY, iFLARING, iTERRAIN, and iNOTES pages.
Press “Save to Batch Page” to save all of the input fields on all of the
ERCBflare pages to the next empty line on the iBATCH page
D
•
•
Copy the scenario line on the iBATCH page to other blank lines to create
the exact same input variables. Then change selected variables on the
copied lines to create the sensitivity assessment
•
Add a run flag in the first column of each iBATCH row where 1=run and
0=don’t run. All lines with a “1” in the run column will be loaded and
processed during a batch mode operation. All lines with a “0” in the run
column will be skipped. Therefore, the run flag allows a user to keep a
large list of flaring assessment configurations on iBATCH page, but only
process those configurations marked for assessment.
7 ERCBflare Batch Operations
107
•
Press the “Run Batch” on the iBATCH page to run all of the scenarios on
the with run flags set to 1. WARNING long run times may result if many
lines are marked with a 1 in the run column.
•
Summary output information is copied back to the iBATCH for each row
entry and to the same row number on the oBATCH page.
In order for ERCBflare to accommodate early versions of MS Excel that are
limited to 255 columns of data, the batch page is divided into inputs (iBATCH)
and outputs oBATCH. Output on the oBATCH page corresponds to the
identical row number on the iBATCH page.
FT
ERCBflare does not perform house-keeping on either the iBATCH nor oBATCH
page. If rows are inserted or deleted on the iBATCH page, the user should also
do so on the oBATCH page to avoid confusion.
iBATCH PAGE AS A DATABASE
A
The iBATCH page can be used as a database (a collection of ERCBflare inputs)
in a single location. This allow a single ERCBflare spreadsheet file to be used for
many flaring scenarios configurations rather than having many ERCBflare
spreadsheets with only a single flare configuration in each file.
R
Once an iBATCH page has been populated with entries, a user can scan the page
and readily visualize differences and similarities between scenarios.
The best way to start a batch mode database is to enter all of the input fields in
ERCBflare on the iFACILITY, iFLARING, iTERRAIN, and iNOTES pages.
Press “Save to Batch Page” to save all of the input fields on all of the
ERCBflare pages to the next empty line on the iBATCH page
•
To start an new configuration, the data on the iBATCH page can be
copied and edited or the entries on the iFACILITY, iFLARING,
iTERRAIN, and iNOTES pages can be updated. In the latter case, press
“Save to Batch Page” button once a scenario update has been completed.
The scenario will be copied to the next empty line on the iBATCH page.
•
Add a run flag in the first column of each iBATCH row where 1=run and
0=don’t run. It is a safe practice, to enter “0” on all lines being stored on
the iBATCH. This prevents accidentally processing the entire database
when the “Run Batch” button is pressed.
D
•
7 ERCBflare Batch Operations
108
•
To reload a single scenario, select any cell on the row of the scenario to be
reloaded, then press the “Load Current Row” button. All of the inputs
from the iBATCH page will be copied to the respective cells on the
iFACILITY, iFLARING, iTERRAIN, and iNOTES pages.
EXAMPLE iBATCH SENSITIVITY
SETUP
FT
The steps below will be a guide through a simple batch mode sensitivity
configuration. In this tutorial example, we use the pre-loaded “Site A”
configuration.
A
1. On the iBATCH page, select a cell on the row with the scenario name
“Site A”, then press the “Load Current Row” button at the top of the
iBATCH page. This will copy all of the inputs and settings from the
iBATCH page to the respective cells on the iFACILITY, iFLARING,
iTERRAIN, and iNOTES pages.
2. On the oMODELLING page, press the non-routine modelling option
“5. AERSCREEN-HBH” to run and AERMOD analysis using the
screening meteorological data using the hour-by-hour flaring option.
D
R
3. Once the modelling has been completed, the oMODELLING page should
display the results, as shown below. The results indicate that maximum
concentration is 1311 µg/m³. Since this concentration is greater than the
SO2 objective of 450 µg/m³, it is desirable to determine the stack height
required in order to meet the objective. This stack would represent an
upper bound on the necessary stack height and only one possible design
change option
7 ERCBflare Batch Operations
109
Dispersion Model Used
3
Aermod Version Used
AERSCREEN-(HBH) Max Concentration Only
12345
Meteorological File UsedBflare_v130114\metfiles\GRASS.sfc
AERMOD Output File Base
# Hours in Meteorological File
2635
Dispersion Model Predictions
Annual
These results account for the duration of
the flaring and 1h hour averaging time.
SO2 Predictions
Units
Averaging time is (minutes): 60 and Total Flare duration is (minutes): 4320.0
Maximum Concentration
µg/m³
Distance to Maximum Concentration
m
Wind Speed
m/s
PG Stability Class
PG
Distance Concentration drops below Obj#1
m
Distance Concentration drops below Obj#2
m
Distance Concentration drops below Obj#3
m
µg/m³
RBC Maximum Concentration RBC (99.0th)
µg/m³
Average 1h
(Volum e/Duration)
Minim um 1h
(QMIN)
Worst Case of Parallel or Elevated Terrain Results
1311
816
6280
6893
1.0
1.0
F
F
10000
10000
-1
-1
10000
10000
NA
NA
NA
NA
336
5722
1.0
F
-1
-1
10000
NA
NA
FT
Climit Maximum Concentration CMAX (99.9th)
Maxim um 1h
(QMAX)
4. Make 5-copies of the Site A scenario on the iBATCH page. This can be
accomplished using two methods:
a. On the iBATCH page, select the entire row with the scenario name
“Site A”, press “ctrl+C” to copy the data. Scroll down to an empty
row, and select the cell in column A. Press “ctrl+V” to paste data.
Repeat this step four more times
A
b. On the iFLARING page, press “Save to Batch Page” five times.
R
5. On the iBATCH page, scroll to the end row where the data was saved.
Then scroll right to column BD “Flare Stack Tip Exit Height”. Change the
entries from the provided height of 27 m to 30 m, 40 m, 50 m, 55 m and
60 m, respectively for each of the new lines.
D
6. Add a run flag in the first column of each iBATCH row where 1=run and
0=don’t run. Therefore, enter ‘0” for all other lines on the iBATCH page
and enter “1” for the five lines that were just added. At the top left of the
iBATCH page you should see “5-Number of Scenarios Set to Run”
7. Double Check:
a. Column B: the selection of Temporary Approval vs Routine flaring
option
b. Column C: the selection of Continuous/Planned/Unplanned flaring
option
c. Column D: the selection of the calculation model to be used
d. Column E: the path to the meteorological data file has been entered
correctly.
7 ERCBflare Batch Operations
110
8. Press the “Run Batch” button to run ERCBflare for each of the 5
sensitivity runs just created. ERCBflare will consequetively: load the
inputs from the iBATCH page; run ERCBflare using the inputs and the
modelling options specified in columns B,C,D, and E; then save the results
to the oBATCH page on the same row number as the inputs.
D
R
A
FT
9. In this case, we are interested in the maximum predicted concentration
values listed in Column BV. The results for a 55 m stack are 486 µg/m³
and the results for a 60 m stack are 431 µg/m³.
7 ERCBflare Batch Operations
111
8.
References
Alberta Environment and Sustainable Resource Development (ESRD). MM5
(2002-2006) Meteorological data for dispersion models.
http://environment.alberta.ca/01119.html
FT
Alberta Environment and Sustainable Resource Development (ESRD). 2006b.
Multi-Model Extraction Utility (MMEU) & 2002–2006 Alberta
Meteorological Data Set. Climate Change, Air and Land Policy Branch,
Alberta Environment.
http://environment.alberta.ca/01120.html
Alberta Environment and Sustainable Resource Development (ESRD). 2009a.
Air Quality Modelling Guidelines (AQMG). Revised 2013.
http://environment.gov.ab.ca/info/library/8725.pdf
Burcat,A., Ruscic,B., 2005, Third Millennium Ideal Gas and Condensed Phase
Thermochecmical Database for Combustion with Updates from Active
Thermochemical Tables.
http://www.osti.gov.bridge
A
Canadian Association of Petroleum Producers (CAPP). 2011. Framework: Sour
Non-Routine Flaring, draft 29 July 2011.
R
Energy Resources Conservation Board (ERCB). 2010. Dispersion Modelling Exit
Parameters for Flares and Incinerators. Developed by Michael Zelensky.
March 1, 2010.
Energy Resources Conservation Board (ERCB). 2011. Directive 60: Upstream
Petroleum Industry Flaring, Incinerating and Venting, Revised
November 3, 2011.
http://www.ercb.ca/docs/documents/directives/Directive060.pdf
D
Energy Resources Conservation Board (ERCB). 2012. Dispersion Modelling Exit
Parameters for Non-Routine Flares. Developed by Michael Zelensky.
March 1, 2011.
ERCBflare. 2010. Directive 060 Spreadsheets - March 22, 2010
http://www.ercb.ca/regulations-and-directives/directives/directive060
Gas Producers Suppliers Association (GPSA), 2012, Engineering Data Book, 13th
Edition, SI, Vol 1 & 2
Geobase. 2000. Canadian Digital Elevation Data. Government of Canada, Natural
Resources Canada, Earth Sciences Sector, Centre for Topographic
Information.
http://www.geobase.ca/geobase/en/find.do?produit=cded
Geobase. 2009. Land Cover, Circa 2000- Vector. Government
of Canada, Natural
Resources Canada, Earth Sciences Sector, Centre for Topographic
8 References
112
Information – Sherbrooke
http://geobase.ca/geobase/en/find.do?produit=csc2000v
Gordon,S., McBride, B.J., 1971, Computer Program for Calculation of Complex
Chemical Equilibrium Composition, Rocket Performance, Incident and
Reflected Shocks and Chapman-Jouguet Detonations, NASA SP-273.
Hubbard,R.. 2009. The Role of Gas Processing in the Natural-Gas Value Chain. Society
of Petroleum Engineers, Journal of Petroleum Technology, August 2009.
Irwin, J.S. 1979. A Theoretical Variation of the Wind Profile Power-Law
Exponent as a Function of Surface roughness and Stability, Atmospheric
Environment 13, pp. 191-194.
FT
Kostiuk,L., Johnson,M., Thonas,G. 2004. University of Alberta Flare Resarch
Project, Final Report, November 1996-September 2004.
http://www.mece.ualberta.ca/groups/combustion/flare/papers/Final%20Report2004.pdf
Mannan,S. 2005. Lees’ Loss Prevention in the Process Industries, 3rd Ed. Elsevier Inc.,
Burlington, MA
Scire, J.S., D.G. Strimaitis and R.J. Yamartino. 2000. A User’s Guide for the
CALPUFF Model (Version 5.0). Concord, MA: Earth Technologies Inc.
TransCanada Corporation. 2012. Gas Quality Specification, TransCanada and
other pipelines.
A
http://www.transcanada.com/customerexpress/docs/assets/Gas_Quality_Specifications_F
act_Sheet.pdf
United States Environmental Protection Agency (U.S. EPA). 2011. User’s Guide
for the AMS/EPA Regulatory Model AERMOD. EPA-454/B-03-001
(September 2004). March 2011.
D
R
Zabetakis,M.G.. 1965. Flammability Characteristics of Combustible Gases and
Vapors. U.S. Dept. of the Interior, Bureau of Mines, Bulletin 627.
8 References
113
Appendix A
wsstkht
fneq90
fneq91
NR_Texhaust
NR_XL
p_sat
R
t_sat
Description
Calculates a real 1 hour time average from a sequence of modelled sub-hourly
duration events modelled as continuous sources producing modelled 1 hour time
averages
Calculates the wind speed at stack height given the wind speed at reference
height.
Uses Gordon (et al 1971) function 90 to calculate the molar heat capacity
Uses Gordon (et al 1971) function 91 to calculate the enthalpy
Uses the Newton-Raphson method to find the flare flame temperature
Uses the Newton-Raphson method to find the Bruzowski flare model XL
parameter
Calculates the saturation pressure for steam using the Gibbs free energy
formulation
Calculates the saturation temperature for steam using the Gibbs free energy
formulation
Uses Redlich-Kwong formulation to determine the compressibility factor for
gases
A
Function
calmaxcr
FT
The ERCBflare was designed to use a platform that provided transparency for all
calculations. This transparency allows the user to follow the calculations and as
necessary repeat the calculations using external means. There are instances where
the benefits of programming where used to perform calculations that would not be
possible in a simple spreadsheet or make the spreadsheet unnecessarily complex.
In these instances, a programmed function was created to perform the
calculations. Otherwise, ERCBflare uses combinations of cell and worksheet
references for all calculations.
D
FNz
FUNCTION calcmaxcr(…)
This function calculates the maximum 1 hour average concentration from a list of
sequential 1 hour average modelled concentration and implied durations for the
results. Typical dispersion models operate on 1 hour time average basis, whereas
the desired modelling duration may be sub-hourly. Sub-hourly duration
emissions are modelled by ignoring the real duration, and adopting the model’s
1 hour average resolution. The result of the modelling (Chour) are subsequently
corrected by prorating the modelled 1 hour time average by desired real duration
(D2). C2 is the corrected 1 hour time average.
8 References
114
C2 =
min(60, D2 )
Chour
60
(A.1)
Spreadsheet function call:
=CalcMaxCr(rcdat, rtdat, tlim)
Where rcdat is a range variable (eg, A1:C1) representing the modelled output
concentrations assuming 1 hour time average, rtdat is a range variable (eg A2:C2)
representing the real duration, tlim is total real duration. If the units of rtdat are
minutes, then tlim should be entered in minutes tlim=60.
FT
This function can be used for a single prediction:
=CalcMaxCr(A1, B1, 60)
Where A1 is the modelled 1 hour average concentration, B1 is the real duration in
minutes.
TRANSIENT BLOWDOWN MODELLED AS PUFFS
Equal Duration Puffs
Equal Mass Puffs
A
Instantaneous Rate
250
Transient Blowdown of 148.0 e³m³ with an Initial Flow Rate of 655 e³m³/d
Exponential Time Constant, TAU= 36.0 min
3 puffs of equal mass
150
100
D
3 puffs of 428 min duration
180
97.9% of mass released in 1284.3 min
60
0
0
50
120
Flaring Rate (e3m3/d)
R
200
Time (minutes)
This function can also be used with inputs of consecutive predictions. For
instance, for the blowdown sequence above, the green line represents the real
durations (14 min, 24 min, and 100 min) for three modelling runs at rates
(165 m3/s, 99 m3/s and 24 m3/s, respectively). Each rate is modelled
independently assuming a continuous emission duration. The modelled output is
8 References
115
provided as 1 hour average concentration for each emission rate, (eg, 100 µg/m³;
75 µg/m³; 25 µg/m³, respectively). Using the CalcMaxCR(…) function the real
1 hour average (worst case) concentration is calculated to be 62.5 µg/m³:
Range(A1:C1) = (100; 75; 25)
Range(A2:C2) = (14; 24; 100)
62.5=CalcMaxCR(A1:C1,A2:C2,60)
FT
CalcMaxCR(...) determines the worst 1 hour average concentration. In the
example above, a 1 hour period is assumed to start at the beginning of the threestep sequence. Therefore, a 1 hour period is composed of 14 min at 100 µg/m³,
24 min at 75 µg/m³ and 22 min (ie the fraction of the 1 hour period remaining) at
25 µg/m³. Alternatively, a 1 hour period could be composed of 24 min at
75 µg/m³ and 36 min at 25 µg/m³, or 60 min at 25 µg/m³.
A
In the above example, it doesn’t appear necessary to examine all of the sequential
combinations since the maximum concentration of 100 µg/m³ is at the start of the
sequence. However, because of non-linearity included in the modelling that takes
into account the hour-by-hour emission rates, downwash, and terrain effects, the
maximum sequence element concentration could be at any step in the sequence.
For instance:
D
R
Range(A1:C1) = (50; 100; 70)
Range(A2:C2) = (14; 24; 100)
82=CalcMaxCR(A1:C1,A2:C2,60)
8 References
116
D
R
A
FT
' ------------------------------------------------------------' CALMAXCR() function
'
' B.W.Zelt, Ph.D., P.Eng.
'
' 27-Sep-2012
'
' User input of ranges
' then call the CalcMaxC() function
Function CalcMaxCr(rcdat As Range, rtdat As Range, tlim)
Dim cdat() As Double
Dim tdat() As Double
isize = 0
For Each a In rcdat
isize = isize + 1
Next
ReDim cdat(isize)
ReDim tdat(isize)
isize = 0
For Each a In rcdat
isize = isize + 1
cdat(isize) = a.value
Next
isize2 = 0
For Each a In rtdat
isize2 = isize2 + 1
If (isize2 <= isize) Then
tdat(isize2) = a.value
End If
Next
If (isize <> isize2) Then
CalcMaxCr = "Ranges are different size"
Else
CalcMaxCr = CalcMaxC(cdat(), tdat(), isize, tlim)
End If
End Function
8 References
117
R
A
FT
' ------------------------------------------------------------' CALMAXC() function
'
' B.W.Zelt, Ph.D., P.Eng.
'
' Calculates the maximum period (TLIM) average concentration from
' a time series of C() and durations T().
'
' 27-Sep-2012
'
' UPDATES:
'
'
Function CalcMaxC(cdat() As Double, tdat() As Double, ndat, tlim)
cmaxi = 0
For i = 1 To ndat
T1 = 0
cmaxj = 0
For j = i To ndat
If (T1 + tdat(j) > tlim) Then
If (tlim - T1 > 0) Then
cmaxj = cmaxj + cdat(j) * (tlim - T1) / tlim
End If
GoTo done1
Else
cmaxj = cmaxj + cdat(j) * tdat(j) / tlim
End If
T1 = T1 + tdat(j)
Next j
done1:
If (cmaxj > cmaxi) Then
cmaxi = cmaxj
End If
Next i
CalcMaxC = cmaxi
End Function
FUNCTION wsstkht(…)
D
This function calculations the windspeed at stack height given the input variables
reference windspeed (m/s), reference windspeed height (m), Pasquill-Gifford
stability class, and stack height (m).
Spreadsheet function call:
=wsstkht(wref, Zref, iPG, hs)
This function performs the following calculation:
 max(hs ,10 m) 
ws = wsref 


href


plx 0
(A.2)
Where the exponent plx0 is determined by the following lookup data as displayed
on the PROPERTIES page.
8 References
118
PasquillGifford
Stability
Plx0
0.07
0.07
0.10
0.15
0.35
0.55
A
B
C
D
E
FT
F
FUNCTION fneq90(…)
This function performs the basic calculation of the molar specific heat capacity:
c po (T)
(A.3)
A
R
=a1 + a2T + a3T 2 + a4T 3 + a5T 4 + a1*
R
Where the lead constants a1, a2, a3, a4, and a5 are chemical specific and listed on
the PROPERTIES page (see section label Table D1). The lead constants are
listed at a reference temperature of 25 °C and are available in a low temperature
range and a high temperature range (b1, b2, b3, b4 and b5, respectively) for
chemical mixtures. There is a discontinuity between the two temperature ranges.
To avoid this behaviour, a correction factor,a1*, is used to make the curves
continuous.
D
Spreadsheet function call:
=Fneq90(T, RANGE(a_low), RANGE(b_high),
use_correction(TRUE or FALSE))
Where cpo (kJ/(kmol·K)) is the molar heat capacity, R (kPa·m3/(kmol·K)) is the
universal gas constant, T is the temperature for molar heat capacity (Kelvin),
RANGE(a_low) is a range variable pointing to the chemical specific molar heat
capacity a series low temperature parameters (see PROPERTIES page),
RANGE(b_low) is a range variable pointing to the chemical specific molar heat
capacity b series high temperature parameters (see PROPERTIES page), and
use_correction is an optional parameter to use the correction or not use the
correction (default is TRUE). The RANGE must include the a6 parameter and a1*
(in the seventh position). The a6 parameter is not used for molar heat capacity but
is used in fneq91() for the enthalpy calculation.
8 References
119
References:
Gordon,S., McBride, B.J., 1971, Computer Program for Calculation of Complex
Chemical Equilibrium Composition, Rocket Performance, Incident and
Reflected Shocks and Chapman-Jouguet Detonations, NASA SP-273.
FUNCTION fneq91(…)
FT
This function performs the basic calculation of the enthalpy:
a
ho (T )
a
a
a
a
=
a1 + 2 T + 3 T 2 + 4 T 3 + 5 T 4 + 6 + a1*
RT
T
2
3
4
5
(A.4)
A
Where the lead constants a1, a2, a3, a4, and a5 are chemical specific and listed on
the PROPERTIES page (see section label Table D1). The lead constants are
listed at a reference temperature of 25 °C and are available in a low temperature
range and a high temperature range (b1, b2, b3, b4 and b5, respectively) for
chemical mixtures. There is a discontinuity between the two temperature ranges.
To avoid this behaviour, a correction factor,a1*, is used to make the curves
continuous.
Spreadsheet function call:
R
=Fneq91(T, RANGE(a_low), RANGE(b_high),
use_correction(TRUE or FALSE),
use_rangeadjust(TRUE or FALSE))
D
Where ho (kJ/kmol) is the enthalpy, R (kPa·m3/(kmol·K)), T is the temperature
(Kelvin), RANGE(a_low) is a range variable pointing to the chemical specific a
series low temperature parameters (see PROPERTIES page), RANGE(b_low) is
a range variable pointing to the chemical specific b series high temperature
parameters (see PROPERTIES page), and use_correction is an optional
parameter to use the correction or not use the correction (default is TRUE). The
RANGE must include the a6 parameter and a1* (in the seventh position). The
use_rangeadjust parameter is an optional parameter (default is TRUE) to include
or not include the a6 parmameter. The use_rangeadjust is used for intermediate
calculations to determine the use_correction range smoothing parameter.
References:
Gordon,S., McBride, B.J., 1971, Computer Program for Calculation of Complex
Chemical Equilibrium Composition, Rocket Performance, Incident and
Reflected Shocks and Chapman-Jouguet Detonations, NASA SP-273.
8 References
120
FUNCTION NR_Texhaust(…)
This function performs the basic calculation of the exhaust temperature of gases
from the flame using Newton-Raphson method. The exhaust temperature is a
function of molar heat capacity and enthalpy of the gases. These parameters are a
function of temperature. Therefore an iterative solution is required to solve for
temperature.
Spreadsheet function call:
FT
=NR_Texhaust(Tguess, Enthalpy_Goal, RANGE(a_low),
RANGE(b_high))
The basic code for the function is:
A
Public Function NR_Texhaust(guess As Double, goal As Double, alo As Range, ahi As Range)
Dim a As Double
Dim H1 As Double
Dim cp As Double
Dim diff As Double
Dim slope As Double
Dim intercept As Double
Dim Ru As Double
Dim Tref As Double
Dim H0 As Double
Dim err As Double
Dim maxloops As Integer
Dim i As Integer
R
maxloops = 100
err = 0.0001
Ru = ThisWorkbook.Worksheets("oCALCULATIONS").Range("k_RU").value
Tref = 273.15 + ThisWorkbook.Worksheets("oCALCULATIONS").Range("Tref").value
H0 = fneq91(Tref, alo, ahi) * Ru * Tref
cp = fneq90(guess, alo, ahi) * Ru
H1 = fneq91(guess, alo, ahi) * Ru * guess - H0
intercept = H1 - cp * guess
diff = H1 - goal
D
isdone = -1#
i=0
While (isdone < 0)
guess = (goal - intercept) / cp
cp = fneq90(guess, alo, ahi) * Ru
H1 = fneq91(guess, alo, ahi) * Ru * guess - H0
intercept = H1 - cp * guess
diff = H1 - goal
If (Abs(diff) < err) Then isdone = 1
i=i+1
If i >= maxloops Then isdone = 1
Wend
guess = (goal - intercept) / cp
NR_Texhaust = guess
End Function
8 References
121
Function NR_XL(…)
The Brzustowski Flare Model at is used in the ERCBflare to determine to position
and dimension of the flare flame. The position parameter XL is a function is
calculated based upon a guess of the position and a dimensionless flare position
parameter SL.
Spreadsheet function call:
FT
=NR_XL(XL_guess, SL_goal)
maxloops = 100
err = 0.1
A
Public Function NR_XL(guess As Double, goal As Double)
Dim a As Double
Dim b As Double
Dim C As Double
Dim diff As Double
Dim slope As Double
Dim intercept As Double
Dim err As Double
Dim maxloops As Integer
Dim i As Integer
a = 1.04 * guess ^ 2
b = 2.05 * guess ^ 0.28
C=a+b
diff = C - goal
slope = 2 * 1.04 * guess + 0.28 * 2.05 * guess ^ (0.28 - 1)
R
isdone = -1#
i=0
While (isdone < 0)
guess = guess - diff / slope
D
a = 1.04 * guess ^ 2
b = 2.05 * guess ^ 0.28
C=a+b
diff = C - goal
slope = 2 * 1.04 * guess + 0.28 * 2.05 * guess ^ (0.28 - 1)
If (Abs(diff) < 0.1) Then isdone = 1
i=i+1
If i >= maxloops Then isdone = 1
Wend
guess = guess - diff / slope
NR_XL = guess
End Function
References:
Brzustowski, T.A.. 1976. Flaring in the Energy Industry. Prog. Energy Combust.
Sci. pp. 129-141.
8 References
122
FUNCTION p_sat(…)
The saturation pressure of steam can be calculated given the temperature using
Gibbs free energy formulation. The Gibbs coefficients are listed on the
PROPERTIES page.
Spreadsheet function call:
=p_sat(temperature)
Where Temperature (Kelvin) is the temperature of the steam.
FT
The basic code for the function is:
A
'ref: The International Association for the Properties of Water and Steam, 2007
'Revised Release on the IAPWS Industrial Formulation 1997 for the Thermodynamic Properties of Water and Steam
'Dimensionless saturation equations
Function p_sat(T As Double, gibbs As Range) As Double
Dim sum As Double
Dim ni As Double
Dim agibbs As Variant
Dim a As Double
Dim b As Double
Dim C As Double
Dim v As Double
agibbs = gibbs.value
v = T + agibbs(9, 2) / (T - agibbs(10, 2))
D
R
a = v * v + agibbs(1, 2) * v + agibbs(2, 2)
b = agibbs(3, 2) * v * v + agibbs(4, 2) * v + agibbs(5, 2)
C = agibbs(6, 2) * v * v + agibbs(7, 2) * v + agibbs(8, 2)
' Psat in MPa
p_sat = 2# * C / (-b + Sqr(b * b - 4# * a * C))
p_sat = p_sat * p_sat
p_sat = p_sat * p_sat
p_sat = p_sat * 1000# ' return kPa
End Function
Reference:
The International Association for the Properties of Water and Steam, 2007
'Revised Release on the IAPWS Industrial Formulation 1997 for the
Thermodynamic Properties of Water and Steam
http://www.iapws.org/
8 References
123
FUNCTION t_sat(…)
The saturation temperature of steam can be calculated given the pressure using
Gibbs free energy formulation. The Gibbs coefficients are listed on the
PROPERTIES page.
Spreadsheet function call:
=t_sat(Pressure)
Where Pressure (kPa) is the pressure of the steam.
FT
The basic code for the function is:
A
'ref: The International Association for the Properties of Water and Steam, 2007
'Revised Release on the IAPWS Industrial Formulation 1997 for the Thermodynamic Properties of Water and Steam
'Dimensionless saturation equations
Function t_sat(p As Double, gibbs As Range) As Double
Dim sum As Double
Dim ni As Double
Dim agibbs As Variant
Dim d As Double
Dim e As Double
Dim f As Double
Dim g As Double
Dim b As Double
agibbs = gibbs.value
R
' p in MPA
b = (p * 0.001) ^ 0.25
e = b * b + agibbs(3, 2) * b + agibbs(6, 2)
f = agibbs(1, 2) * b * b + agibbs(4, 2) * b + agibbs(7, 2)
g = agibbs(2, 2) * b * b + agibbs(5, 2) * b + agibbs(8, 2)
D
d = 2# * g / (-f - Sqr(f * f - 4# * e * g))
' t_sat in K
t_sat = agibbs(10, 2) + d - Sqr((agibbs(10, 2) + d) ^ 2# - 4# * (agibbs(9, 2) + agibbs(10, 2) * d))
t_sat = t_sat * 0.5
End Function
Reference:
The International Association for the Properties of Water and Steam, 2007
'Revised Release on the IAPWS Industrial Formulation 1997 for the
Thermodynamic Properties of Water and Steam
http://www.iapws.org/
8 References
124
FUNCTION FNz(…)
This function calculates the compressibility of a gas given its critical temperature
and critical pressure using the Redlich–Kwong equation of state. The Redlich–
Kwong equation of state is an empirical, algebraic equation that relates
temperature, pressure, and volume of gases.
Spreadsheet function call:
=fnZ(Temperature, Pressure, Critcal_Temperature,
Critical_Pressure)
FT
Where T (Kelvin) is temperature, P (kPa) is pressure, Critical Temperature
(Kelvin) and Critical Pressure (kPa)
The basic code for the function is:
A
' http://www.chem.mtu.edu/~tbco/cm3450/Compressibility_from_Redlich_Kwong.pdf
' Redlich-Kwong equation
' based on Cutlip and Shacham, 2008, pp. 101-103 (see Seader, Henley & Roper, 3rd Ed, 2011)
Function fnZ(T As Double, p As Double, Tc As Double, Pc As Double)
Dim Tr As Double
Dim Pr As Double
Dim a As Double
Dim b As Double
Dim q As Double
Dim r As Double
R
On Error GoTo error_handler
Tr = T / Tc
Pr = p / Pc
a = 0.42747 * Pr / Tr ^ (5# / 2#)
b = 0.08664 * Pr / Tr
r=a*b
q=b*b+b-a
D
fnZ = mcroot(1, -1, -q, -r)
Exit Function
error_handler:
fnZ = "**ERROR**"
End Function
8 References
125
Function mcroot(a3 As Double, a2 As Double, a1 As Double, a0 As Double)
'
' Computes the maximum real root of the cubic equation
' a3 x^3 + a2 x^2 + a1 x + a0 = 0
'
Dim a As Double
Dim b As Double
Dim C As Double
Dim d As Double
Dim z As Double
R
A
FT
a = a2 / a3
b = a1 / a3
C = a0 / a3
p = (-a ^ 2 / 3 + b) / 3
q = (9 * a * b - 2 * a ^ 3 - 27 * C) / 54
Disc = q ^ 2 + p ^ 3
If Disc > 0 Then
h = q + Disc ^ (1 / 2)
Y = (Abs(h)) ^ (1 / 3)
If h < 0 Then Y = -Y
z=Y-p/Y-a/3
Else
theta = Atn((-Disc) ^ (1 / 2) / q)
c1 = Cos(theta / 3)
If q < 0 Then
s1 = Sin(theta / 3)
c1 = (c1 - s1 * 3 ^ (1 / 2)) / 2
End If
z1 = 2 * (-p) ^ (1 / 2) * c1 - a / 3
M = a + z1
r = (M ^ 2 - 4 * (b + M * z1)) ^ (1 / 2)
z2 = (-M + r) / 2
z3 = (-M - r) / 2
z = z1
If z2 > z Then z = z2
If z3 > z Then z = z3
End If
mcroot = z
End Function
Reference:
D
Cutlip and Shacham, 2008. Based on Seader, Henley & Roper, 2011. Separation
Process Principles, 3rd Ed.
http://www.chem.mtu.edu/~tbco/cm3450/Compressibility_from_Redlich_
Kwong.pdf
8 References
126
Credits
Mapping Code
FT
Projection transformation to and from latitude/longitude and Universal Transverse
Mercator (UTM) coordinates were based upon code from General Cartographic
Transformation Package (GCTP). The General Cartographic Transformation
Package (GCTP) is a system of software routines designed to permit the
transformation of coordinate pairs from one map projection to another. The GCTP
is the standard computer software used by the National Mapping Division for map
projection computations.
Reference:
ftp://edcftp.cr.usgs.gov/software/gctpc/
A
Nearest Neighbour Code
R
The nearest neighbour algorithms are translated for VBA by Sergey Bochkanov
using the ALGLIB project code. This program is free software; it can be
redistributed it and/or modify under the terms of the GNU General Public License
as published by the Free Software Foundation (www.fsf.org). ALGLIB is a crossplatform numerical analysis and data processing library. It supports several
programming languages (C++, C#, Pascal, VBA) and several operating systems
(Windows, Linux, Solaris).
D
Nearest neighbor search is an important task which arises in different areas pattern recognition. ALGLIB package includes nearest neighbour subpackage,
which implements nearest neighbor search by means of kd-trees. Kd-trees allow
to perform efficient search in low-dimensional spaces (from 1 to 5), but have
lesser performance in high-dimensional spaces.
Reference:
http://www.alglib.net/
8 References
127
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement