TNO Safety software RISKCURVES Version 9 Quick Start Guide and User Manual
TNO Built Environment & Geosciences Department of Industrial and External Safety Princetonlaan 6 PO. 80015 NL-3508 TA Utrecht, the Netherlands Fax. +31 88 86 62050 Email:
[email protected]
© 2013 TNO
RISKCURVES © 2013 TNO All rights reserved. No parts of this work may be reproduced in any form or by any means - graphic, electronic, or mechanical, including photocopying, recording, taping, or information storage and retrieval systems - without the written permission of the publisher. Products that are referred to in this document may be either trademarks and/or registered trademarks of the respective owners. The publisher and the author make no claim to these trademarks. While every precaution has been taken in the preparation of this document, the publisher and the author assume no responsibility for errors or omissions, or for damages resulting from the use of information contained in this document or from the use of programs and source code that may accompany it. In no event shall the publisher and the author be liable for any loss of profit or any other commercial damage caused or alleged to have been caused directly or indirectly by this document.
Contents
3
Table of Contents Foreword
0
Chapter I Introduction
7
1 TNO software ................................................................................................................................... products 7 2 Installation ................................................................................................................................... 8 System requirem .......................................................................................................................................................... ents 8 The protection.......................................................................................................................................................... key 8 Installation and.......................................................................................................................................................... de-installation 8 Upgrading from.......................................................................................................................................................... previous versions 8
3 What ................................................................................................................................... is RISKCURVES 12 Which task can .......................................................................................................................................................... RISKCURVES perform 12 What is the required .......................................................................................................................................................... input 13 What kind of results .......................................................................................................................................................... are obtained? 13
Chapter II Quick Start Guide
21
1 A new................................................................................................................................... Graphical User Interface 21 2 The concepts ................................................................................................................................... behind the tree nodes 22 3 Quick................................................................................................................................... start: Create a new project 27 1 Add a background .......................................................................................................................................................... m ap 28 2 Verify calculation .......................................................................................................................................................... settings 28 3 Define m eteorological .......................................................................................................................................................... conditions 29 4 Define population .......................................................................................................................................................... distribution 30 5 Define Stationary .......................................................................................................................................................... or Transport equipm ent locations 36 6 Add Scenarios .......................................................................................................................................................... to equipm ent location 37 7 Entering consequence .......................................................................................................................................................... m odel set data 39 8 Perform ing .......................................................................................................................................................... the risk calculation 41 9 Evaluate results .......................................................................................................................................................... of the calculation 42 10 The use of .......................................................................................................................................................... Cum ulation sets 43 11 The use of .......................................................................................................................................................... Com parison sets 44 12 The use of .......................................................................................................................................................... Analysis points 44
Chapter III The user interface in detail
47
1 Menu................................................................................................................................... bar 49 2 Toolbar ................................................................................................................................... 50 3 Project ................................................................................................................................... tree 52 4 CalculationSet ................................................................................................................................... definition 54 5 Equipment ................................................................................................................................... definition 55 6 Scenario ................................................................................................................................... definition 56 7 Analysis ................................................................................................................................... points 58 8 Result................................................................................................................................... panel tabs 59 9 Graph................................................................................................................................... display panel 60 Presenting Model .......................................................................................................................................................... Results 61 Base functionality .......................................................................................................................................................... graphs 62
10 (Autohide) ................................................................................................................................... Scenario selection panel 64 11 Command ................................................................................................................................... button panel 65 12 Node ................................................................................................................................... input panel 66 © 2013 TNO
4
RISKCURVES 13 Graph................................................................................................................................... selection box 67 14 Profile ................................................................................................................................... expert button 68 15 Map display ................................................................................................................................... panel 69 Presenting geographic .......................................................................................................................................................... calculation results 71 Positioning equipm .......................................................................................................................................................... ent 73 Map functionality .......................................................................................................................................................... 73
16 Report ................................................................................................................................... panel 76 17 Model................................................................................................................................... log panel 78 18 Legend ................................................................................................................................... panel 80
Chapter IV Advanced features
83
1 Options ................................................................................................................................... 83 2 Display ................................................................................................................................... units 85 3 Presentation ................................................................................................................................... settings 87 4 Expert ................................................................................................................................... Parameter settings 89 5 Meteorological ................................................................................................................................... distribution 91 6 Vulnerability ................................................................................................................................... settings 92 7 Environment ................................................................................................................................... settings 93 8 Accuracy ................................................................................................................................... settings 94 9 Chemical ................................................................................................................................... Databases 95 Chem ical database .......................................................................................................................................................... editor 96 Selecting a chem .......................................................................................................................................................... ical from the database 97 View ing/Editing .......................................................................................................................................................... properties of chem icals 99 Chem icals convertor .......................................................................................................................................................... 104
10 Mass................................................................................................................................... and volume calculator 105 11 Mortality/probit ................................................................................................................................... calculator 106 12 Geo-referencing ................................................................................................................................... images 107 13 Risk ................................................................................................................................... transects 110
Chapter V Technical backgrounds
113
1 QRA................................................................................................................................... Definitions 113 Calculation Set .......................................................................................................................................................... 113 Calculation Settings .......................................................................................................................................................... 113 Accuracy param .......................................................................................................................................................... eters 113 Vulnerability.......................................................................................................................................................... param eters 116 Environm ent.......................................................................................................................................................... param eters 120 Meteorological .......................................................................................................................................................... data 121 Population .......................................................................................................................................................... 121 Equipm ent .......................................................................................................................................................... 122 Scenario .......................................................................................................................................................... 123 Modelset .......................................................................................................................................................... 123 Cum ulation set .......................................................................................................................................................... 123 Com parison.......................................................................................................................................................... set 123 Analysis points .......................................................................................................................................................... 123 Dam age definition .......................................................................................................................................................... 124 Societal Risk.......................................................................................................................................................... 125 Individual Risk .......................................................................................................................................................... 125 Iso Risk Contours .......................................................................................................................................................... 126 Societal Risk.......................................................................................................................................................... (FN) Curve 127 Project file .......................................................................................................................................................... 128
© 2013 TNO
Contents SR Maps
5
.......................................................................................................................................................... 128
2 The ................................................................................................................................... consequence models within a modelset 133 Gas release .......................................................................................................................................................... 134 Gas release ......................................................................................................................................................... from a vessel or pipe 134 Gas release ......................................................................................................................................................... from a long pipeline 135 Liquefied gas .......................................................................................................................................................... release 136 DIERS top......................................................................................................................................................... venting (vessel only) 137 Vapour release ......................................................................................................................................................... from vessel or pipe 138 Pressurized ......................................................................................................................................................... liquefied gas release from vessel or pipe 138 Spray release ......................................................................................................................................................... of pressurized liquefied gas from vessel or pipe 139 Instantaneous ......................................................................................................................................................... flashing liquid release 141 Liquefied ......................................................................................................................................................... gas from long pipeline 142 Liquid release .......................................................................................................................................................... 143 Pool evaporation .......................................................................................................................................................... 144 Atm ospheric.......................................................................................................................................................... dispersion 146
3 Model ................................................................................................................................... input parameters 149 4 Combined ................................................................................................................................... models 150 5 Cumulation ................................................................................................................................... of sources 152
Chapter VI Appendices
157
1 List of ................................................................................................................................... chemicals 157 2 Low ................................................................................................................................... level error messages 158 3 Known ................................................................................................................................... limitations 160
Index
© 2013 TNO
163
Introduction
1
7
Introduction This manual is delivered in Adobe’s PDF-format. You are free to print the manual for your own use with respect to the license conditions of the software. You can access this manual directly from within the software from the “Help” menu or via the installed entry in the Windows START-menu. If you discover any omissions, errors or inconsistencies, we kindly ask you to contact us directly via email (
[email protected]) or use the built-in email feature accessible from the “About” box.
1.1
TNO software products TNO Department of Industrial and External Safety delivers two different software products: EFFECTS and RISKCURVES. This manual describes the RISKCURVES product that is currently available for safety related calculations. Within the Risk Calculation Core, many EFFECTS models are used for calculation of the consequences. For that reason, some parts of this manual also reflect to those EFFECTS models. EFFECTS EFFECTS performs "consequence" calculations to predict the physical effects (gas concentrations, heat radiation levels, peak overpressure's etc.) of the escape of hazardous materials. Results are presented in either textual or graphical format. Models in EFFECTS are based upon the Yellow Book, third edition, second print 2005 [1]. EFFECTS can also model complex releases by linking individual models in such a way that they describe all physical phenomena that may occur during that release. For example a liquid release will consist of a release model, connected to an evaporation model, which is then linked to a dispersion model that calculates the concentration profiles in the environment. Finally it might be linked to an explosion model to calculate the ultimate effects due to peak overpressure's or heat radiation if the chemical is flammable and ignites. RISKCURVES RISKCURVES is a fully-featured computer program to perform Quantitative Risk Analysis (QRA). It is capable of calculating individual risk, societal risk, and consequence areas of multiple potential accident scenarios. Scenarios can be entered either as a predefined damage zone, or be calculated by internal consequence models. Furthermore, scenarios can be defined on one location (as a "stationary" equipment), or on a route (as as a "transport" equipment). RISKCURVES wil also analyse risk by means of the most dominant contributor, construct all types of societal and individual risk curves, display risk contours, calculate transport risk per kilometer of route and visualise relevant data in its GIS viewer. More information about RISKCURVES can be obtained from our sales department. Contact via email:
[email protected] or http://www.tno.nl/RISKCURVES.
© 2013 TNO
8
RISKCURVES
1.2
Installation
1.2.1
System requirements RISKCURVES 9 is developed for the Windows 7 OS platform. It will also run on Windows XP or Vista platforms, but some features and screen layout are less optimal. There are no special requirements with respect to memory or disk space other than a free USB-port to accommodate the protection key. For more complex calculations, RISKCURVES benefits from additional internal memory and a faster processor. RISKCURVES 9 runs on Windows 7 64-bit as a 32 bit application and has been thoroughly tested on 64 bit environment.
1.2.2
The protection key To prevent unauthorised use, the software is protected with a special protection key (“dongle”). The dongle should be connected to a free USB-port of the PC during use of RISKCURVES. The protection key represents the ownership of the license and will only be replaced by TNO if proven defective. In case of loss or theft of the key, replacement is the sole responsibility of the user. The protection key is remotely programmable by TNO and contains information about the owner and specification of the granted license (program, version, options etc.). Before its first use or if the license conditions change (due to the purchase of an upgrade or additional option), the key will need to be reprogrammed once. This process requires access to email functionality and instructions will be included with the delivery of the software.
1.2.3
Installation and de-installation Installing and de-installing the software needs sufficient administrative rights to do so. You may need to contact your IT-department for support on this requirement. Apart from installation, administrative right are not required. The software comes with a full-featured setup program. Please follow the instructions carefully and make sure the protection key is NOT inserted during installation. The default driver for the protection key that is installed is from “Hasp/Sentinel”. If the same driver but in a newer version is required by another software program, the installation for RISKCURVES can be omitted. If in that situation RSIKCURVES does not recognise the protection key correctly, please contact the TNO helpdesk:
[email protected] De-installation is done by selecting the appropriate entry in the installed program group under the Windows START menu. Alternatively, one can use the "Uninstall" option within the "Programs" group in the windows control panel.
1.2.4
Upgrading from previous versions The projects of RISKCURVES 9 are not backward compatible with previous versions 7. Depending on the version of the project, one or more steps are necessary to convert the project to the format of RISKCURVES 9. Note that the previously stored calculation results will not be included in the converted project. In cases where a calculation model has newly added input parameters in the converted version, these fields will show up empty. Pressing “Defaults” will fill-in the default values without overwriting the already entered fields.
© 2013 TNO
Introduction
9
Upgrading from RISKCURVES 7 Projects RISKCURVES differs in project file format from RISKCURVES 7. The new files are identified by their “.riskcurves” extension. The RISKCURVES 7 project files use the “.clc” extension. To use RISKCURVES 7 projects in RISKCURVES 9, they need to be converted with the RISKCURVES Project Converter. The converter can be started from the Windows START menu under “RISKCURVES ”, or in the Windows file explorer by right-clicking on an RISKCURVES 7 project file and selecting “convert”. Once the required RISKCURVES 7 project file is selected, clicking the “Convert” button will create a new project file suitable for RISKCURVES in a definable project folder. The original RISKCURVES 7 file will not be overwritten. All input needed to recalculate will be included, including population and meteo definition. Rresults are not converted since they are probably not valid since the calculation accuracy and method has changed. Note that some models in RISKCURVES 9 require additional parameters compared to RISKCURVES 7. Furthermore, the damage definitions have more strict checking on the definitions than the previous consequence interfaces. The new RISKCURVES now requires to use increasing distances for decreasing lethality levels. Due to this test, some translated scenario's might be presented in red, indicating wrong or missing input. By default all scenarios with input data will be converted.
© 2013 TNO
10
RISKCURVES
User defined chemicals RISKCURVES 9 uses a new database model. User chemicals defined in previous versions have to be converted using the Database Converter. This converter can be found in the startmenu "Chemicals Converter". To convert user defined chemicals from your old database (.rdb file), start the converter and click 'Open File'. Select the database file with your user defined chemicals and click 'Open'. You can now select chemicals in the left-hand list, and drag them (or click the '>>' button) to the right-hand list: this will import the selected chemicals into the new database labeled as user-defined chemical. The right-hand list will show you all chemicals in the new user database.
The filter field can be used to search for chemicals. If you've imported too much chemicals by mistake, you can delete them later in the Database Editor.
Upgrading from RISKCURVES 4 (DOS version) There is no build-in support for RISKCURVES project older than version 7.
© 2013 TNO
Introduction
11
To be able to use version 4 input files, it is advised to convert these input files by the update procedure of RISKCURVES 7 itself. This will provide a RISKCURVES 7 (.clc and .inf) format file, which can be read within the new RISKCURVES 9.
© 2013 TNO
12
1.3
RISKCURVES
What is RISKCURVES RISKCURVES is a computer program package to perform a Quantitative Risk Assessment (QRA) for activities with hazardous materials. “Risk” is defined in this context as the probability per unit of time (frequency) that humans in the vicinity of the hazardous material may suffer lethal consequences due to an unwanted release. A Quantitative Risk Assessment (QRA) analyses the risks of accidents involving with dangerous substances, resulting in lethal victims, injuries and/or material damage to surroundings. In order to be able to compare risks, quantitative values are given for Individual Risks and Societal Risk. RISKCURVES uses an intuitive modular approach, allowing the use of topographic maps or aerial pictures to define potential “Loss of Containment Scenarios”. Advanced geographic presentations can be created by using the internal GIS presentation system. Based upon the Yellow Book [1], Green Book [2], and Purple Book [3], the effect and consequence models included within RISKCURVES provide a sound, scientific and transparent basis to perform a QRA. The number of degrees of freedom for a QRA is huge. The publication CPR 18E (Purple book: Guidelines for quantitative risk assessment) provides important guidelines on choosing equipment, (Loss of Containment) scenario’s to be evaluated, effect-calculation models and background information like meteorological data. Al information from this Purple Book is implemented within RISKCURVES, providing a coherent and consistent QRA calculation tool.
1.3.1
Which task can RISKCURVES perform In a QRA specific accident scenarios are defined, specifying subsequent events. For example: an outflow, pool formation, evaporation, dispersion eventually leading to lethal effects. The calculation of chances and effects of the identified events will lead to quantitative values for Individual Risks and Societal Risks. Individual Risk are usually presented as “Iso Risk Contours”: lines on a topographic map which represent point with equal PR value: e.g. the 10-6 contour. RISKCURVES is capable of performing physical effect, probability and consequence calculations to calculate individual and societal risk. It can also calculate risk caused by the transport of hazardous materials. It facilitates data entry and presentation of the results by using a highly visual approach. Results are presented by means of individual risk contours, societal risk curves (FN-curves) and optionally provides Societal Risk Maps. (A Societal Risk Map is a visualisation of the societal risk at a specific location: illustrating either absolute level or relative contribution at that location). The number of degrees of freedom for a QRA is huge. The publication CPR 18E (Purple book: Guidelines for quantitative risk assessment) provides important guidelines on choosing equipment (so-called sub selection method), the typical "Loss of Containment" scenario’s to be evaluated, effect-calculation models and background information like meteorological data.
© 2013 TNO
Introduction
1.3.2
13
What is the required input The typical questions to be raised when performing a QRA are the “What” and “Where” questions: What are the typical Loss Of Containment (LOC) scenarios, and where are they situated. The geographic location of the scenario will also determine the specific environmental parameters: temperature, pressure, humidity, surface roughness (influencing dispersion) and meteorological data (probability of wind coming from specific direction with specific speed and specific stability class). Within CPR-18, it is prescribed which equipments and which scenarios should be evaluated within a QRA. Within RISKCURVES, the user can simply add any equipment which is present within the system boundaries of the site to be studied. The location of the equipment needs to be provided as equipment coordinates or a route. It is strongly advised to use a digital background map of the area as underground. Once the equipment is positioned, RISKCURVES can add the corresponding LOC scenarios (G1, G2, G3), including the failure rate frequencies and event or phenomena dependent probabilities. A LOC scenario is provided as a pre-calculated consequence zone, called damage definition (previous version used the indistinctive term “Consequence Interface”) or as a consequence calculation model. For the latter case, EFFECTS consequence models are being used within the calculation core of RISKCURVES. It is also possible to copy / paste EFFECTS end model calculations into a RISKCURVES scenario. In general, the following information is required to perform a QRA: 1. Definition of one or more accident scenario(s) which includes the applicable frequencies of the accident scenarios. RISKCURVES 9 now distinguishes an equipment: containing the location of a possible event, the possible scenario's at that equipment: defining the frequencies, and an underlying modelset: defining the damage zones. 2. Definition of Meteorological probability distributions: this includes stability class / wind directions; for multiple stability classes (Pasquill A..F) and wind sectors. When unknown, a standard (equal) distribution can be used. 3. It is very convenient (but not obliged) to use a digital background map to position the scenarios and use as a background layer when presenting results. 4. Environmental conditions typical for the location of the study area (temperatures, humidity, solar radiation) need to be provided. 5. Vulnerability conditions describe relations between phenomena and resulting damage (lethality). 6. In case of societal risk calculations you will need a population distribution
1.3.3
What kind of results are obtained? RISKCURVES will perform all chance, effect- and damage/consequence calculations according to CPR 18E, and will calculate the Individual Risk and Societal Risk. A person who is on a specific location will have a chance to be a lethal victim of an accident at installation A and an accident at location B. Both chances can be added together and presented on a geographic map as Iso Risk Contours.
© 2013 TNO
14
RISKCURVES
The individual risk criteria assumes 100% presence and an unprotected situation outside. A so-called “Iso Risk Contour” can be drawn by connecting all points with equal Individual Risk.
The Individual Risk can also be presented in a so-called FX curve, which presents the fraction lethal versus distance from the release point, for different wind-directions.
© 2013 TNO
Introduction
15
Risk contours are available on the level of a calculation set, cumulation sets, comparison sets and individual equipments. A Risk transect can be provide for a specific line track. Such a transect will provide the risk as a function of the place along this track.
So called analysis points can be added a any location to present Risk Ranking reports at that spot. Such a report will present a ranking of all scenarios based on their contribution to the (individual) risk at that point. The Societal Risk Curve (FN-curve) presents the cumulated risk that a group of specific size will be killed. The FN curve is depicted as a two dimension graph, using a logarithmic scale on frequency F (Y-axis) and number of victims N (X-axis) axis’s. The curve is interpreted using a “Guide value”, which is a line that should preferably not be crossed. RISKCURVES will present a “Guide Ratio R” value, indicating the distance to this guide value (a guide ratio >1 implies exceeding the guide line), and also presents the “Expected value E” which is the size of the area below the FN curve.
© 2013 TNO
16
RISKCURVES
A FN curve appears to be not very easy to understand or explain. The curve is the result of spatially distributed risk sources that may influence a geographically distributed population distribution, whereas the result only present a curve. Questions that often raise are: “Do we have a problem” and “Where is this problem” or “What is causing this problem”. To be able to answer these kind of questions, a Societal Risk Map was developed and these presentations are now available within RISKCURVES 9. Two different maps can be presented: the SR area map, which indicates whether the guide ratio is higher than one on a specific spot. The SR area map illustrates affected zones, and height of the societal risk at a specific spot. SR (Societal Risk) Maps is basically a geographical "Area Specific Societal Risk" presentation of a societal risk, being a two dimensional curve. As a result of the demand for a visualization of the societal risk, a new type of presentation was developed in 2007. The question was raised when a societal risk calculation is fed with geographical based information on population, and geographical based scenario locations, why can we not see a geographical distribution of the societal risk. Such a presentation would be very convenient for emergency response (were are the people who are threatened by accident) or urban planning activities (how much space left for population without exceeding societal risk limits: the guide value)
© 2013 TNO
Introduction
17
To provide answers to both question two types of graphs were developed: the Societal Risk Area Map and the Societal Risk Contribution map. The Societal Risk Area map gives an indication of which areas are affected and the height of the risk whereas The Societal risk Contribution Map gives an indication which cells contribute to the societal risk
The bases for the presentation is that every grid cell from the population grid has its own FN curve. In the case of the Contribution map, this curve relates to the victims within this population cell. The higher the risk of this cell (expressed as the expected value of the curve) the more red the color will be.
© 2013 TNO
18
RISKCURVES
For the contribution map, the expected value is used to translate the two dimensional FN graph into a color. The type of coloring can be adjusted, it appears that using a 6 color levels (use legend ) provides the best contrast, but other coloring might improve the visualization.
This way the curve represents the full societal risk of scenario's for the area. Note that this area bounded FN curve will never exceed the overall FN curve for all cells.
For the societal Risk Maps it is important to understand that the risk is determined from the receivers point of view (instead of from source).
© 2013 TNO
Introduction
19
Furthermore, because of the nature of the method, cumulating of various risk sources is possible: transport & stationary installations, small & large scenario’s
The idea behind this new type of visualisation is that this provides a supplementary view of what is happening, and the maps facilitate considering societal risk in early stage of planning process: - the SR Area map shows areas with restrictions - the SR Contribution map shows which areas contribute most (emergency response)
© 2013 TNO
20
RISKCURVES
The underlying FN graphs per location, that are use to derive these map presentations, are presented for every "Analysis (risk ranking) point". The societal risk at the location will display the FN curve of all scenarios that are affecting this location, the contribution FN graph will display the FN curve for the population within the population grid cell.
© 2013 TNO
Quick Start Guide
2
21
Quick Start Guide This guide gives you the opportunity to quickly get acquainted with the possibilities of RISKCURVES by defining some simple examples and tasks, and explaining the operation of the software from a users’ point of view. The guide is not a detailed reference guide, nor a complete “how to” user manual but helps user to quickly understand the concepts of RISKCURVES version 9. At any time you may refer to the full technical reference manual which is provided with RISKCURVES, or consult the built in help system by pressing
for more detailed information.
2.1
A new Graphical User Interface Compared to the previous version 7, the graphical user interface (GUI) of the program has been completely redesigned. Version 7 used a calculation phase dependent “Tabbed” interface, which tended to obscure input. The new RISKCURVES 9 uses a tree view oriented user interface, which reflects the hierarchy of input data. Furthermore, the screen layout resembles the GUI of the latest version of our consequence tool EFFECTS: The left hand side of the screen is input, and right hand side present (tabbed) results, dependent of the “active” (currently selected) item. The new interface provides intuitive methods for copying and pasting user definitions (such as scenarios) but remains uncluttered and clean. Access to standard actions, such as adding or deleting nodes is provided by context menu’s (popup menu: right mouse click) and standard shortcuts (Del, Ctrl-C, Ctrl-V). The standard screen is divided in three parts: a (hierarchical) tree view, an input parameters panel, and a (tabbed) result panel. The input and results panel always reflect their contents to the active (selected) node of the tree. Switching from active node will also change the contents of the input and result panels, eliminating the use of a “present” or “edit” button.
© 2013 TNO
22
RISKCURVES
The top of the main form contains typical menu items, providing conventional access to main functions and a toolbar. The left bottom side contains the three main buttons < Clear>, < Defaults> and < Calculate>.
2.2
The concepts behind the tree nodes The hierarchical tree view which illustrates the input, contains some important concepts or typical definitions, which will be introduced in more detail below.
© 2013 TNO
Quick Start Guide
23
The top node, called Riskcurves Project is a typical placeholder of all user input. The project corresponds to the information stored in a typical file, reflected by the name of the node. The name of the topnode reflects the current project filename. The caption style of a node also reflects the current state: italics imply "not calculated yet", a red caption indicates incomplete or wrong input.
The tree view illustrates the hierarchy which is automatically occurring while defining input for a QRA: A calculation set is a typical input definition for a single QRA calculation: it contains all input that influencing the result. Since users often want to compare the change in risk due a modification (of population, scenarios), a RISKCURVES project can contain multiple Calculations Sets in one project (and thus file).
© 2013 TNO
24
RISKCURVES
Results of a calculation are influenced by their general parameters, which are combined in Calculation settings”; these settings will be applied to all scenarios belonging to the calculation set. Furthermore, the calculation of a “Societal Risk” requires the definition of a population distribution. Population is always associated to a calculation set, and can be defined as a grid (cell based distribution) or polygons (geometric shapes containing population). The equipment node is used to define the geographic positioning of scenarios; a typical tank, vessel, installation or transport route can be defined at this level. Note that the background map can be used to select a coordinate or define route points. Each equipment can contain multiple scenarios: for vessels typical scenarios may include a leak, a full bore rupture, and a catastrophic (instantaneous) release scenario. Each scenario however, has its own failure frequency, describing the chance of this accident happening, and consequence description. RISKCURVES is capable of using its own internal (EFFECTS) consequence model, which can be a fire model, dispersion model or explosion model, but users can also define a damage definition. These Damage definitions can be used to enter results from consequence calculations from external models, or use damage zone definitions which may be standardised or prescribed by authorities. Copying and Pasting of Nodes The use of the hierarchical tree node allows the possibility to copy scenarios (same leak on another installation), copy equipments (same set of scenarios belonging to a vessel copied to another location) , or even copy entire calculation sets (same calculation but with altered frequency, population etc).
A calculation set is a combination of system setting, a meteorological definition, population and accident (Loss of Containment) scenarios definitions for which Individual Risk and Societal Risk are being calculated.
A calculation set will have results in terms of Individual Risk Contours and Societal Risk Graphs and Societal Risk Maps. A calculation set is a typical input definition for a single QRA calculation: it contains all input that influencing the result. Since users often want to compare the change in risk due a modification (of population, scenarios), RISKCURVES can contain multiple Calculations Sets in one project (and thus file).
© 2013 TNO
Quick Start Guide
25
A calculation set always contains the sub nodes Calculation settings, a Meteo data node, Population (if societal risk calculation is required), stationary equipment and transport equipment, because these contents together determine the result of a calculation. Calculation settings is a typical collector or grouping node.
It doesn’t have its own parameters, but combines several groups of parameters, to be applied to all input contained in a calculation set. Typical parameters are “Accuracy” describing parameters influencing calculation accuracy and speed, “Vulnerability” settings describing the relation between physical phenomena and damage (lethality), and “Environment” parameters, describing ambient temperature, humidity, solar radiation etc. for the typical location. The meteorological data definition contains the choice for the meteorological station to be used. Any meteorological data set contains probabilities for typical weather classes (Pasquill stability class, wind-speed, day or night) occurring at the location (see meterological distribution). The number of weather classes defined will determine how many damage definitions / consequence models are contained under a scenario (e.g only D5 and F2 or 6 different classes!). Population definition node contains the definition of population by means of grids (a matrix like definition of cells) or polygons (area definition with number of inhabitants). Population can be added by using the Population Import Wizard, or by manually adding a polygon and defining an are with population. See defining Population. The total cumulation of all grids and polygons under the grouping node will be used to create a total population grid, used within the calculation sets Societal Risk calculation.
Both day an night grid will use a separate "Inside fraction" determining the fraction of the people that are inside houses and have a some degree of protection (see vulnerability settings) When using "temporary polygons", it is possible to use a dedicated "inside fraction" and "utilization fraction" (a presence factor). Temporary population can be used to include the presence of large crowds (e.g. festivals, sport events) during a FRACTION of the time. This is particular relevant if large numbers of people are outside (thus unprotected).
© 2013 TNO
26
RISKCURVES
Note: When using many (say more than 10) temporary polygons that can be exposed to the same event (when they are close to one another, so within the potential lethality footprint of a single event), this procedure can get time consuming because all potential combinations of these areas need to be evaluated !!!. As an example, just for three temporary population area’s we need to evaluate: A and B and C exposed, A and B exposed, A and C exposed, B and C exposed, only A, only B , only C, and no area (just base population) exposed, where every combination has its own probability of occurrence!!
Equipment: a location or route on which scenarios are being analysed (distinguishing STATIONARY and TRANSPORT equipment). Note that these nodes can be expanded, they are placeholders or grouping nodes for a list of coordinates, or routes.
Scenario: a Loss Of Containment scenario occurring at an equipment (either a stationary location or a transport route), which has a specific failure frequency, and contains consequence definitions: a description of the scenario in terms of substance, quantities, release situation or resulting damage. A Modelset is the placeholder for the actual consequence definition. It contains either a damage definition or consequence calculation, which is defined for a number meteorological conditions
It is possible to define altered input values for specific weather class conditions by selecting the weatherclass from the combobox.
A cumulation set can be used to make a dedicated cumulation of risk sources that does not contain all equipment or all scenario's, presented corresponding SR or IR results.
© 2013 TNO
Quick Start Guide
27
Simple right click the cumulation node and select < Add Cumulation set> . The corresponding input panel allows to select or deselct any scenario or equipment from the list. A Comparison set allows to compare results for Calculations Sets or Cumulation sets; it will provide multiple graphs and contour.
2.3
Quick start: Create a new project The first step in performing a QRA with RISKCURVES is creating a project. Start RISKCURVES (see “Installing the software and starting RISKCURVES”) and choose File | “New” from the main menu or press the “New project” toolbar button. The user is asked for a project name, and an empty project tree will be created.
To create and run a QRA calculation, the following steps need to be taken: 1. Add a background map 2. Verify Calculation settings 3. Define meteorological conditions 4. Define population distribution 5. Define Stationary or Transport equipment locations 6. Add Scenarios to equipment location 7. Entering consequence model set data 8. Performing the risk calculation
© 2013 TNO
28
RISKCURVES
9. Evaluate results of the calculation 10.Optionally, define Cumulation sets , to make a subset of scenarios 11.Optionally, define Comparison sets , to compare different calculations or cumulations 12.Optionally, define Analysis points, to compare risk at specific locations
2.3.1
1 Add a background map The use of a topographic background maps is very useful and highly recommended. Got to the background node and select < Right mouse> < Add background>. A new (Red) node will be created. Select the node and use the browse button in the input panel to select a background file. Currently supported formats are: JPG, PNG, BMP, TIF and TIF as pixel oriented files, and DXF and SHP as vector formats.
Note that pixel formats require the use of a ESRI worldfile to be able to determine scale and location of the image, whereas the vector formats SHP and DWG already include scale and position information. RISKCURVES supports the use of multiple backgrounds, so a background can be composed of adjacent or overlapping images.
2.3.2
2 Verify calculation settings Without any additional user input, RISKCURVES will perform calculation with default settings for “Accuracy” “Vulnerability” and “Environment” parameters.
© 2013 TNO
Quick Start Guide
29
The “Environment” block is most likely to be modified since it describes typical environmental conditions applicable for the region were the QRA is to be performed: parameters like ambient temperature, water temperature, humidity, surface roughness, solar radiation flux, latitude and cloud cover are country and even location dependent.
The values that are entered here will be "pushed" into each modelset that will perform a consequence calculation. One can define or alter default environment to be used when creating a new project by using menu “Edit” “Options” “Default environment”.
2.3.3
3 Define meteorological conditions The probability of a risk occurring at a specific location is highly influenced by the probability of the wind blowing from the accident location towards that location. In order to take this into account, a meteorological definition has to be supplied. Meteo data consists of the definition of typical Pasquill stability class with a wind speed (e.g. D5 or F2), the probability of that class occurring, and the probability for the wind-directions for that class and is applicable for the region where the scenario’s are to be defined. This data is usually supplied by meteorological station at airports etc. and can be predefined for met-stations at your country. A new meteo-station definition can be added under menu “Edit”, “Option” “default meteo distribution” and the browse button:
© 2013 TNO
30
RISKCURVES
All definitions that are provided in “Options” can be selected in the combobox in the input panel of the Meteo data node:
Once a meteo station location has been entered, the red label will turn to black, illustrating that acceptable input has been provided. Note that the provided (Dutch) meteo station definitions all contain 6 weather classes, but it is also possible to use only two any other number of different Pasquill classes. Using only two classes imply that calculation time will be reduced since the consequence models need to perform two calculations. The spatial distribution of occurance of specific wind directions can be visualised in a windrose view
2.3.4
4 Define population distribution This step is only required if a user wants to calculate societal risk, which includes risk of actual exposed population. Population always distinguishes separate population during day and population at night definitions. To add population, select the node and use < Right mouse> or < Add Population Polygon>.
© 2013 TNO
Quick Start Guide
31
Note that as of version 9, it is possible to combine multiple grids or polygons; the resulting end population will be created on the base of ALL grids an polygons defined in the yotal population node, using the defined population grid cell size.
Population import Wizard A population grid can be imported from an ASCII based grid, ESRI grid format (provided by local authorities) or the previous RISKCURVES Vs 7 .POP file. Apart from grid based input files, population can be also created from polygons loaded a shape file. To import an external grid, select the "import population" from the popup menu at a population node.
Follow the suggestions by either selecting a Population Grid (cell based distribution) or select Population Polygons, which are separate area definitions that can be edited separately after importing. When using highly detailed shape files, containing real "building" descriptions, it is advised to translate this into a grid (because of the huge number of shapes these files can contain), when using "region based" shapes, it can be useful to import these as separate polygons. For grids, the type of file to import needs to be defined: it can be an ESRI grid , ASCII / CSV table, RISKCURVES Vs 7 POP file or created from a Shape file.
© 2013 TNO
32
RISKCURVES
An ESRI grid file contains a header, describing the dimensions (number of rows, columns and cell size) and location (position of lower left corner) of the grid, followed by the grid values themselves. The header also contains a value, which is treated as empty cell, this value is often defined as as -999 or -9999. A version 7 POP file also contains the complete grid definition including location and dimensions. An ASCII table assumes the data to be available as separate lines, containing X coordinate, Y coordinate, Population, possibly separated by spaces or other delimiting characters. The import screen offer the possibility to define the decimal separator, and a field separator character, and to skip one or more header lines.
© 2013 TNO
Quick Start Guide
33
A SHAPE file contains descriptions of polygons with population info about those regions: When using a shape file as a grid, all shapes will be combined into one grid based definition. This requires the definition of a grid cell size and selection of fields for daytime / nighttime population.
After importing the file, the boundary definitions of the grid can be provided by defining lower left and upper right corner of the grid:
© 2013 TNO
34
RISKCURVES
Population Polygons A population polygon is a definition of areas with specific population information, they can be created by drawing a shape on the background map or importing a shapes from a SHAPE file. Importing polygons from SHAPE file When importing a shape file as polygons, all shapes will be added separately as population polygons.When defining by means of a shape file, the fields containing relevant info in the shape tables need to be defined. Select the name of the field containing the description of the region, and the field that contains the number of people (day/night). Furthermore, the population value provided can contain a density (value is population per area: select the corresponding unit by using the right mouse button on the units description) or an absolute number of people.
© 2013 TNO
Quick Start Guide
35
After importing polygons, it is possible to edit the shapes, and potentially define specific areas as "temporary population", which implies that a specific utilization fraction can be entered. These “Temporary populations” are intended for usage in special situations like festivals, sport events or other situations where non-permanent presence of large amounts of people can occur during a FRACTION of the time. This is particular relevant if large crowds are outside and have no protection by houses. The usage of multiple "temporary" population polygons also implies that that multiple areas can be affected by an event, leading to the situation of combination of victims. The current calculation procedure also checks for potential occurrence of MULTIPLE “Temporary” (even if they have 100% presence) populations, and accounts for the potential PROBABILITY of multiple polygons being exposed, with potential COMBINED NUMBER OF VICTIMS. Note: When using many (say more than 10) temporary polygons that can be exposed to the same event (when they are close to one another, so within the potential lethality footprint of a single event), this procedure can get time consuming because all potential combinations of these areas need to be evaluated !!!. As an example, just for three temporary population area’s we need to evaluate: A and B and C exposed, A and B exposed, A and C exposed, B and C exposed, only A, only B , only C, and no area (just base population) exposed, where every combination has its own probability of occurrence!!
Manual definition: Zoom in on the area of interest (use mouse wheel for zooming, right mouse drag for moving the map) and select the edit button. Start pinpointing coordinates on the map, thus defining the shape (polygon) of the habituated area. Select the edit button when definition is finished and enter the number of people within this area during day and during night. For standard usage select “is temporary” as NO.
© 2013 TNO
36
RISKCURVES
Give the polygon or grid a specific and recognizable name: select the population polygon, and click on the text “population polygon” or press to be able to modify the name of the branch. Note: Renaming a tree node can also be used on Calculation Set, Equipment or Scenario !
2.3.5
5 Define Stationary or Transport equipment locations Select the Stationary equipment branch and use < Right mouse> . Rename the “Equipment” using or selecting the string and provide a useful descriptive name.
Select the map view, zoom in to the location where the equipment is placed, hover the mouse to the exact location and select < Set Release point> . The current world coordinates of the mouse will be entered in the input fields “X coordinate”and “Ycoordinate” of release. Select the “Show release point” toolbar button location with a label and cross on the map.
to illustrate the
© 2013 TNO
Quick Start Guide
37
Transport equipment: Select the Transport equipment branch and press and again rename the “Equipment” using or selecting the string and provide a useful descriptive name. . Select the map view and zoom in to the area where the route is to be defined. Press the “Edit” button in the transport equipment input panel and start pinpointing route points on the map. Watch all route coordinates being added to the table when selecting route points on the map.
Finish the route definition by pressing “End Edit” again. Note that it is still possible to manually modify the coordinates. The “correction factor” column can be used for switches on railroad tracks or locations where a local altered failure frequency needs to be applied.
2.3.6
6 Add Scenarios to equipment location Once stationary equipment locations or transport equipment routes have been defined, typical LOC (Loss Of Containment) scenarios belonging to the equipment can be added. Select the equipment and press and select the type of scenario to be added from the branch of models:
EFFECTS models are consequence calculations performed by single phenomena consequence models. They can either be based on atmospheric dispersion of toxic or flammable gasses or based on heat radiation (Bleve , poolfire of jetfire phenomena). Combined models support multiple phenomena; if a material is both flammable and toxic, or direct and delayed iginition can occur, these combined LOC model chains will distinguish several phenomena.
© 2013 TNO
38
RISKCURVES
The combined models are supplied for Gaseous, Liquid and Two phase materials, and are available for specific release cases. A release can be either an instantaneous release (called G1 scenario in the Purple Book), a release within 10 minutes (G2 scenario) or a leak scenario with a specific hole size (G3 scenario). If the user doesn’t know the state of , one can select the Unified LOC model, which determines the state itself, and provides a choice to evaluate Damage definition s can be used to enter pre-calculated consequence areas. The damage models are also dedicated to a specific phenomenon. Another possibility to add scenario is by using the floating panel: Select an equipment node, and hover the mouse over the white line on the left border of the RISKCURVES window. A model selection panel will unfold, illustrating different possibilities by family name:
After the scenario has been added, the definition itself needs to be provided. A scenario definition consists of two elements: a frequency part and a consequence part.
© 2013 TNO
Quick Start Guide
39
The tree visualises this as two nodes of the scenario: the scenario node and the corresponding (consequence) modelset. For a scenario, main parameters are base frequency (expressed as chance of occurrence per year), a possible correction factor (which can be used to represent risk reduction actions), and a daytime fraction. The daytime fraction can be used to express the situation that an activity only takes place during day or night time. By default, this fraction should be the average occurrence of daytime situation, according to the meteorological data definition (e.g. for Netherlands 44% is daytime). If another fraction is used, this implies that the activity is predominantly shifted into day or night time. Combined models also require entering a fraction for direct ignition, delayed ignition, BLEVE and explosion phenomenon. For single phenomenon models, is it assumed that this fraction is already included in the base frequency. Pressing the < Defaults> button will quickly enter feasible frequency / probability values here, but is not advised because failure frequencies tend to be very specific for the typical situation.
2.3.7
7 Entering consequence model set data Dependent on the type of model (single phenomenon model, combined model or damage definition) a dedicated input parameter list will be presented. An first example is provided for a BLEVE damage definition: This input is defined by a fireball radius (100% lethality inside and outside), a 35 kW/m2 radius (same lethality as within fireball), and a lethality versus distance response table which defines unprotected outside lethality.
© 2013 TNO
40
RISKCURVES
This lethality table needs to be entered in a logical ascending distance / descending lethality order. If input is invalid, the table caption will turn red. Note that the weatherclass combobox can be used to define either ALL (default) or ONE specific meteorological condition. Start with entering all (Default), and IF specific damage distances occur (such as expected in case of toxic dispersion phenomenon), select distinguished weather classes and enter dedicated distances. Note: If a weather class specific parameter is displayed in a blue color, it means that it is identical to the default situation. This illustrates the fact that in the background this parameter is linked to the default model (see EFFECTS for details about model linking) A consequence model definition is basically identical to an EFFECTS model definition: the input panel displays all relevant input parameters. In fact, it is also possible to copy/paste EFFECT end models into RISKCURVES. (An end models implies that the model ends up with any lethality information, eg. a single outflow model is no end model)
The number of required input parameters for an EFFECTS model can be changed depending of the setting of complexity: Simple, Normal or Expert. The three toolbar buttons on top of the main window will define this state. It is advised to start using “Simple mode”, which only requires main parameters (Chemical, amount of material released) to be entered, and only use “Normal” or “Expert” if one wants to divert from standard method. In “Expert” mode, all parameters that influence the result of the calculation are shown, providing the possibility to alter parameters like ambient temperatures or other default value parameters defined in “System Settings”.
© 2013 TNO
Quick Start Guide
41
If an EFFECTS model has been selected as scenario type, this consequence model will be calculated for a number of weather conditions, equal to the typical Pasquill classes defines in the meteo data node. This results in a Set of models in which every model can have specific input. The weather class dedicated input can be accessed by selecting the appropriate weather class from the combobox.
Note that the way the scenario node is displayed, reflects the current state: -
a red scenario means data is incomplete or incorrect
an italic presentation means that input has changed and the node needs to be (re) calculated a blue presentation of a specific weather class model implies that the data is linked to the default weather class model
2.3.8
8 Performing the risk calculation After defining scenarios a calculation can be performed. The calculation can be started by
© 2013 TNO
42
RISKCURVES
pressing the button on the bottom of the screen. Dependent of the number of (modified or uncalculated) scenarios, a calculation can take seconds, minutes or hours for large (hundreds) scenario sets. Note that consequence (EFFECTS) models will not be recalculated if only a location or frequency has been changed, a scenario will not be recalculated if only location changed, and equipment is skipped if nothing has changed etc. Only modified input needs to be redone, where the calculation of a societal risk, which is a accumulation of several scenario/ equipment contributions, will always be redone.
During calculation, several progress bars will be presented, to give an idea about the current progress status. If for some reason, a scenario or equipment is skipped, the calculation will proceed with the next scenario, and store the result of previous calculation!
2.3.9
9 Evaluate results of the calculation After the calculation is finished, the Log tab will display any abnormalities, using a Yellow color for warnings, and Red for errors. The severity will also be represented by the LED icon on top of the log window.
© 2013 TNO
Quick Start Guide
43
Note that the contents of the Log window reflects the current active node. If a calculation set is active, all Logs of underlying nodes will be included. To see dedicated results for one equipment, or even one scenario, this node has to be activated (selected). The same selection method applies for all other results: specific results from equipments (Individual risk contours and FN curve) can be evaluated by selecting the equipment, a scenario has results in terms of individual risk per wind-direction (FX graphs) or even calculated consequence distances can be evaluated by selecting the required weather class.
Main results of course can be found on “Calculation Set” level: The complete set of scenarios and equipments will result in a Individual Risk Contour map, and Societal Risk curves. The Societal risk of transport scenarios is depicted in transport FN cuvres, applicable for a section of the route. These graphs are part of the results of individual transport equipments (per route).
2.3.10 10 The use of Cumulation sets Very often one is not interested in the fully accumulated results of all scenario’s, but want to know the contribution of a specific subset of scenario, e.g. only flammable scenario’s or accumulation of specific vessels or equipments. Such a subset can be made using a Cumulation Set. Define a new set by selecting the “Cumulation sets” node and selecting . Give it a descriptive name (e.g. “Only Flammables”) and use the checkboxes to select which equipments or separate scenarios should be incorporated within this accumulation. After pressing the button, which only takes a few seconds, the subset results will be presented.
It is important to realise that a Cumulation set can also be used to ADD different calculation sets. This way, it is possible to combine calculations for different parts of a site, for example containing different production processes, and add them all together in one result as a cumulation. This cumulation result will include Iso Risk contours, Societal risk graphs and SR maps.
© 2013 TNO
44
RISKCURVES
2.3.11 11 The use of Comparison sets The new RISKCURVES has the possibility to perform multiple QRA calculations, and compare results. This can be used to validate the influence of a changing population, or generally: a changed risk situation. To perform multiple QRA calculations, the most rigorous way would be to copy and paste an entire calculation set: select the node for the calculation set, press and (or use edit copy / paste) and a complete calculation set will be added. Again, use descriptive names for the different calculation sets, e.g. “Larger storage capacity” or “Including new Urban Development population” and modify the contents of the copied calculation set accordingly. After calculation (which may take some time again), these results can be compared using the comparison Set.
However, in many cases it is possible to add a new or modified scenario to the standard calculation set and use a Cumulation set to exclude this from being added to the result. Since all defined “Cumulation Sets” are also included within a Comparison set, this can be used to verify the influence of a modified scenario, without the need to copy the entire calculation set. Realise that copying a full calculation set will result in big projects with many duplicate scenarios.
To start comparing different calculation sets or cumulation sets, define a “Comparison set” by selecting the node and selecting . All calculation sets and cumulation sets of the project will be visible here. Use the checkboxes to include or exclude a set. When comparing individual risk contours, only one level of interest will be shown. This particular level can be modified within the presentation settings.
2.3.12 12 The use of Analysis points For every calculation or cumulation set, analysis points can be defined, providing the possibility to analyse the contribution of scenarios at specific locations.
© 2013 TNO
Quick Start Guide
45
An analysis point can be added to any calculation or cumulation set. To add an analysis point, use < Right mouse> < Add analysis point> on the analysis points node, or use < Right mouse> , < Add analysis point> on top of the map to pinpoint a coordinate from the map. The results will be visible after a calculation, and presented in a table in the report tab.
© 2013 TNO
46
RISKCURVES
© 2013 TNO
The user interface in detail
3
47
The user interface in detail RISKCURVES 9 has abandoned the tab based approach from the previous version, and now uses a hierarchical tree view, which is also similar to the latest EFFECTS user interface. The screen consists of standard elements like a menu bar and toolbar, and a main screen which is divided in three zones: the project tree, the input panel and the result panel. The project tree depicts the various components that are required to perform a QRA calculation. The tree support cut and paste functionality and the object hierarchy illustrates how LOC equipments are part of a calculation, scenarios take place at an equipment etc. The input of the “active” (currently selected) node of the tree is presented in the input panel. The contents of this panel will change dependent of the type of node that is selected. The same dependency applies for the result panel, which will always display results (report, graphs, maps and log) for the currently selected node. The figure below shows an arbitrary user interface screen that might be visible during any stage of a calculation and with all possible options enabled. The user interface has been designed in such a way that it follows the rules of a standard Windows user interface as close as possible. Click on the Item letters or screen area to get detailed information about the control item:
© 2013 TNO
48
RISKCURVES
The figure above shows the new Graphical User Interface (GUI) of the software. The arrows point to the most important controls of the GUI. All areas/controls are indicated with a letter ("A"..."L"). This letter is also used in the paragraphs to identify which part of the GUI is described. A. Menu bar B. Toolbar C.Project tree D.Model input panel E. Results panel tabs F. Graph display panel G.(Autohide) Scenario selection panel H. Command buttons I. Model input parameters J. Profile selection box K. Profile expert button Furthermore, other screens that can be selected through the Results screen tabs are: · Contour display panel
© 2013 TNO
The user interface in detail
49
· Report panel · Model Log panel
3.1
Menu bar The menu bar contains the menu items to control the main functions of the user interface and is set up in the way a common MS-Windows application is supposed to work.
The menu is separated into 5 main categories: Menu File . . The File menu provides access to: New:
Creates a new empty project. Clears the memory.
Open:
Opens an existing project.
ReOpen:
Shows recent files and allows to select any of these recent files.
Save:
Stores the current project contents to disk
Save as:
Stores current contents under a new name
Menu Edit These menu items provide standard clipboard functionality (Cut, Copy, and Paste) for all items. As a clipboard can only hold one type of data at a time, you will have the possibility to copy either the profile, or the (GIS screen) Contours or the Report as HTML data to the clipboard. These can be pasted in the normal way in any document. The copy functionality is particular useful when copying parts of the project tree, such as complete scenarios or even calculation sets. Copying and pasting can also be used between two instances (running applications) of RISKCURVES, and allows to use population definitions, equipments or scenarios from one project to another. It is also possible to copy paste end models from EFFECTS (version 8/9) into RISKCURVES 9. This is only feasible for so-called end models: models that end up with lethality levels. In general: all models that are available within RISKCURVES consequence calculation core have support for copy / paste from EFFECTS. (A release model is NOT an end model). This feature may become handy if one wants to evaluate the behaviour of a specific model within EFFECTS before using it in a QRA. Another important item in Edit is “Options”. Selecting this item will show the options screen, where users can define default settings, chemical database, default units, environmental, vulnerability and accuracy settings. A description of this feature is given in options description
© 2013 TNO
50
RISKCURVES
Menu View Allows to enable disable the view of different segments of the toolbar, select the complexity level or activate graphic or map view features.
Menu Tools This Tools menu allows access for the following tools: · · · · ·
Mass and Volume calculator Mortality / probit calculator The remote dongle update program The RISKCURVES Vs7 to Vs9 project convertor The Chemicals convertor
Menu Help Provides access to the help file, help file table of contents and version release notes.
3.2
Toolbar The toolbar contains buttons for quick access to common functions and is divided in several groups dependent upon their functionality.
From left to right it contains buttons for: Group 1: File and print
· · · ·
New project Open an existing project Save a project Save a project as…
Group 2: Copy & paste
These are the standard cut, copy & paste tools. Note that the contents of the clipboard is determined by the currently active region of the screen. For a profile or contour, the clipboard contents will be an image, when the active component is a node, complete nodes of the project tree can be copied and pasted.
© 2013 TNO
The user interface in detail
51
Instead of the buttons, the standard windows Ctrl-C (copy) and Ctrl-V (paste) hotkeys are often more convenient. Group 3: User complexity settings
These three buttons can be used to switch complexity level of the list of input. Note that especially for the combined models, the list of input can be extensive. Because many of the input parameters will always be used in default setting, or are taken form the environment or system default parameters, the user required input can be simplified to much less input parameters. Currently, three levels of complexity are supported: Simple, Normal and Expert mode. The last mode will always show all input parameters that influence the calculation. Group 4: Profile tools
A profile graph can be the societal risk FN curve, but might also present consequence model results such as a time or distance depending values, as heat radiation versus distance, or concentration versus time. Cross hair cursor: provides the possibility to show a crosshair, which will illustrate the X,Y values of the point under the cursor Ruler: activates the ruler, which can be used to measure the distance between two points Group 5: Contour tools
The grid tool shows a grid definition in the map, which can be handy for reading out positions Will show the location of equipment locations and analysis points. This is an on/off toggle button. Any equipment location or analysis point can be illustrated with a dot on the contour image. Transport routes will also be displayed as a line in the same map layer. Show crosshair: Illustrates the coordinate of the location at the cursor, and if a "grid" layer is active (selected in the legend) , provides information about the value of the location at the cursor Geo-reference the background image. Will invoke a screen that can be used to georeference a pixel based background.
© 2013 TNO
52
RISKCURVES
A pixel oriented graph needs to have a definition for the size and positioning of the image in real world coordinates. For this purpose, EFFECTS uses the ESRI standard georeference method which requires a wordfile definition for every image. Currently supported pixel formats are JPG, TIFF, BMP and PNG files. Ruler: activates the contour ruler, which is a measuring device to be used for obtaining absolute sizes of clouds, areas, or distances to objects on a background map. Transect: provides the possibility to determine the individual risk along a line section. By defining a transect line (click and drag the cursor to define a track) the transect panel, which is located below the legend panel, will display the risk along this track. Lock zoom: this toggle button can be used to force the map view to keep the same field of view on every component. Full extent: rescales the map to the full extent (all objects visible)
3.3
Project tree The project tree contains a list of all equipment and corresponding scenarios in the project.
© 2013 TNO
The user interface in detail
53
The user may switch between different active nodes by simply clicking on the appropriate branch of the tree. Selecting the node will consequently display its properties, and its results or reports in the result panel Node that some nodes are grouping nodes, and need to be expanded before the actual contents is visible. Stationary equipment is a placeholder for all typical point sources, whereas transport equipment contains scenario's to be defined as a line source: pipeline, railroads, highways etc. Node in Italics or red The caption of the node may illustrate the current state of the contents: red indicating incomplete (not all values entered) or providing an error after a calculation, an italic caption means that this node has not been calculated with the current contents (it may be new, or changed after the previous calculation).
Renaming nodes
© 2013 TNO
54
RISKCURVES
Most nodes can be renamed, and given an appropriate name. Select the node and press or select the node and subsequently select the label of the node. It is strongly suggested to use descriptive names for equipment and scenario's, e.g. "Storage vessel" and "Instantaneous rupture" scenario. The sorting of the nodes is always based on alphabetical order. This can be used to give a logical ordering by adding numbers in the name of the model.
Copying and Pasting scenario's Any non-grouping node support a copy-paste action. Use Edit ..Copy / Edit.. Paste (or C and -V) while a node has been selected.This will create a copy of the node, containing the same input as the original. It is also possible to copy an entire calculation set containing all scenario's. This can be very convenient when creating alternative development situations. One calculationset might contain the current population, whereas a second set includes new urban development plans. Furthermore, a new calculationset will be included in a comparison set, allowing to display results from multiple calculation situations. Removing equipments or scenarios from the project A node can simply be deleted by pressing the button while the node is selected. A dialog will ask for a confirmation for the removal of the node.
Hotkeys for collapsing or expanding nodes The use of the numerical keypad keys + (plus: expand to previous) or - (minus: collapse) can be used to expand or collapse the tree quickly. The * (star) will expand the full tree.
3.4
CalculationSet definition A calculation set is the placeholder for a complete QRA calculation, creating individual risk results and possibly societal risk results. A calculation set is a combination of system setting, a meteorological definition, population and accident (Loss of Containment) scenarios definitions for which Individual Risk and Societal Risk are being calculated.
© 2013 TNO
The user interface in detail
55
A calculation set will have results in terms of Individual Risk Contours and Societal Risk Graphs and Societal Risk Maps. A calculation set is a typical input definition for a single QRA calculation: it contains all input that influencing the result. Since users often want to compare the change in risk due a modification (of population, scenarios), RISKCURVES can contain multiple Calculations Sets in one project (and thus file). A calculation set always contains the sub nodes Calculation settings, a Meteo data node, Population (if societal risk calculation is required), stationary equipment and transport equipment, because these contents together determine the result of a calculation.
In standard situations, one usually works with ONE calculation set. However, it is possible to use multiple (independent) calculation sets within one project (thus a RISKCURVES project file). Multiple calculation sets can be used to compare different calculations, e.g. one with a base population and one with modified population due to urban development plans.
Calculation set parameters : Perform societal Risk calculation: Is a Yes/No choice: A Societal Risk Calculation, resulting in so-called Societal Risk graphs, (FN curves) requires the availability of population information. If Yes is selected, a population definition needs to be available, either as grid or as polygons. Note that an Individual Risk calculation, resulting in a map with Iso Risk Contours, will always be performed Create Societal Risk Maps: Is a Yes/No choice, if these calculations are activated, the map view will also display SR maps Cumulate transport routes in FN graphs: Is a Yes/No choice, by default No. If the user selects to cumulate transport routes, the total FN curve will include the results for the complete route. if cumulation is skipped, transport scenario's will provide Transport FN graphs, valid for a specific section of the route (by default 1 km).
3.5
Equipment definition An equipment is defined either as STATIONARY (equipment is at one coordinate) or as a TRANSPORT definition, where the equipment requires route definition. Add an equipment by pressing .
© 2013 TNO
56
RISKCURVES
Adjust the new (red) name "Equipment" and use a good description. A stationary equipment has two parameters containing the coordinates.
Use the background map to pinpoint the exact location of the equipment by using For TRANSPORT equipment, a route can be defined by pointing clicking coordinates on the background map. See positioning equipment
3.6
Scenario definition A scenario definition contains information about the typical "Loss of Containment" event, and is always located at an equipment (either transport or stationary). The choice for the scenario type determines the typical event to be modelled. An important choice is to define a known consequence footprint, called a damage definition, or to let internal EFFECTS model calculate the damage zones, depending on the installation definition (the amount, type and storage conditions of the chemical). Add a scenario by using or use the scenario selection panel By default, the scenario will be named according to the type of scenario added. The effects (consequences) part of a scenario is contained within a modelset definition (depicted by an fX icon). The number of model calculations within a modelset is depending on the meteorological definition. Rename the scenario and give a good descriptive name by using or selecting the node label.
© 2013 TNO
The user interface in detail
57
Scenario input requires: Base frequency: The failure frequency for the scenario, expressed per year. Although this parameter has a default, it is highly recommended to modify this according to the actual failure frequency. Frequency correction factor: A scenario frequency might deviate from "standardised" situations, due to risk reduction measures, dedicated situation etc. By using a correction factor instead of adjusting the base frequency, adjustments can be made more traceable. Frequency equally distributed day/night: By default, it is assumed that a frequency of an scenario is equally distributed over nighttime and daytime; that is according to its meteorological occurrence. However, some activities (loading unloading etc) may have a certain preference for either day or nighttime. By changing this choice to "No", users can define user specified (so deviating from meteorological distribution) value. Fraction frequency in daytime hours: The base frequency is the total frequency for daytime and nighttime. Users can define activities to take place only at daytime hours (fraction daytime = 100%), only at nighttime (fraction = 0%) or any other value. The value entered here will determine which part of the total frequency is used for daytime situation. Chance direct ignition: The probability that a direct (immediate) ignition event takes place. This parameter is only relevant for scenario's in which multiple phenomena (poolfire and vapour cloud explosion, jetfire and vapour cloud explosion) are possible (combined models). In case of damage definitions, the calculation is restricted to a single event: an explosion damage definition already assumes that the explosion takes place. By default a value of 0.8 is used, but this value can be altered, because it is dependent of the type chemical (flammability) or release rate. Chance delayed ignition: The probability that a delayed ignition (flash fire and/or vapour cloud explosion) event takes place. Note that the sum of Direct ignition and Delayed ignition does NOT have to be one: 1 - (Fraction direct + Fraction delayed) = fraction No Ignition
© 2013 TNO
58
RISKCURVES
Bleve fraction: The probability of a Bleve event taken place. A Bleve can only occur with a instantaneous release and may be one of the immediate ignition events (other immediate ignition event can be a poolfire). This parameter only has influence in case of multiple phenomena and is only applied in case of instantaneous two phase releases of flammable materials. Fraction with explosion phenomena: Given the occurrence of a vapour cloud explosion (which is regarded upon as a delayed ignition), this event may have overpressure effects. This parameter describes which fraction of those events will have overpressure effects. Note that, different from other QRA tools, TNO assumes that all delayed ignitions will have a flash fire phenomenon, and only a part of thes flash fires will ALSO have overpressure effects.
3.7
Analysis points Analysis point can be used to report risk contribution at specific user definable locations. Any analysis point will produce a risk ranking per scenario, based on risk contribution at that location. Furthermore, the societal risk FN graph of all scenarios affecting that location will be presented, illustrating the severity of the societal risk at that location. This FN graph per location is the base for the societal risk area map. The FN contribution graph will illustrate the societal risk curve for the typical population within this population grid cell of this coordinate. These location specific FN graphs are used as the bases for the SR contribution map. Analysis point will be illustrated on the map when the "analysis point" is the active component in the tree or whenever the "Show equipment locations" toolbar option has been selected.
An analysis point can be defined from any map view illustrating the Iso Risk Contours, such as calculation set or the cumulation set. To add an analysis point, use < Right mouse> < Add analysis point> on the analysis points node, or use < Right mouse> , < Add analysis point> on top of the map to pinpoint a coordinate from the map. The results will be visible after a calculation, and presented in a table in the report tab.
© 2013 TNO
The user interface in detail
3.8
59
Result panel tabs On top of the right half of the screen, four tabs provide access to the different result viewers:
For more information on various result screens, refer to: · · · ·
Map display panel Graph display panel Report panel Model Log panel
Note that the log panel also contains a LED light warning sign, which is used to illustrate the status of the log messages
© 2013 TNO
60
3.9
RISKCURVES
Graph display panel The profile result will present typical graphic representations that are available for the selected node. This implies that the contents of the graph is dependent of the active node.
One of the most important graphs is the Societal Risk Graph which is available for a Calculation Set or Equipment. Note that a transport equipment contains a slider which can be used to illustrate the FN curve for a specific section of the route. Depending on the contents of the graph, the X or Y scale may be adapted to logarithmic view, when this is commonly used for displaying the typical graph. Typical graphs available are: 1. Calculation set: The FN curve for all stationary equipment. If the option "cumulate transport FN" is checked, the societal risk of the full route is ALSO included in this graph !! 2. Equipment: the FN curve for this equipment. This might be a transport FN curve, that has a slider to select the section of the route to display
© 2013 TNO
The user interface in detail
61
3. Scenario: In expert mode, the FX graph is illustrated here. The FX graph is a "Individual Risk versus Distance" graph presentation, available in all winddirections. It can be seen as a polar representation of the risk of this particular scenario. All these polar results together, positioned on their corresponding location, eventually create the Individual Risk Map (presented as contours) 4. Modelset: A modelset will contain result for different weather classes (D5 day, F2 night etc). On a consequence model level, all consequence model results will be shown in this panel as well.
3.9.1
Presenting Model Results For evaluating the detailed results of a consequence model calculation, the appropriate weatherclass needs to be selected in the weather class combobox. Depending on the type of model (single phenomenon or combined consequence model), the number of graphs can differ, but usually these profiles will illustrate a time or distance depending behaviour of a result parameter.
The graph selection box can be used to browse through all graphs that can be provided by the model whereas the small button next to the profile selection box enables the possibility to view "Profile expert graphs"
© 2013 TNO
62
3.9.2
RISKCURVES
Base functionality graphs The graph presenter is equipped with a convenient zoom and scroll functionality which is entirely operated with mouse: Zooming into a graph can be done by selecting a zoom area with your LEFT mouse. To do so, point and click in the graph, hold the left mouse button and drag the mouse from top left to right down while holding the left mouse button. The program will show a rectangle. When you release the left button, the area in the rectangle will be zoomed.
To unzoom the current graph, select an arbitrary zoom area from bottom right to left up (the opposite way around as you zoom, which is from left to right down). Alternatively the profile can be zoomed/unzoomed scrolling the mouse wheel the same way as the contours can be zoomed. The profile is zoomed on the point the mouse cursor is pointing at that moment. Moving a graph is achieved by dragging with the RIGHT mouse button. Drag the chart while clicking the right-mouse button (the cursor will change to a hand), and the current viewing area can be changed. Note that an action will undo this modification of the viewport area and will revert the graph back to the (automatically scaled) graph boundaries. Edit, Copy and Freeze Pressing the right mouse button on top of the profile graph will open a popup menu with options Edit, Copy and Freeze. The Edit choice will invoke the build in graphic editor which provide access to all settings of the graph, including properties as titles, scales and legend placement, but also contains an export tab, which enables the possibility to save the graph either as data or any specified file type graph.
© 2013 TNO
The user interface in detail
63
The edit dialog can also be accessed by doubleclicking on the profile graph. The Copy choice will put a high resolution copy of the current view on the clipboard. The Freeze selection will create a "clone" of the current graph which is no longer connected to the underlying model. This can be convenient if one wants to evaluate or compare diffrent versions of a calculation. Axis units All axis units of the profile graph can be changed by right clicking on the axis itself, just like all other numerical values.
© 2013 TNO
64
RISKCURVES
Crosshair and Ruler The crosshair tool, which can be activated by pressing the in the toolbar, provides a moving crosshair, which displays the current X and Y values of the mouse cursor. The ruler option (toolbar button ) will show a ruler in the graph. The boundaries of the ruler can be moved with the mouse, providing a way to measure the distance between specific points.
3.10
(Autohide) Scenario selection panel On the left side of the input screen, a small "grip" is shown. Whenever the mouse is hovered above this area, the scenario selection panel will unroll. Whenever the mouse moved outside this area, the panel will automatically hide again.
This panel provides a direct access to all available scenario's. Note that adding a scenario is only possible if an equipment is the active node: a scenario needs to be added to a location or a route !! Note that the panel is oriented by model family: Combined models, Dispersion models, Fire models and Damage definitions (no calculation but a predefined damage zone). Selecting a model here will have the same result as selecting it through the popup menu under Equipment: Add scenario.
© 2013 TNO
The user interface in detail
3.11
65
Command button panel This panel contains the three important buttons to perform calculations:
The buttons perform basic actions on the model input screen. The clear and default buttons will perform actions on the current selected session and leave other sessions unchanged. The calculate button acts on the entire project and will recalculate all currently unavailable results. The buttons have the following tasks: The “Clear” button This button will clear the input screen by making all fields empty. Any unsaved data in the current session screen will be lost The “Default” button When this button is clicked, it will fill the input area with default data. Note that the “Default” button will NOT overwrite any existing data. This means that the function of this button is that it will add default data to input screens that have some fields left blank. If you want to substitute all data in a screen with default data you will have to press the “Clear” button (see next paragraph) prior to the “Default” button. This is particularly important when using the "Simple" or "Normal" mode, with a limited number of input parameters. Note that a consequence model can not run if it does not have all input parameters entered. If the user is working in "simple" or "normal mode" (see toolbar) some of the input parameters MAY be hidden for the user. Before the calculation can be performed, all hidden empty parameters are filled with their default values.
© 2013 TNO
66
RISKCURVES
The "Calculate" button Pressing this button will start a calculation of the project. The calculation will be performed for all currently modified scenarios or equipment that contain valid data. Currently modified scenarios, or equipment, can be recognized by italic representation in the project tree. If the calculation result in warnings or error, the log screen will automatically open, illustrating the status of a calculation in color codes and a light. During a calculation, a progress window will illustrate the progress, and give some feedback on the current scenario or equipment being calculated. Note that unchanged scenario's will not be recalculated; their modelset results will remain unchanged. For a societal risk calculation, the results will need to be translated to total victims and cumulated towards a total FN graph. Recalculate ALL: Calculate Recalculation of all scenario's can be forced by pressing plus the calculate button. This will force complete recalculation of all effect models, scenario's and equipment contained in the current project. Calculate from here It is also possible to force recalculation of a specific node: e.g. a single scenario, a single equipment or a single calculation set. Select the node and press < Right mouse> : Calculate from here. Note that if a single scenario is recalculated, but is part of a larger set, the calculation set results will not be automatically updated.
3.12
Node input panel Whenever a node has been selected, all required input (input parameter) for the active node of the project tree will be displayed here. You can enter data by clicking on a white edit box and type the data you want to use. By pressing the or + key, you can navigate through the input fields. Depending on the active node: a Calculation Set, Population, Equipment, Scenario's or Modelset the contents of the input panel (required input) will change. When defining a modelset, this panel will contain all typical input for the effect consequence model. Typical changes in the input of a combobox might influence availability of fields: selecting an evaporating pool will disable the "liquid fraction" input: which will be grayed out. Furthermore, color highlighting is used to reflect the state of an input field: all linked parameters (copied from default weatherclass) are depicted with blue description, whereas non-linked parameters remain black. Furthermore, if parameters are missing, the label font will be is red, illustrating missing input. Parameters that are in a light yellow edit box can be recognized as "expert" parameters, and will only be visible if the button is pressed. These parameters will normally be taken from their corresponding default settings which can be environment settings like temperatures etc, or expert parameter for all other defaults.
© 2013 TNO
The user interface in detail
67
Unit conversion input parameters The input screen offers a convenient way to use any unit you like for entering the data. Right click on the unit and select the required unit. Note that chemical dependent parameters (such as Lower Explosion Limit or LEL value) will perform the required mass/volume translations automatically:
3.13
Graph selection box When a calculation is performed, a model can deliver more than one graphical result. As we can only display results of the same type in one graph simultaneously, the program will store all different types of graphs in a selection box. For example, a BLEVE (EFFECTS) model might present a heat radiation versus distance diagram as well as a mortality-distance diagram.
© 2013 TNO
68
RISKCURVES
Since only one graph can be visible at any moment, you can find the other graphs here. When you press the down arrow, all other graphs become visible (see below)
Depending on which graph you choose, the graphical area (see graphical presentations) will be updated automatically. Note that some models, like the combined models, have an extensive list of profiles, requiring to use the scrollbar in the selection box !
3.14
Profile expert button The small button next to the profile selection box enables the possibility to view "Profile expert graphs". This feature allows the user to select multiple profiles, Pressing this button will open a new window, with all currently available profiles listed on the left side. The user can "tick" any graph.
If profiles with different units are selected, the graph will use both left and right (even bottom/ top) axis to present the different graphs. All features, available for a standard profile graph, such as (popup menu) Edit, Copy, Freeze and Unit axis setting are also available in the Profile expert.
© 2013 TNO
The user interface in detail
3.15
69
Map display panel Map panel: a GIS presentator The contour display panel will provide a GIS presentation of all geographic oriented results, optionally above a background map.
© 2013 TNO
70
RISKCURVES
So what’s GIS anyway? GIS is the abbreviation of "Geographical Information Systems". GIS is everything about objects that are linked to their geographical location and linked to extra information that can be displayed in a map theme. For example: a database may contain the location of houses (X,Y coordinate) while another database contains extra information like the price of this house, and the material that it is built from. GIS brings it all together by creating a map that shows location of the object (the house) and showing a colored price range. For example houses below are displayed as the green dots while houses above are displayed in red. A map consists of one or more layers. When using high end GIS systems, people have many separate map layers available that contain roads, pipes, waterways, underground electricity cables, glass fiber cables for computer communication, the water supply system, street plans, terrain topography and much more. Combining two or more of these layers result in a specific map theme. A map theme can, for example, consist of a layer that contains the terrain topography and a layer that contains the waterways. This will give information about the flow of rivers and canals with respect to terrain topography. When contractors start digging, they mostly make use of maps that contain the terrain topography combined with pipelines, water supply system, computer communication cables etc. to avoid damage while digging. RISKCURVES works more or less in the same way as a common GIS system. It can make use of several map layers and you can manipulate these layers and the properties of every layer.
© 2013 TNO
The user interface in detail
71
The sequence in which the map layers are drawn, greatly influences the way that the map looks. For example if the first layer contains the road network and the second layer is a bitmap containing the terrain topography, the road network might be invisible because it is “hidden” behind the bitmap. This is because vector files (like the road map) are mostly transparent, while bitmaps are opaque (not transparent). Keep this in mind when you create a map of several layers. By default, background maps are always the first (back most) layer. The order of the layers can be manipulated by dragging a layer in the legend panel towards a different location. Currently supported formats for background maps are: · SHP (Shape) file format, these files typically contain vectorised information like lines, polygons or points, associated with fields (stored in a seperate database), that can be used as an indicator for the geographic object. · DXF is a cad oriented exchange format. Most CAD programs can export drawings (e.g. overview of an industrial plant) as DXF file. · Pixel oriented images: JPG, TIF, PNG, BMP files. Pixel oriented files always need a geo reference file that contains the translation of pixel coordinates to real world coordinates. A utility to create these files is included in the program.
3.15.1 Presenting geographic calculation results If a model calculation output also contains geographic oriented information (e.g. size of the toxic cloud, size of a fireball) these result will automatically be presented in the RISKCURVES system. You will often add one or more and map backgrounds (topographic maps, Google earth screen capture) to the map presentation area. RISKCURVES will automatically add all layers that contain the results. Note that the layers can be activated / deactivated by the checkboxes in the legend. Below is a picture that shows a specific GIS layers on top of a topographic map.
© 2013 TNO
72
RISKCURVES
Contour Legend and colors In the legend area all map layers will become visible as an checkable item. The colors that are used for Iso Risk Contours, or grid presentations, can be modified within "presentation settings" For a comparison set, multiple contours will be drawn of one particular contour value (e.g. the 10-6 contour) will be drawn. The value for this "multiselect contour" can be modified within presentation settings.
© 2013 TNO
The user interface in detail
73
3.15.2 Positioning equipment Positioning equipment, or defining routes or population polygons The background map will become very useful when defining equipment: a stationary equipment requires a coordinate. This can be taken from the map by using the popup menu: < Right Mouse> < Set release point>. This will automatically copy the current mouse cursor position to release coordinates. It is strongly advised to zoom in the map in order to position equipment accurately. For a transport equipment, a route needs to be defined, which is performed by simply pointing/clicking points on the route. Pressing the button will invoke the drawing mode: add or move point by simply dragging a point (move) or clicking a new point on the map. A selected point will be drawn in red. Use the scroll wheel to zoom in to a high level and be able to define points accurately. Deleting points can be done by selecting a point on the map and pressing or selecting the table row and pressing . Finalize definition of a route by pressing < End Edit> in the route input panel.
A route also has correction factors, which are used for local adjustment of the failure frequency, for example on railroad crossings, switches etc.
3.15.3 Map functionality The map view uses the same mouse shortcuts as the graph display for zooming and scrolling. Apart from the left right mouse drag, the scroll-wheel can also be used to zoom in or out of an area. Whenever the contour panel is activated, seven toolbar buttons can be used for specific features:
The grid tool shows a grid definition in the map, which can be handy for reading out positions
© 2013 TNO
74
RISKCURVES
Will show the location of equipment locations and analysis points. This is an on/off toggle button. Any model that has a geographic presentation in the contour viewer, will have a "Release X" and "Release Y" parameter. This location can be illustrated with a dot on the contour image. Transport routes and analysis points will also be displayed. Show crosshair: Illustrates the coordinate of the location at the cursor, and if a "grid" layer is active (selected in the legend) , provides information about the value of the location at the cursor Geo-reference the background image. Will invoke a screen that can be used to georeference a pixel based background. Note that a pixel oriented graph needs to have a definition for the size and positioning of the image in real world coordinates. For this purpose, EFFECTS uses the ESRI standard georeference method which requires a wordfile definition for every image. Currently supported pixel formats are JPG, TIFF, BMP and PNG files. Ruler: activates the contour ruler, which is a measuring device to be used for obtaining absolute sizes of clouds, areas, or distances to objects on a background map. Transect: provides the possibility to determine the individual risk along a line section. By defining a line (click and drag the cursor to define a track) the transect panel, which is located below the legend panel, will display the risk along this track. Lock zoom: this toggle button can be used to force the map view to keep the same field of view on every component. Full extent: rescales the map to the full extent (all objects visible)
Furthermore, a popup menu is available within the map view:
© 2013 TNO
The user interface in detail
75
Set release point: By selecting this option, the coordinate which is currently under the mouse cursor, will be entered in the "X coordinate/ Y coordinate" equipment input fields. Add analysis point: By selecting this option, the coordinate which is currently under the mouse cursor, will be added as an analysis point. This function is only available on calculation sets and cumulation sets
Gray scale background: Is only applicable for pixel background images, and will set the current background in grayscale to improve visibility of the (colored) contours. Show GridValues: Illustrates the coordinate, and if a Grid layer is active, the value of the cell under the cursor will be presented:
Show grid: activates the grid definition in the map, which can be handy for reading out positions Show Release point: activates the equipment locations layer, showing the locations of equipment locations and analysis points. . Draw transect: point and drag a line to display the risk values along this line as a risk transect. Edit transect point: provides the possibility to manually adjust coordinates, to obtain exactly the same values for strt and end point of the transect. Georeference, Georeference point and Georeference expert:
© 2013 TNO
76
RISKCURVES
These buttons are only available when a background map component is active. Basically, the geo referencing of a pixel map consists of two aspects: the size translation, and the absolute coordinate's. Although it is very often feasible to use a (country specific) absolute referencing system, maps can also be used relatively, by simply identifying the main source location as (0,0). By using this tool, any point (zoom in to get an accurate positioning) a selected point can be associated with a specific coordinate. Note that this action will alter the world-file (see georeferencing), used to relate the background pixel map to real world coordinates
3.16
Report panel The report panel contains a full list of input AND result parameters that were generated while running the calculation. Depending on the active node of the tree, the contents of the report will differ. For a calculation set or cumulation set, a list of all scenarios, including frequency and maimum effcet distances, and societal risk ranking report will be presented.
© 2013 TNO
The user interface in detail
77
For a scenario including a consequence calculation, the full list of the EFFECTS model calculation results is available. Note that for combined model scenarios, this list may be very long, because a model chain might contain several submodels, such as outflow, evaporation and toxic/explosive dispersion models, and several typical fire/heat radiation phenomena models like a Bleve or Poolfire model. On a modelset level the result will be presented for all different weather conditions defined.
© 2013 TNO
78
RISKCURVES
If the user has selected multiple models i the model navigation list, the report will display results of selected sessions in different columns. Any differences in input will be marked bold, allowing to quickly compare calculations and see differences in input. The report view is a full HTML document, which can easily be copied to your local office application. Furthermore print and print preview is supported by the internal HTML viewer.
3.17
Model log panel The model log presents a logging of all messages occurring during a calculation. Furthermore, a color code is used to illustrate warnings and error messages. When calculating complex physical effects or consequences, there are numerous reasons why a calculation can go wrong. Physical conditions might not match those required to run the selected model, erroneous input might be entered or simply a bug in a model encountered. When the program traps an error, this error is send to the model log. When the model log contains warnings or errors after a calculation the user will get a notification that something unexpected has happened and the log viewer is opened. The severity of the error is illustrated by the led light on top of the view:
© 2013 TNO
The user interface in detail
79
A Yellow indicator illustrates a warning, for example a model was used outside it validity domain, root finding method does not find a solution, or system messages that a subroutine was doing an illegal action but which could be corrected. It is not an error but needs user judgment. Warning messages may include hints that no societal risk has been found. A Red indicates an error, which can be straightforward messages like "Can not calculate because of parameters being empty", but it may also report that input conditions determine a situation for which the model will not run or a scenario that was skipped.. Basically an error messages implies that no reliable (end)results are available for the corresponding scenario or consequence model. The message is often combined with a suggestion of how to solve the problem. Note: the log window supports sorting on columns, so by sorting on severity, the most important warnings or errors may be listed on top. Note: Every message will only show once. Furthermore, if a warning or error is raised, the program will always switch to the log viewer to force the user to read the messages.
The Time column illustrates the date or time when the error occurred. The # field presents the number of times that the problem (error or warning) occurred.
© 2013 TNO
80
RISKCURVES
The first error that happened is usually the most important. If errors were trapped in more than one model calculation, the first/second error of every session is usually the most important. The MODEL LOG log is cleared every time you perform another calculation, and is always associates with ONE model calculation. Since combined model chains consists of several submodels, warnings may be associated with multiple models. Model codes used in the log: The combined model often incorporates 4 types of dispersion models which will be abbreviated in the messages: HGDE: Heavy Gas Dispersion Explosive mass model, (Inst indicated Instantaneous mode, Pool indicates Poolevaporation mode) HGDT: Heavy Gas Dispersion Toxic model NGDE: Neutral Gas Dispersion Explosive mass model NGDT: Neutral Gas Dispersion Toxic model
3.18
Legend panel The legend panel, displayed on the right side of the map, will display all layers currently active in the internal GIS viewer. Activation or deactivation can be performed by selecting the menu option "View" "Panels" "Legend" / "Transect graph" Specific presentation layers can be activated or de-activated by selecting the checkbox of the layer in the legend.
© 2013 TNO
The user interface in detail
81
- The ordering of the layers can be altered by dragging the layer legend item to the top or bottom direction. The topmost layer will be the first layer to be drawn, subsequent other layers will be projected on top of the preceding layer. For this reason, the background maps should always be the top layer. - The bottom part of the legend panel can be used to present a risk transect: a XY graph presentation of the risk along a definable track on the map. Double clicking a legend item will display the internal editor for the GIS viewer. This editor can be used to to change display colors (note that default colors should be set in presentation settings), transparency of the layers etc.
© 2013 TNO
82
RISKCURVES
© 2013 TNO
Advanced features
4
Advanced features
4.1
Options
83
In the edit menu a menu-item Options is provided, selection if this item will show the options screen.
Reload last file If this item is checked, RISKCURVES will always open the last used file, allowing you to quickly continue with the project you were working on. Number of files to remember The file menu contains a list of the recently used RISKCURVES project files. The number of files stored in this history list can be modified with this value. Restore application position By selecting this item, it is possible to restore the size and position of the RISKCURVES application window. The program will always open in the same size and same relative screen position that was last used. Restore toolbar positions If the user has modified the position of the toolbars (they can be dragged by using the grip on the left side of the toolbar), using this setting will repaint the toolbar on the same place when RISKCURVES is re-activated.
© 2013 TNO
84
RISKCURVES
Default chemical database By default, RISKCURVES comes with two databases: the YAWS database, providing physical properties for about 120 chemicals and the the extended DIPPR database, containing properties for over 1500 chemicals. Furthermore, users can make copies of the database, where specific properties are modified, or chemicals gave been added. This setting can be used to determine the default database that will be used whenever a new model is created. Note that is is always possible to switch databases for any effects model by using the browse button to the right of the chemical combobox:
See Chemical database for more information on the chemical database Default Display Units RISKCURVES is equipped with a automated unit conversion system. All units can be switched by simply right-clicking the unit label and selecting an alternative unit. Apart from this option, users can define their own default set.
Default Presentation settings These settings contain typical parameters influencing the graphical presentations of RISKCURVES. All parameters can be individually adjusted in the presentation settings within the project. For any new project, the defaults defined here will be applied. Expert Parameter Defaults Any effects model contains several input parameters, where the current complexity level determines the number of parameters displayed. So-called "expert parameters" are more dedicated parameters, illustrated in a yellow background, that usually don't need adjustment. The defaults that will be used if these expert parameters have not been explicitly defined, will be taken from the "expert parameter defaults" that can be adjusted here. Environment settings These settings store some typical environmental parameters that are used within the program, if the user is not working in Expert mode. Furthermore, whenever the button is pressed in expert mode; the values provide here will be entered in the corresponding property fields. Expert parameter settings Apart from environment dependent values like ambient temperature etc, some other system parameters are often used within the models. Values that re used within the models can be modified in the parameter settings dialog.
© 2013 TNO
Advanced features
85
Dispersion Sigma definitions The neutral gas dispersion model uses so-called Sigma's to the Gaussian dispersion. For every Pasquill stability class these sigma's have been defined in the Yellow book. Although highly discouraged, it is possible to change these settings, because other countries may have different values for these classes. WARNING: CHANGING THESE VALUES WILL INFLUENCE THE RESULTS OF THE NEUTRAL GAS DISPERSION MODEL !
Location of User modified settings Note that any RISKCURVES project will contain all parameters used INSIDE the project itself. Even transferring the project to another system will not change results upon calculation, WITH EXCEPTION of the chemical database. Note that this file (see chemical database for the location of this file) needs to be copied to another system to be using identical "User defined" substances.. All defaults, entered within the options menu, including user defined new meteorological definitions,(specific meteo station data) will be stored in a user settings file, .user.config (which is a XML file) which is located inside the users application folder. The typical location depends on the windows version, but it can be found using the %Appdata% query in the Windows Find box or Windows Explorer address field. The %appdata% folder contains subfolders EFFECTS and RISKCURVES, containing the dedicated configuration files. The typical locations for user configuration files can also be found using the "RISKCURVES diagnostics" tool, which is installed along with RISKCURVES.
4.2
Display units EFFECTS 8 is equipped with an automated unit conversion system.
© 2013 TNO
86
RISKCURVES
All units can be switched by simply right-clicking the unit label and selecting an alternative unit. The input screen offers a convenient way to use any unit you prefer for entering the data. Right click on the unit and select the required unit. Note that chemical dependent parameters (such as Lower Explosion Limit or LEL value) will perform the required mass/volume translations automatically:
Apart from this build-in unit conversion option, users can define their own default Unit set. If one prefers to use British Standard Units or other local (non-SI) units, it is possible to define this as the standard unit for all models and all axis. After saving of these settings, all screens will be using the specified units.
© 2013 TNO
Advanced features
4.3
87
Presentation settings These settings contain parameters determining presentations of RISKCURVES. Presentation settings will always be applied for the complete project, and not just for one Calculation set. The defaults can be accessed from the options menu, users can redefine or adapt presentation settings within any project. Important choices are the typical values for Iso Risk Contours to draw, the Individual Risk level to use in comparison graphs, and definition of the guide value line.
Guide value definition RISKCURVES now supports the use of country specific guide value definitions. Note that the guideline is also presented in the FN graph as a straight line. The slope, and orientation of this threshold line can be defined by using the parameters: 1. Guide value starting at # victims: the X-axis starting point 2. Guide value starting at frequency: the Y-axis starting point 3. Guide value transport starting at: the Y axis value to use for a transport FN curves 4. Guide value maximum: the X axis end point
© 2013 TNO
88
RISKCURVES
5. Guide value slope: the slope of the guide value line (e.g. Netherlands uses slope 2, implying a risk aversion, the UK uses a slope of 1; no risk aversion)
Color legend translation of grids Furthermore, the color legend to use for specific grid presentations can be defined here. The concept is to define minimum and maximum levels, the number of levels and color range to illustrate a (risk) value into a color.
Note that the translation of societal risk area maps is based on the norm ratio of the societal risk at a location, which is compared to the guide value. These norm ratio values are colored according to logarithmic scale: -2 means 100 times lower than the guide value, =2 means 100 time too high, and zero means ratio 1.
Line color palette The color palette itself is used for the definition of line colors for contours and graphical presentations. The first 6 values are used for contour colors, the colors from 7 to 20 are used for subsequent lines in the graph panel. The last two colors are used as color for the "show locations" and "show analysis points" color.
© 2013 TNO
Advanced features
4.4
89
Expert Parameter settings The expert parameters editor is used to store some of the default values which will be entered into a model calculation when working in "simple mode" or when the button will be pressed.
StandardPipeRoughness The roughness of a pipe is used in pipe flow pressure drop calculations, default 4.5E-5 m Hole contraction coefficient This contraction coefficient for sharp edges is used in outflow calculations. Default is 0.62 Pipe contraction coefficient This contraction coefficient for pipe endings edges is used in outflow calculations. Default is 0.82 Concentrating averaging time toxics This value is used to calculate an time averaged concentration for toxic loads. Default is 600.0 sec. For a (semi-) continuous source this is the duration over which the concentration will be ‘averaged out’, to deal with the effect of the meandering of the wind.
© 2013 TNO
90
RISKCURVES
The averaging time for toxic concentration is related to aspects of the receiver. For local irritant chemicals the effects can occur within few seconds (few breathings) and for systematically irritant chemicals within few minutes (few times pumping of blood through body). Therefore the standard value is chosen to be 60. Concentrating averaging time flammables This value is used to calculate an time averaged concentration for flammable substances. Default is 20.0 sec. For a (semi-) continuous source this is the duration over which the concentration will be ‘averaged out’, to deal with the effect of the meandering of the wind. The minimum value for the averaging time is 18.75 s [Yellow Book], this compares to the value for an instantaneous source, which is also used for the calculations of the contour for the flammability limits and the explosive mass. Toxic Inhalation Heigth This values is used as default height to calculate the toxic dose. Fraction confined mass in Multi energy explosion method The multi energy method for explosions has an important parameter "Fraction confined mass". Default this one is set to 8.0 %. Although this value is quite unrealistic, it appears to give answers comparable to the old TNT method. CurveNumber for Multi energy explosion method The multi energy method for explosions has an important parameter "CurveNumber". Default this one is set to 10. Although this value is quiet unrealistic, in combination with 8% confined, answers are in the same order of magnitude as the old TNT method. The multi-energy method is based upon experimental graphs in which the required value depends upon the distance from the vessel and the type of explosion. 10 different types of explosion are considered, and have a curve associated to them. Those are: 1. Very weak deflagration 2. Very weak deflagration 3. Weak deflagration 4. Weak deflagration 5. Medium deflagration 6. Strong deflagration 7. Strong deflagration 8. Very strong deflagration 9. Very strong deflagration 10. Detonation
© 2013 TNO
Advanced features
91
Probabilty FlashAndExplosion In a gas cloud explosion, the flashfire may be accompanied by overpressure effects. This parameter determines the probability that flash AND explosion occur. Default is 0.4 Default mixingheight Used in dispersion calculations. Default value 500.0 m
4.5
Meteorological distribution A meteorological distribution contains the probabilities of weather conditions and wind directions occurring. Definitions are usually named after a corresponding meteorological station, like airport names. All Dutch weather stations are available within the standard installation, but users will often need to define their own stations. Once station have been added, these locations can be selected in the meteo data project node. This editor shows a table which present the probabilities that wind for a specified stability class from a wind sector occurs. Different from previous version, the new editor now has separate Day/Night columns, and values are relative percentages. The sum of all day and all night definition should be 100% together. If this condition is not archived, a red Meteo Distribution indication will be shown, and the total day or total night might turn red, indicating invalid values.
User can define simple data sets with only two weather classes. The button < Add Weather Class> will add two (a day and night)columns, titles according to the selected Pasquill class and wind speed.
© 2013 TNO
92
RISKCURVES
Columns can be deleted by pressing < RightMouse> button and selecting button. Apart from modifying the cells for the specific wind direction weather class combination, the total occurring probability for a column can be modified. This will remain the current wind direction distribution for that weather class untouched and can be convenient when adjusting probabilities of weather class occurring. Note: In previous RISKCURVES versions the distribution daytime/nightime was incorporated within the meteorological definition, the current distribution is based on 100% daytime total and 100% nighttime total. The daytime / nighttime ratio is defined in a separate parameter Meteorological Daytime Fraction
4.6
Vulnerability settings Within the vulnerability settings, typical parameters defining translation of effects to damage are grouped together. The defaults to be used in a new project can be accessed from the options menu, users can redefine or adapt vulnerability settings within any project. See paragraph QRA Definitions Vulnerability parameters for a detailed description on all values.
© 2013 TNO
Advanced features
4.7
Environment settings This editor provides the possibility to modify environment settings as displayed below:
© 2013 TNO
93
94
RISKCURVES
These environment are being used as standard values for model definitions, and can only be overruled when using the "expert mode" situation. Since RISKCURVES now supports dedicated day/night calculations, environment parameters have distinguished day or night specification. Whenever a new sceanrio will be created, the appropriate day/night conditions will be pushed in models for D5 day or D5 night. With respect to the risk calculations, the parameter "Meteorological Daytime Fraction" is important. This parameter defines the number of hours during 24 hour that are defined as daytime situation. Note that separate D5 day and D5 Night calculations will be performed for two situations (as for any stability class occurring during day and nighttime). This parameter will adjust the typical occurrence of day/night situation according to the countries meteorological condition. Apart from this "calculation set" typical parameter, any scenario can be defined as occurring more or less during daytime. See Environment parameters for a detailed description of all parameters.
4.8
Accuracy settings Accuracy settings are used to group parameters that influence the accuracy of the calculation. One should be aware that there might be a tradeoff between accuracy and calculation speed. The defaults to be used in a new project can be accessed from the options menu, users can redefine or adapt accuracy settings within any calculation set.
© 2013 TNO
Advanced features
4.9
95
Chemical Databases By default, EFFECTS comes with the YAWS database, providing physical properties for about 100 chemicals. The extended database contains over 2000 chemicals. Database sources: YAWS , DIPPR or USER defined Because the user can alter the databases, there is a potential danger that a model might crash or calculates false results due to erroneous data that was entered in any database. To avoid a situation that a model will not run due to an erroneous database, database always contains the original data. These databases are called “YAWS” and "DIPPR". As of version 8.1 both datasets are stored within the same physical file, the availability of a DIPPR license will provide access to the DIPPR chemicals as well.Without an active license, DIPPR chemcials cannot be accessed. The database editor is configured so that you can not change the YAWS or DIPPR data, but you can copy values from either database into your own "USER set". The following paragraphs will explain how you can create your own database and maintain it using the database editor. Physical and thermodynamic properties of all (available) chemicals come from the chemical database. The standard YAWS database comes with more than 100 chemicals, see list of chemicals. The DIPPR database contains more than 2000 chemicals and is based on the DIPPR® 2010 chemical database. Please note: that if you have a license for both databases, calculation results from YAWS may differ from DIPPR generated results, simply because the chemical properties may differ. Furthermore, the DIPPR database takes into account the "non ideal gas behavior" by using a second virial coefficient. Due to this, results from gas outflow models may differ more than expected based on typical density parameters.
© 2013 TNO
96
RISKCURVES
Location: The database itself is stored in a file "Chemicals.tci" stored in the windows % ProgramData% folder under \TNO\). This implies that EFFECTS and RISKCURVES might be using different databases version. Synchronizing User defined chemicals: If users have added "User" chemicals, and colleagues on other PC's want to use the same definitions, the "chemicals.tci" file will need to be copied to the other computer! In future versions, with potential new versions of the DIPPR database, a conversion tool will be provided to transfer user defined chemicals into the new database.
4.9.1
Chemical database editor Using another chemical database set When the program starts, it will always connect to the standard database automatically. You can use chemicals from the chemical database sets (YAWS, DIPPR or USER DATA set). To use another database set click on the browse button next to the chemical name. This will open the chemical database dialog, where you can change the database set in the top left corner. Select a chemical from that database and click 'Ok' to use that chemical. Warning! Because you can switch between database sets between models and sessions, there is a potential danger of creating inconsistent calculation results. Therefore the program will always print the name of the database that was used at the time the calculation was performed. If you suspect a situation as described here, check the database names! The database editor performs no checks. If you enter physically incorrect data, the program will trap many errors in a model calculation that might be very difficult to trace.
DIPPR: Non ideal gas behaviour Because the DIPPR database contains a "Second Virial Coefficient", which describes the non ideal (compressibility) behaviour of a chemical, results of calculation with the DIPPR databases may differ from YAWS based calculation, especially in the case of gas release calculations. Viewing the chemical database This editor shows the contents of the database. The chemical database editor provides a tree view, where chemicals are depicted by a tree branch in the left-hand panel. Selecting a branch opens the chemical, where each chemical has constant, temperature dependent, and possibly toxic properties.
© 2013 TNO
Advanced features
4.9.2
97
Selecting a chemical from the database To select a chemical, use the search field in the top-left of the database editor menu. This name field will act as a filter on the chemical names, and tree resulting tree will include all chemicals matching the entered characters.
© 2013 TNO
98
RISKCURVES
Select the chemical which is required in the tree, and its basic properties will be illustrated in the right panel. Properties are divided in "Constant" "Temperature dependent" and "Toxic properties". Note that the button will automatically jump to the International Chemical Safety Card of the current chemical, if the corresponding ICSC number is known. This site will provide detailed safety information, including commonly used occupational exposure limits values like IDLH values. Select the sub-branch you wish to inspect from the chemical in the chemical tree.
© 2013 TNO
Advanced features
4.9.3
99
Viewing/Editing properties of chemicals Before you are going to view or even edit the properties of a chemical, you must select the database that contains the chemical of which you want to change or view properties. NOTE: ONLY CHEMICALS CAN BE MODIFIED! Because the program will use the standard YAWS or DIPPR database by default, it will show non-editable fields for all provided data. To modify the data, a copy of the record must be made, which will be put in the "USER DATA" set. This can conveniently be done by right-clicking on a chemical in the chemical tree, and selecting "Copy ". This will create a full copy of an existing chemical in the USER DATA set. Note that copies of original DIPPR chemicals stay subject to your DIPPR license.
© 2013 TNO
100
RISKCURVES
Viewing constant / temperature dependent properties The chemical tree list is divided into a 3 sub-branches. The first is called “Constant Properties” and contains all temperature-independent properties like molar mass, critical temperature etc. All temperature-dependent properties are listed under the other chemical substance branch. The names are self-explanatory. Click on the required parameter (such as "Liquid vapour pressure") to illustrate the formula's and graphs for the property. The checkboxes in the 'Known Ranges' view control the graphs that are displayed.
© 2013 TNO
Advanced features
101
Viewing graphs of the chemical parameters The graph which illustrates the temperature dependent behaviour of the selected chemical parameter provides some of the typical graphical features: zooming by dragging a box, using a crosshair to read out values, or even change scales:
Viewing Toxic parameters The toxicity parameters view (last branch) contains all values for the "A", "B" and "N" probits. Note: that the values for A and B are dependent of the units used ! By using the right mouse button on top of the unit box, the unit values can be changed, correcting the a and b values
© 2013 TNO
102
RISKCURVES
Editing temperature-dependent properties To edit a temperature-dependent property select it's entry in the left-hand tree. If the temperature-dependent property you wish to edit is not in the list, right-click on the 'Temperature Dependent Properties'-entry, and select it from the 'Add' menu.
Minimum valid and maximum valid temperatures (Ranges) In the right-hand pane you can see the different ranges known for the selected property. Right click in the 'Known Ranges' area to add a new range, or delete a current one. Double click on a temperature to edit it's value. As any temperature dependent property is a formula, the ranges indicate between which temperatures a function is indicated as "valid". The word "valid" can have two meanings within this concept. First: most temperature dependent parameters are interpolated or extrapolated from experimental data and the two temperatures shown here are the temperatures between which experiments have been performed and the data is validated. This does not necessarily cover the whole range that the function could be valid. For example, theoretically, a vapor pressure is valid until the critical temperature. Above that temperature, a liquid-vapor equilibrium does not exist. So the maximum valid temperature can be lower than the critical temperature when experiments have not been carried out until the critical temperature. Second: As explained above, the maximum valid temperature can also be a true physical limit, like a melting point or a critical temperature. These values were entered in the previous "General" data entry screen.
© 2013 TNO
Advanced features
103
Enabled: In the DIPPR database, temperature range overlapping properties have been disabled. To be able to activate another range, you can toggle this setting by double-clicking on the field.
Equation This is the actual equation that is used to calculate the temperature dependent data. You can change the 'Equation' drop-down list to the desired equation, in the 'Parameters' area to it's left, you can double click to edit the parameters for the chosen equation. Editing toxic properties To edit toxic properties, select the 'Toxic Properties' node in the left-hand tree view. Existing properties can be edited by double-clicking their value. New properties can be added by selecting them in the right-click menu. With every toxic property, it's possible to store the source, and a comment.
Editing constant properties and chemical identification information Editing the constant properties, and the chemical identification information is much the same as editing toxic properties.
© 2013 TNO
104
RISKCURVES
Every chemical has a source and a comment stored with them, editable on the chemical identification page (select chemical name in the left-hand tree). Existing values can be changed by double-clicking, new entries can be added with the right-click popup menu. Create a new chemical A new chemical can be created by selecting the user database, and right-clicking in the lefthand tree area. Select 'New Chemical' from the popup menu. Properties can be added and edited as described above. Since only user-chemicals can be edited, it is possible to copy a chemical from another database into the user database. To do this, select the database in the drop-down menu, and search for the chemical. In the left-hand tree, right click on the chemical name and select 'Copy chemical'. The editor will now switch to the newly created user chemical.
4.9.4
Chemicals convertor Users of the previous Riskcurves version 7 may have put considerable effort in defining their own chemicals or adding properties to existing chemicals. Since the chemical database format has been changed, these "User defined" records need to be translated into the new database. For this purpose, a chemical convertor is provided. The convertor can be started form the menu "Tools" ... "Chemicals Convertor". Use the "Open file" button to brows to the database file containing the required chemical records. This file is usually located in a "shared data" folder under the EFFECTS or RISKCURVES program files directory and has extension "RDB".
© 2013 TNO
Advanced features
105
Upon opening this file the complete list of chemical will be presented in the chemicals list of the left of the screen. A multiple selection is possible. Press the button in the center to import the selected chemicals into the new user database.
4.10
Mass and volume calculator In some cases a model asks for a volume while you only have a mass available, in other cases it is the other way around. The volume and mass calculator converts masses into volumes and vice versa for both gases and liquids. It also supports unit conversions. This calculator can be accessed from the "View" menu.
© 2013 TNO
106
RISKCURVES
The calculation is invoked as soon as something is typed or changed in the “Mass of …” or “Volume of …” boxes. To change from mass conversion to volume conversion simply switch to the other edit box. NOTE! Depending whether the chosen pressure and temperature fall within legal physical limits, a result might be visible or not. For example if you choose a temperature above the critical temperature, you will see no answer for a liquid mass/volume conversion. Thus, in some cases you might discover that there is no result. In most cases this means that you are trying to perform a conversion outside the physical limits (e.g. above the critical temperature, no liquid phase exists, hence no mass of a liquid can be calculated).
4.11
Mortality/probit calculator The mortality calculator (or Response calculator) is used for quick estimations of the fraction of mortality when exposed to a concentration of a toxic chemical. It also supports unit conversions between three different dose units. The calculator can be accessed from the "View" menu.
© 2013 TNO
Advanced features
107
Dependent upon the units of concentration and the molecular mass of the chemical that the probits are to be calculated for, the a value of the probit a will differ. The probit calculator can convert the value of the toxic probit as calculated for each of the three commonly- used units for concentration: kg/m3, ppm or mg/m3. Simply right click on the unit of the probits, and it will display other possible units, and convert the current displayed value.
4.12
Geo-referencing images Vector data in Map Layers exists in a real-world or map coordinate system, measured, typically in meters. The x-coordinates increase from left to right and the y-coordinates increase from the bottom to the top. This is quite different from a raster image represented by a Pixel Oriented Layer such as bitmaps and JPG files. A bitmap is a raster image that is organized and measured by rows and columns. Each cell has a row number and a column number. If the origin is located in the upper left corner of the data, that cell would be identified as row 1, column 1. Pixel based Image layers do not have a coordinate system. This means that the mapping system assumes the lower left corner of the map as coordinate (0,0) and the upper right coordinate as (x,y) where x and y are the number of pixels of the bitmap in the x and y direction. For Vector Layers and Pixel Layers to be displayed simultaneously, the rows and columns of the image must be mapped into the x,y plane of a map coordinate system. An image-to-world transformation that converts the image coordinates to map coordinates must be established. Some image formats store geo-referencing information in the file header of the image or, in the case of images that do not contain this geo-referencing information, facilities exist in other products available from ESRI, for creating a file that contains the necessary transformation parameters. The file that contains the transformation parameters is called a world file. The world file always takes precedence over any header information. The utility supports five different pixel format files: TIF, PNG, JPG, GIF and BMP files. The corresponding world reference files have the extension TFW, PGW, JGW, GFW and BPW.
© 2013 TNO
108
RISKCURVES
NOTE: the image files and corresponding world-files need to be kept together in the same directory!! The user has two different options to provide the information to be able to translate the pixel information: Method A Use a georeference ruler to define a distance and define one georeference point. By using the georeference ruler, the user can draw a reference line (a ruler) anywhere on the map. This reference line should be a location on the map with a known distance. In Google Earth screen dumps, this may be the distance identification in the lower left corner. Use the button to activate the ruler. After pinpointing the distance with the mouse, a dialog will open that can be used to define the corresponding "real world length" of the pixel based background. Note that it might be convenient to zoom in on the area were a distance is known. The end point of the ruler can be dragged with the mouse to define a known distance. The georeference length dialog can be used to define the corresponding conversion destination dimension (=real world length) of the ruler. Note that is is possible to modify the original dx (x distance) and dy (y distance) as selected with the ruler. This option assumes that the chart has an isometric X- and Y axis scaling: e.g. 10 pixels on the X-direction will have the same distance as 10 pixels in the Y direction. Press to save the reference file. After this, it is possible to define a reference coordinate: simply on the background image and select "Georeference point" to enter a known coordinate of a selected point.
© 2013 TNO
Advanced features
109
Method B, suing the "georeference expert" Define the boundaries of a background map: By using the "Georeference expert" option ( : popup menu Georeference expert) the user can define the Xstart, Xend, Ystart and Yend values (in meters) of the loaded background map. The selection coordinates and dimensions panel displays the currently used boundaries, which may be pixel boundaries. Note that whenever a pixel image is loaded without a corresponding world file, the default value of 1 pixel = 1 meter will be used, and the upper left corner will be used ass the 0,0 coordinate. The conversion destinations (bottom) panel can be used to define the required boundaries of the background map. After selecting the button, the adjusted reference file will be saved.
© 2013 TNO
110
4.13
RISKCURVES
Risk transects Risk transects are graphical presentations of the individual risk along a defined line on the map. Start defining a transect by selecting the "draw transect" toolbar button or the "draw transect" entry in the map popup menu. Simply drag on the map to draw a line and after defining the line, the corresponding transcet graph will be drawn in the right bottom window. Note that the transect graph can be hidden or shown, use the "View", "Transect graph" menu entry to activate/deactivate the view.
© 2013 TNO
Advanced features
111
Since the transect graph is rather small in its default position, this panel can be dragged and resized as a floating panel. All graphic functionality for graphs is also available for the transect graph panel.
© 2013 TNO
112
RISKCURVES
© 2013 TNO
Technical backgrounds
5
Technical backgrounds
5.1
QRA Definitions
5.1.1
Calculation Set
113
A calculation set is a combination of system setting, a meteorological definition, population and accident (Loss of Containment) scenarios definitions for which Individual Risk and Societal Risk are being calculated.
A calculation set will have results in terms of Individual Risk Contours and Societal Risk Graphs and Societal Risk Maps. A calculation set is a typical input definition for a single QRA calculation: it contains all input that influencing the result. Since users often want to compare the change in risk due a modification (of population, scenarios), RISKCURVES can contain multiple Calculations Sets in one project (and thus file). A calculation set always contains the sub nodes Calculation settings, a Meteo data node, Population (if societal risk calculation is required), stationary equipment and transport equipment, because these contents together determine the result of a calculation.
5.1.2
Calculation Settings Calculation settings is a typical collector or grouping node.
It doesn’t have its own parameters, but combines several groups of parameters, to be applied to all input contained in a calculation set. Typical parameters are “Accuracy” describing parameters influencing calculation accuracy and speed, “Vulnerability” settings describing the relation between physical phenomena and damage (lethality), and “Environment” parameters, describing ambient temperature, humidity, solar radiation etc. for the typical location.
5.1.3
Accuracy parameters Accuracy settings contain parameters that influence the accuracy of the calculation, and very often also calculation speed, since speed and accuracy are somehow connected.
© 2013 TNO
114
RISKCURVES
Lowest significant frequency This is the lowest frequency that will be taken into consideration. If the frequency of an event is lower than this value, it will be neglected.
Cell size for risk grid This parameter defines the grid resolution of the contour map. Note that contours will be calculated based on a Iso Risk grid. By default a cell size of 10 meter is used. However if the range of the affected area becomes very big (routes of many kilometers, toxic events with effect areas upon many kilometers) this might lead to enormous grid (> 10^6 cells). In those situations a re-sampling will be performed avoiding large memory usage. As of version 9.012, this parameter is also used to influence the accuracy of the societal risk calculation: After the consequence calculation, the resulting lethality footprint is translated into a "societal risk response" grid to be able to superimpose (a wind direction and stability class dependent) lethality on the population. The cell resolution of these grids can be adjusted with the provided value. However, for large effect phenomena, this would lead to huge memory loads, because all weather classes and every potential wind-direction has its own list of or "affected" cells. For that reason, scenario's that would use more than 100*100 cells as its "response footprint" will be forced to use an increased cell size. Furthermore, scenario’s with shorter maximum effect distances (less than 100 mtr) will always be calculated at the standard accuracy of 10 mtr cells. The usage of this relatively small "response grid" size ensures that even when using a population distribution in much bigger cells (50 mtr), an accurate estimation of the number of lethal victims is achieved. (e.g. in case of a partial overlap of the lethal footprint with the population grid) Number subsectors FN calculation By default RISKCURVES defines 12 wind directions. This implies however that RISKCURVES can “miss” some population concentrations when it performs societal risk calculations.
© 2013 TNO
Technical backgrounds
115
This is shown in the left drawing. Two clouds for a 30° wind direction (=360/12) miss the important object resulting in risk underestimation. In the right drawing, 15° wind direction (=24 wind directions) were chosen resulting in a hit of the object. The value in this field defaults to 9 (108 wind directions).
Number of subsectors for FX calculation About the same explanation as for FN calculations holds for individual risk calculations. Again, a hypothetical person could be “missed”, especially where he stands far away from the cloud and the cloud tip is relatively small when compared with the sector circumference. In general when this factor is increased (values up to 20 are useful) the individual risk will decrease as Riskcurves will calculate more accurately the risk caused by overlap/underlap of a cloud compared with the sector width. The drawback is more calculation duration. More narrow clouds have a smaller risk than wide clouds which is obvious as wide clouds can overlap to adjacent sectors. This methodology takes it all into account at a value of 20.
Inter accident distance FX When calculating individual risk contours along a transport route, contours can become “caterpillar” shaped instead of a smooth line. This is caused when the inter-accident distance used to calculate individual risk is too large. The factor defaults to 50 mtr which means that possible accident points are located 50 meter from each other. When using scenario's with effect distances lower than 50 mtrs, is is advised to narrow the inter accident distance.
Inter accident distance FN This factor is used to influence the inter-accident distance during societal risk calculations. Consider the following example:
The above example illustrates a road transport. RISKCURVES will generate accident points and calculate the size of the gas clouds.
© 2013 TNO
116
RISKCURVES
In the left situation the accident points are separated too far and therefore the calculation misses an important object because the clouds can not reach it. In the right in situation, more accident points are generated and the object is hit. The factor defaults to 50 mtr which means that possible accident points are located 50 meter from each other. When using scenario's with maximum effect distances much lower than 50 mtrs, is is advised to narrow the inter accident distance. Maximum number of accident points To limit an enormous number of calculations when RISKCURVES want to generate accident points for long routes but small consequences, this is the maximum number of accident points it will use
5.1.4
Vulnerability parameters Vulnerability parameters define how a specific physical effect is translated towards damage.
For toxic materials, this is derived from their toxic probits which are stored in the chemical database, but for flame contact, heat radiation and overpressure, dedicated damage translation can be defined.. Lethal fraction flash fire This is the fraction of mortality used within the footprint of a flame envelope determined by a LEL footprint. Again, the height of the flammable cloud is not taken into account. Leave it to fraction 1 (100%) unless you have good reasons to change it. Lethal fraction flame contour This is the fraction of mortality used within the footprint of a poolfire or jetfire (no matter the height) Leave it to fraction 1 (100%) unless you have good reasons to change it. Heat radiation level total destruction This parameter defines the heat radiation level that will be associated with total destruction: 100% lethality inside and outside and is used in both individual risk FX and societal risk FN calculations Heat Radiation Exposure Duration This value determines the maximum duration of exposure to heat load, as used in consequence calculations. Default is set to: 20.0 seconds Protection factor clothing
© 2013 TNO
Technical backgrounds
117
The protection factor applied for clothing, used for societal risk calculations on heat radiation. A probit calculation will be applied on heat radiation, leading to a lethality. This lethality is corrected with this factor to obtain the damage in case of societal (protected) calculations Heat radiation damage probits By default, the vulnerability model (probit function) as described in the Green Book [4] has been used for the exposure to heat radiation:
with q = the heat radiation level in [W/m 2] and t = the exposure duration in [sec], which is assumed to be maximum 20 sec (defined by parameter max heat radiation exposure duration). The probit value is transferred to a fraction of mortality (0..1) afterwards. This implies a probit A of -36.38, Probit B = 2.56, and probit N = 4/3 Because some countries are accustomed to use other probits, these A, B and N values can be modified. The methodology described above is valid for individual and societal risk, but for inside population a protection of 100% is assumed, as long as the level is lower than the heat radiation total destruction level .
How to convert a probit to a fraction of mortality The probit value Pr as mentioned several times in the chapters before varies between 2 and 9. To convert the probit value to a percentage of mortality, the table below is used. The probit values are listed within the table itself. From the side and the top of the table, the percentage of mortality can be read. For example: A probit value of 4.01 (second row) corresponds with a value of 16% mortality.
%
0
1
2
3
4
5
6
7
8
9
0 10 20 30 40 50 60 70 80 90 99
3.72 4.16 4.48 4.75 5.00 5,25 5.52 5.84 6.28 7.33
2.67 3.77 4.19 4.50 4.77 5.03 5.28 5.55 5.88 6.34 7.37
2.95 3.82 4.23 4.53 4.80 5.05 5.31 5.58 5.92 6.41 7.41
3.12 3.87 4.26 4.56 4.92 5.08 5.33 5.61 5.95 6.48 7.46
3.25 3.92 4.29 4.59 4.85 5.10 5.36 5.64 5.99 6.55 7.51
3.36 3.96 4.33 4.61 4.87 5.13 5.39 5.67 6.04 6.64 7.58
3.45 4.01 4.36 4.64 4.90 5.15 5.41 5.71 6.08 6.75 7.65
3.52 4.05 4.39 4.67 4.92 5.18 5.44 5.74 6.13 6.88 7.75
3.59 4.08 4.42 4.69 4.95 5.20 5.47 5.77 6.18 7.05 7.88
3.66 4.12 4.45 4.72 4.97 5.23 5.50 5.81 6.23 7.33 8.09
© 2013 TNO
118
RISKCURVES
Pressure level total destruction This value is used to define the peak pressure level at which inside and outside lethality is assumed to be 100% (total destruction zone). Default value is 300 mBar (0.3 Bar) Lethal fraction pressure total destruction zone Defines the lethality within the total destruction pressure level zone. By default 100% (fraction 1) Pressure damage based on The translation to damage by overpressure can be defined by 1. Using two pressure levels: total destruction and inside (glas) fragments. 2. Using a probit based on Peak pressure: Pr = A + B * ln(PeakPressure^N) 3. Using a probit based on exposed pressure impulse Pr = A + B * ln(Pressure Impulse^N) In case 3, the pressure impulse is calculated as (0.5 * peakpressure * positive phase duration). Method 3 cannot be applied when using the TNT overpressure calculation, because that method does not provide a positive phase duration answer; one needs to use the Multi Energy method for method 3.
Peak pressure inside damage This pressure defines the minimum pressure level for inside damage. All areas with pressure between "total destruction" and "inside damage" levels, will be treated with the corresponding inside damage lethality level. The lethality fraction will only be applied in societal risk calculations, on inside population. The pressure level will also be used as a treshold level for pressure contours presented by TNT or Multi Energy models. Lethal fraction Pressure inside damage All areas with pressure between "total destruction" and "inside damage" levels, will be treated with this corresponding inside damage lethality level. The lethality fraction will only be applied in societal risk calculations, on inside population. Perform toxic indoors calculation The toxic exposure inside can be calculated based on the actual concentration time profile and ventilation rate. This calculation is invoked by selecting "Yes" in this setting. The calculation is performed inside the Dispersion Toxic dose models which will also present a Inside lethality grid (expert parameter). The inside lethality is strongly influenced by passage time of the cloud, and ventilation ratio. Fixed indoors outdoors toxic ratio
© 2013 TNO
Technical backgrounds
119
By default, the QRA calculation uses a ratio of 1/10: lethality inside is one tenth of lethality outside. For long release durations, high exposures, or high ventilation ratios, this may be a very optimistic assumption: even an outside dose which is much higher than 100% lethality still has maximum 100% lethality, thus 10% lethality inside. Indoor Ventilation ratio This parameter is used in the calculation of inside lethality by toxic exposure. The ventilation ratio highly affects inside toxic exposure. The default value is 1 times per hour, representing natural ventilation. Note that for mechanical ventilation situation values ranging from 2.5 (living room) to 10 (bathrooms, moist environment) are common. Toxic Exposure duration The exposure duration is used to calculate a toxic dose, integrating the concentration (modified including the probit constants) as function of time over that period. (see inclined lines in graph at start of exposure). The duration of exposure is needed as the dose increases the longer one is exposed to an effect. Normally, a default value of 30min (1800s) is used. If in a given location the effect duration is lower than the exposure duration (the passage time of the toxic cloud is around 60s and the user chose an exposure duration of 1800s) EFFECTS will internally rearrange the exposure duration so there is not a loss of accuracy in the result of the integration process. Example The exposure duration is a powerful tool to model evacuation or sheltering. Say, a release happens and people can find shelter after 10 minutes. If we assume that people can find 100% shelter inside houses we can model this as follows: 1. Set the start of exposure to zero 2. Set the exposure duration to the time that people can find shelter (600s) In this case the model starts the exposure at t = zero, which means that people close to the source of release will suffer from the effects but people further away from the release will be exposed to lower concentrations because the cloud has not reached them yet. All these are taken into account by the model. NOTE 1: Different methods of applying this exposure duration are possible, see "exposure duration based on" parameter. NOTE 2: By using the option "perform toxic indoors calculation" the dispersion model can take into account that people inside houses will still be exposed to (lower) concentrations. NOTE 3: For heat radiation, a dedicated "heat exposure duration" parameter is used, which is default 20 seconds, because the human reaction to intensive heat radiation is much quicker.
© 2013 TNO
120
5.1.5
RISKCURVES
Environment parameters Environmental parameters define typical surrounding environment values, used within various consequence calculations. Because separate calculations are being performed for day/night weatherclass situations, some parameters have two (day and night) values.
Ambient Temperature The average yearly temperature or the temperature you want to use for all calculations. In general, the higher the temperature the larger the effects and consequences. Mostly a value between 9 and 25 degrees Celsius is used.
Ground / Surface/ Bund temperature The average yearly temperature of the subsoil that you want to use for pool evaporation calculations. In general, the higher the temperature the larger the evaporation rate and consequences.
Water temperature The average yearly temperature of the water that you want to use for pool evaporation on water calculations. In general, the higher the temperature the larger the evaporation rate and consequences.
Air relative humidity The relative humidity of the atmosphere due to the partial vapor pressure of water in the atmosphere. The relative humidity influences the atmospheric transmissivity. For the Dutch situation the normal variation is between 50 - 90%.
Fraction of CO2 in Atmosphere The amount of CO2 in the atmosphere can be used for transmissivity calculations of solar light. An average value is 0.03%
© 2013 TNO
Technical backgrounds
121
Atmospheric pressure The outside pressure is used in various dispersion and outflow calculations.
Solar Radiation Flux The pool evaporation model uses an overall heat balance for the pool to calculate the evaporation. This heat balance also includes solar heat radiation. Users can choose whether to use a fixed value for solar heat radiation, or calculate the actual value based on day, month, cloud cover and latitude of the location. The solar heat radiation flux is the actual value for the heat flux as used in pool evaporation calculations. Note that values may range from negative (at night: earth radiates towards sky) to 1500 Watts/m 2 depending on the latitude, cloud coverage, and day of the year. When the actual value has to be calculated, several other input values are required: earth location latitude value, cloud cover and day/month of the year.
5.1.6
Meteorological data The meteorological data definition contains the choice for the meteorological station to be used. Any meteorological data set contains probabilities for typical weather classes (Pasquill stability class, wind-speed, day or night) occurring at the location (see meterological distribution). The number of weather classes defined will determine how many damage definitions / consequence models are contained under a scenario (e.g only D5 and F2 or 6 different classes!).
5.1.7
Population Population definition node contains the definition of population by means of grids (a matrix like definition of cells) or polygons (area definition with number of inhabitants). Population can be added by using the Population Import Wizard, or by manually adding a polygon and defining an are with population. See defining Population. The total cumulation of all grids and polygons under the grouping node will be used to create a total population grid, used within the calculation sets Societal Risk calculation.
© 2013 TNO
122
RISKCURVES
Both day an night grid will use a separate "Inside fraction" determining the fraction of the people that are inside houses and have a some degree of protection (see vulnerability settings) When using "temporary polygons", it is possible to use a dedicated "inside fraction" and "utilization fraction" (a presence factor). Temporary population can be used to include the presence of large crowds (e.g. festivals, sport events) during a FRACTION of the time. This is particular relevant if large numbers of people are outside (thus unprotected). Note: When using many (say more than 10) temporary polygons that can be exposed to the same event (when they are close to one another, so within the potential lethality footprint of a single event), this procedure can get time consuming because all potential combinations of these areas need to be evaluated !!!. As an example, just for three temporary population area’s we need to evaluate: A and B and C exposed, A and B exposed, A and C exposed, B and C exposed, only A, only B , only C, and no area (just base population) exposed, where every combination has its own probability of occurrence!!
5.1.8
Equipment Equipment: a location or route on which scenarios are being analysed (distinguishing STATIONARY and TRANSPORT equipment). Note that these nodes can be expanded, they are placeholders or grouping nodes for a list of coordinates, or routes.
© 2013 TNO
Technical backgrounds
5.1.9
123
Scenario Scenario: a Loss Of Containment scenario occurring at an equipment (either a stationary location or a transport route), which has a specific failure frequency, and contains consequence definitions: a description of the scenario in terms of substance, quantities, release situation or resulting damage.
5.1.10 Modelset A Modelset is the placeholder for the actual consequence definition. It contains either a damage definition or consequence calculation, which is defined for a number meteorological conditions
It is possible to define altered input values for specific weather class conditions by selecting the weatherclass from the combobox.
5.1.11 Cumulation set A cumulation set can be used to make a dedicated cumulation of risk sources that does not contain all equipment or all scenario's, presented corresponding SR or IR results. Simple right click the cumulation node and select < Add Cumulation set> . The corresponding input panel allows to select or deselct any scenario or equipment from the list.
5.1.12 Comparison set A Comparison set allows to compare results for Calculations Sets or Cumulation sets; it will provide multiple graphs and contour.
5.1.13 Analysis points An Analysis point provides the possibility to perform a risk contribution analysis for a specific coordinate. Every point will provide an Individual Risk Ranking report, and when societal risk maps are activated, it will also present the corresponding societal risk curve or societal risk contribution for that location.
© 2013 TNO
124
RISKCURVES
5.1.14 Damage definition The damage definition can be used to enter a known consequence area. The use of a damage definition, avoids the requirement of a consequence calculation by the internal models. If for a typical industrial site, various identical equipment are available, it might be useful to perform one single calculation, and add the resulting damage zones as a damage definition, thus avoiding calculation time. Damage definitions always include ONE specific phenomenon, such as (circular) Bleve and local cloud fire, explosion, jet- and poolfire, or toxic damage zones. depending on the typical definition, the number of parameters might differ. Some definitions use values that are entered in a table. Enter the values (distance is row ascending, lethal fraction should be row descending) and press < down> to go to the next row. Is is also possible to copy/paste these values from a spreadsheet. While entering values in a table, a implicit test is being performed on the consistency of the input, the damage dimensions should be row ascending and lethality row descending. If this requirement is not fulfilled, the definition will have a RED label.
A Bleve damage definition requires input: - radius of the fireball ; within this radius, 100% lethality is assumed. - radius of peak overpressure ; within this radius, 100% lethality is assumed. - radius of 25 kW/m2; within this radius, 100% lethality is assumed. - lethality unprotected outside: within this table, users can define lethality versus distance.
A Local Cloud fire damage definition requires input: - radius of the flashfire; within this radius, 100% lethality (can be overridden with vulnerability setting lethal fraction flashfire) is assumed.
An Explosion damage definition requires as input: - Chance flash AND explosion: for an explosion, both an overpressure and flash fire phenomenon can occur. Provided the situation that there is an ignition, (and thus a flashfire), there is a probability of an overpressure effect happening. This probability is defined is this parameter "Chance Flash AND Explosion" . Note that this assumption might differ from other QRA models; it is assumed that overpressure cannot occur without a flash fire, so we have either a flashfire or a flashfire with overpressure effects (and not either overpressure or fire damage) .
© 2013 TNO
Technical backgrounds
125
-Length and weight of flashfire define the size of the damage zone of the flame envelope, described as an ellipse. -The offset of the flashfire can be used to define the distance from release location where the ignition takes place. The offset defines the distance to the boundary of the flame envelope. - The offset of the overpressure effects can be used to change the center location of the overpressure. By default, the overpressure centre is assumed to be the centre of the cloud. This overpressure offset will move the explosion centre, a positive value will move the overpressure centre away from the release centre. - Readius pressure level total destruction and Inside damage. Overpressure damage zones are being defined as two circles: one (high pressure) zone which will have total destruction (typically 0.3 bar overpressure level) and on giving only inside damage due to glass fragmentation. The associated vulnerability is derived from vulnerability settings lethal fraction total destruction pressure level and lethal fraction inside pressure level
A pool- and jetfire damage definition uses: - Flame Length, width and offset: defines the elliptical shape of the flame itself. The offset is an offset of the tail of the flame, relative to the release location. Within the flame envelope the lethality as defined within vulnerability setting lethal fraction in flame contour is used. - Lethality contour levels: these length, width and offset values are used to define elliptical shaped damage zones.
A toxic damage definition uses: - Lethality contour levels for OUTSIDE and INSIDE: these length, width and offset values are used to define elliptical shaped damage zones.
5.1.15 Societal Risk The Societal Risk (SR) is defined as the chance per year that a group of a specific size becomes lethal victim of an accident with dangerous substances. The Societal Risk is presented in a FN curve, presenting the frequency (chance) on a logarithmic Y axis versus the number of victims on a logarithmic X-axis. It is also referred to as "Group Risk"
5.1.16 Individual Risk The Individual Risk (IR) is defined as the chance per year that a person on a specific location, who is continually and unprotected present at that spot, is victim of an accident with dangerous substances. The Iso Risk Contours on a map present locations were the IR has identical values. In some countries the Individual risk is being referred as "Locational Risk" indicating the the value is valid for one specific location or coordinate
© 2013 TNO
126
RISKCURVES
5.1.17 Iso Risk Contours The individual risk criteria assumes 100% presence and an unprotected situation outside. A so-called “Iso Risk Contour” can be drawn by connecting all points with equal Individual Risk.
The Individual Risk can also be presented in a so-called FX curve, which presents the fraction lethal versus distance from the release point, for different wind-directions.
© 2013 TNO
Technical backgrounds
127
Risk contours are available on the level of a calculation set, cumulation sets, comparison sets and individual equipments. A Risk transect can be provide for a specific line track. Such a transect will provide the risk as a function of the place along this track.
5.1.18 Societal Risk (FN) Curve The Societal Risk Curve (FN-curve) presents the cumulated risk that a group of specific size will be killed. The FN curve is depicted as a two dimension graph, using a logarithmic scale on frequency F (Y-axis) and number of victims N (X-axis) axis’s. The curve is interpreted using a “Guide value”, which is a line that should preferably not be crossed. RISKCURVES will present a “Guide Ratio R” value, indicating the distance to this guide value (a guide ratio >1 implies exceeding the guide line), and also presents the “Expected value E” which is the size of the area below the FN curve.
© 2013 TNO
128
RISKCURVES
A FN curve appears to be not very easy to understand or explain. The curve is the result of spatially distributed risk sources that may influence a geographically distributed population distribution, whereas the result only present a curve. Questions that often raise are: “Do we have a problem” and “Where is this problem” or “What is causing this problem”. To be able to answer these kind of questions, a Societal Risk Map was developed and these presentations are now available within RISKCURVES 9.
5.1.19 Project file This RISKCURVES Project file contains all settings and input to be used for a calculation. The file extension is .Riskcurves and it is stored a zipped XML file.
5.1.20 SR Maps SR (Societal Risk) Maps is basically a geographical "Area Specific Societal Risk" presentation of a societal risk, being a two dimensional curve.
© 2013 TNO
Technical backgrounds
129
As a result of the demand for a visualization of the societal risk, a new type of presentation was developed in 2007. The question was raised when a societal risk calculation is fed with geographical based information on population, and geographical based scenario locations, why can we not see a geographical distribution of the societal risk. Such a presentation would be very convenient for emergency response (were are the people who are threatened by accident) or urban planning activities (how much space left for population without exceeding societal risk limits: the guide value)
To provide answers to both question two types of graphs were developed: the Societal Risk Area Map and the Societal Risk Contribution map. The Societal Risk Area map gives an indication of which areas are affected and the height of the risk whereas The Societal risk Contribution Map gives an indication which cells contribute to the societal risk
The bases for the presentation is that every grid cell from the population grid has its own FN curve. In the case of the Contribution map, this curve relates to the victims within this population cell. The higher the risk of this cell (expressed as the expected value of the curve) the more red the color will be.
© 2013 TNO
130
RISKCURVES
For the contribution map, the expected value is used to translate the two dimensional FN graph into a color. The type of coloring can be adjusted, it appears that using a 6 color levels (use legend ) provides the best contrast, but other coloring might improve the visualization.
© 2013 TNO
Technical backgrounds
131
This way the curve represents the full societal risk of scenario's for the area. Note that this area bounded FN curve will never exceed the overall FN curve for all cells.
For the societal Risk Maps it is important to understand that the risk is determined from the receivers point of view (instead of from source). Furthermore, because of the nature of the method, cumulating of various risk sources is possible: transport & stationary installations, small & large scenario’s
The idea behind this new type of visualisation is that this provides a supplementary view of what is happening, and the maps facilitate considering societal risk in early stage of planning process:
© 2013 TNO
132
RISKCURVES
- the SR Area map shows areas with restrictions - the SR Contribution map shows which areas contribute most (emergency response)
© 2013 TNO
Technical backgrounds
5.2
133
The consequence models within a modelset Consequence models will enable users to assess the physical effects of accidental releases of toxic and flammable chemicals. The consequence models inside RISKCURVES are based on a series of models from the Yellow Book [3rd edition, second print 2005], that allow detailed modelling and quantitative assessment of: · · · · ·
Release rate: discharge from a vessel or a pipe of gas, liquid or pressurized liquefied gas: vapour, liquid, two-phase and spray release Pool evaporation: from land or water surfaces of a boiling or a non-boiling liquid Atmospheric dispersion: neutral gas, heavy gas and turbulent free jet Vapour cloud explosion: the TNO multi-energy method or TNT model Heat radiation from fires: BLEVE, poolfire, or jet fire.
the calculation core of RISKCURVES contains consequence models from EFFECTS, which allows the possibility of linking models. By transferring the output of a previous model to the input of a subsequent model it is possible to reduce manual data transfer and to assess the physical effects of complete release scenarios. Note that the models can also be selected in the scenario selection panel on the left of the screen:
Typical predefined chains of models "Combined models" have been defined, allowing to use a chain of models that perform calculations of all possible phenomena's that can occur for a specific chemical and LOC scenario. For every consequence model, a Yellow Book reference is given for the complete description of the model. Within this manual the features of the models are described shortly. In a separate last section the input and output model parameters are explained.
© 2013 TNO
134
RISKCURVES
The detailed description of the various models, including incorporated formula's and relations, can be found in specific chapters of the Yellow Book: · · · · ·
Chapter 2: Outflow and Spray release, separated into Gas release, Liquefied gas releases and liquid releases Chapter 3: Pool evaporation Chapter 4: Atmospheric (vapour cloud) dispersion Chapter 5: Vapour cloud explosion Chapter 6: Heat flux from fires
In the paragraphs below the various effect models for each group of models will be described, including information about: · · · · ·
the reference for the description of the model; the use and characteristics of the model; explanation of input and output parameters; an example calculation with the model, with explanation of the results; description of the linking with other models.
The physical effects occurring upon a release of hazardous material are calculated with the integrated TNO EFFECTS models, which are based on the models described in the Yellow Book [3]. The EFFECTS modules have been incorporated in RISKCURVES.
5.2.1
Gas release The release models calculate the release rate when loss of containment occurs in certain situations. The situations refer to the physical state of the chemical and the characteristics of the failure. EFFECTS will check whether a suitable release model is selected based on physical state of the chemical and the release conditions like temperature, pressure, location of the hole and the calculated height of the expanded liquid in a vessel. An error message will appear if an incorrect model is selected, indicating the physical state of the chemical under the specific release conditions. Distinction is made between a gas release from a vessel or a (short) pipe connected to a vessel, and a (non-stationary) release from a long pipeline, not (necessarily) connected to a vessel. The "gas release 10 minutes" model will search for a corresponding hole size for a 10 minutes scenario: Which size of the hole is required to get an representative rate which equals the flow required for the scenario in which the full inventory is released in 10 minutes
5.2.1.1
Gas release from a vessel or pipe
Reference The model is described in the Yellow Book [1997], sections 2.5.2.1 to 2.5.2.4.
© 2013 TNO
Technical backgrounds
135
Characteristics and use The model is suitable for the following situation: Hole in a vessel or in pipe connected to a vessel containing (compressed) gas only. The hole in the pipe may either be a small leak or a full bore rupture.
The initial release rate mainly depends on the leak size, the discharge coefficient, the initial pressure inside the vessel and the length of the pipe (in case of a release from a pipe). Because gas flows out of the vessel, and assuming no gas is being supplied, the pressure in the vessel will drop and therefore the release rate decreases in time. The rate of decrease mainly depends on the vessel volume. The expansion of the gas, because the pressure drops from the pressure of the releasing gas to ambient, is taken into account. The model assumes adiabatic outflow.
Releases from a hole in a pipe connected to a vessel are in principal not different from the releases from a hole in the vessel itself, with the exception of the friction resistances of the flow through the pipeline. For this resistance the Fanning friction factor is used, which depends on the length, the diameter and the internal roughness of the pipe.
5.2.1.2
Gas release from a long pipeline
Reference The model is described in the Yellow Book [1997], section 2.5.2.5. and is also called the "Wilson model"
Characteristics and use The model is suitable for the following situation: Total rupture of a long gas pipeline, not (necessarily) connected to a vessel. Distinction is made between outflow from a full-bore rupture and through a small hole in the pipeline. The initial release rate mainly depends on the pipe diameter (= leak size) or hole size, the friction of the flow inside the pipeline depending on the wall roughness and the initial pressure inside the pipeline. Because of the release the pressure inside the pipeline will drop in the region of the leak at first. The pressure drop ‘travels’ along the length of the pipeline, with a velocity equal to the sound velocity. This causes the gas release to become non-stationary until the pressure drop reaches the end of the pipeline. This is the point at which the model stops the calculation. The ongoing release can be assumed to be stationary and continuous until the pipeline is empty.
© 2013 TNO
136
RISKCURVES
Note that model assumes a single sided outflow at the end of a pipe with the provided length. The model is often linked to the "Turbulent Free Jet" model, which needs consistent input with respect to flowrate, pressure and corresponding diameter. Although the model includes an output value "model valid until time" the determining outflow rate ("representative rate") is usually occurring within this time. The representative duration is calculated as the time needed to empty the pipe at this rate, and this duration is allowed to be larger than "model valid until time": this is the estimated duration of the outflow, WHEN IT SHOULD EMPTY AT THE REPRESENTATIVE RATE. In reality the outflow duration will be much longer, because is starts at a high rate, but this rate rapidly decreases to very low values.
5.2.2
Liquefied gas release For a liquefied gas release the following differentiation in type of release is made:
A distinction is made between a vapour, champagne or a pressurized liquefied gas release from a vessel or a (short) pipe connected to a vessel. For a pressurized liquefied gas the following differentiation in type of release is made, depending on the location of the hole related to the liquid level in the vessel.
DIERS (Top venting) The hole is above liquid level in the vapour phase, but below the expanded boiling liquid level. Because of the release of vapour the liquid starts to boil and the expanded boiling liquid may reach the location of the hole. In that case liquid drops will be carried along with the releasing vapour.
Simple vapour release Hole above liquid level in the vapour phase and also above the expanded boiling liquid level. In that case only vapour will release.
Liquefied gas from long pipeline (Morrow) This model is dedicated to calculate the two phase release from a long pipeline. In the combined models (universal release or LOC models), this model is used whenever the pipeline length exceeds 1 km. Note that the release is based on the contents of the (blocked) pipeline itelf.
© 2013 TNO
Technical backgrounds
137
Instantaneous flashing release This model calculates the division into an airborne mass and a rainout mass. It calculates the adiabatic flash (the amount of liquid that can evaporate when cooling to to atmospheric boiling temperature) and uses an empirical correction on this for liquid fraction in the cloud.
Spray release This model calculates the rainout and droplet (aerosol) formation of a two phase continues release.
Pressurized Liquefied gas release (TPDIS, bottom venting) The hole is below liquid level in the liquid phase. Because of the pressure drop inside the pipeline connected to the vessel vapour will be formed inside the pipeline. The resulting release will be a two-phase release; both liquid and vapour releasing. In the case that the pipeline is very short (hole in vessel) only liquid will release.
For pressurized liquefied gases the models to calculate the flash evaporation and the evaporation due to the mixing with air immediate after the release (e.g. spray release) are part of the release rate models. 5.2.2.1
DIERS top venting (vessel only)
Reference The model is described in the Yellow Book [1997], Paragraph 2.5.3.2
Characteristics and use The model is suitable for the following situation: Vessel containing a pressurized liquefied gas with a hole in the vessel above liquid level in the vapour phase, but below the expanded boiling liquid level. In the vessel a vapour-liquid equilibrium holds, with a pressure equal to the vapour pressure at given temperature.
Because of the release of vapour the liquid starts to boil and the expanded boiling liquid may reach the location of the hole. In that case liquid drops will be carried along with the releasing vapour. The release rate is considerably increased by this champagne effect. The expanded boiling level is determined by the hole size, release conditions (pressure) and the properties of the chemical.
© 2013 TNO
138
5.2.2.2
RISKCURVES
Vapour release from vessel or pipe
Reference The model is described in the Yellow Book [1997], section 2.5.3.
Characteristics and use The model is suitable for the following situation: Vessel containing a pressurized liquefied gas with a hole in the vessel or in a pipe connected to a vessel above liquid level in the vapour phase and also above the expanded boiling liquid level. In the vessel a vapour-liquid equilibrium holds, with a pressure equal to the saturated vapour pressure at a given temperature. The model is based on the phenomenon of adiabatic vapour release.
The initial release rate mainly depends on the leak size, the discharge coefficient and the initial pressure inside the vessel. Because vapour releases, the pressure in the vessel will drop, causing the liquid to start to boil off. Because of the boiling the temperature of the liquid in the vessel will decrease and so the vapour pressure decreases. Therefore the vapour release rate decreases in time. Releases from a hole in a pipeline connected to a vessel are in principal not different from the releases from a hole in the vessel itself, with the exception of the friction of the flow through the pipeline. This friction depends on the length, the diameter and the internal roughness of the pipe. In an equilibrium situation the remaining liquid cools down to its boiling point at atmospheric pressure. The expansion of the vapour, because the pressure drops from the pressure of the releasing vapour to ambient, is taken into account.
5.2.2.3
Pressurized liquefied gas release from vessel or pipe
Reference The model used is the 2-phase flow model TPDIS developed by the Finnish Meteorological Institute [Kukkonen]. The model is described in the Yellow Book [1997], section 2.5.3.5.
Characteristics and use
© 2013 TNO
Technical backgrounds
139
The model is suitable for the following situation: Vessel containing a pressurized liquefied gas with a hole in the vessel or a hole in a pipe connected to a vessel below liquid level in the liquid phase. In the vessel a vapour-liquid equilibrium holds, with a pressure equal to the vapour pressure at given temperature.
This 2-phase flow model assumes that for a length of the pipeline connected to the vessel smaller than 0.1 m (and also for a hole in vessel) only liquid release results; no vapour is formed before the release. The initial liquid release mainly depends on the leak size and the pressure in the vessel. Because of the liquid release the liquid level decreases, by which the vapour phase increases. Therefore more vapour will be formed by boiling of the liquid, by which the temperature and hence pressure in the vessel decreases. Because of this phenomena and of the decreasing of the liquid height in the vessel the liquid release rate will decrease somewhat in time. In case of a leak in a pipeline containing a pressurized liquefied gas a pressure drop inside the pipeline may occur. This occurs mainly for a large leak (about equal to the pipe diameter) and for a pipe length larger than about 0.1 m. As a result of the pressure drop inside the pipeline, vapour will already be formed inside the pipeline and the resulting release is 2phase, both liquid and vapour. The TPDIS model will check for itself whether the release will be 2-phase or only liquid.
The basic model assumptions of TPDIS [Kukkonen] are the following. In the model the flow has been divided into three flow regimes: (I) superheated liquid, (II) expanding two-phase fluid and (III) equilibrium two-phase fluid. Homogeneous equilibrium flow has been assumed in the third flow regime. The homogeneity assumption implies that the fluid is a homogeneous mixture of vapour and liquid, and that the phases move with the same velocity. For long pipes, the length of the regime (III) may be nearly equal to the pipe length. The process is assumed to be adiabatic. This is a reasonable assumption, as the outflow is very rapid, and the heat energy conducted through the pipe walls is therefore much less than the energy of the phase transitions.
5.2.2.4
Spray release of pressurized liquefied gas from vessel or pipe
Reference The model is described in the Yellow Book [1997], section 2.5.3.7.
Characteristics and use
© 2013 TNO
140
RISKCURVES
The model is suitable for the following situation: evaporation of a release of pressurized liquefied gas from a vessel or a pipe with a boiling point below ambient temperature; (pressurized) liquefied gas release or champagne release.
The evaporation of a released pressurized liquefied gas will be determined by the following phenomena: – Flash-off. A flash-off means that as a result of the reduction of the pressure to atmospheric pressure the liquid will spontaneously start to boil. The necessary heat of evaporation for this is drawn from the liquid which will cool down to its boiling temperature at atmospheric pressure as a result. – Rain out and air entrainment. The release and the flash-off will generally be so violent that the liquid in the jet will be broken into drops and the jet will mix with air. The liquid drops will fall onto the ground (rain out). Due to the air entrainment, part of the liquid drops will evaporate during falling. Because of the withdrawal of the heat necessary for this evaporation the temperature of the air-vapour mixture will decrease to below the boiling temperature of the chemical. The spray release model calculates the amount of the liquid that will rain out, which forms a pool on the subsurface.
Linking of model data The following parameters will be linked to other models. Link to neutral gas dispersion - (semi-)continuous release: –
net air-borne mass flow rate source dimensions
Link to heavy gas dispersion model - horizontal/vertical jet: –
net air-borne mass flow rate source dimensions temperature after rain out (boiling point) vapour mass fraction after rain out
Link to pool evaporation, boiling pool: –
total mass liquid rained out
© 2013 TNO
Technical backgrounds
141
REMARK: The pool evaporation model is suitable for instantaneous releases, therefore this model has to be used with care in the link with the spray release model. A solution will be the following: From experiments it is known that in most cases all the liquid drops in a spray release will be evaporated by the mixing with air. This means no rain out of liquid. This total evaporation of the spray can be calculated by adjusting the source exit height to higher values. The ongoing dispersion of this total evaporated spray release can be directly calculated with the suitable dispersion model (see above). Within the dispersion model the height of the source has to be adjusted to the actual value.
5.2.2.5
Instantaneous flashing liquid release
Reference The model based on AMINAL- Belgium, "Nieuwe richtlijn voor het berekenen van flash en spray" doc.97/001, which is original source of table 4.8 of Purple Book CPR 18E.
Characteristics and use The model is suitable for the following situation: evaporation of an instantaneous release of pressurized liquefied gas. Instantaneous release means that the entire contents of the vessel or system are released in a very short time.
The evaporation of an instantaneous released amount of pressurized liquefied gas will be determined by the flash-off and by the evaporation due to mixing with air. Based on experiments the total airborn mass equals may be about 2 times the flashed amount, but depends on the adiabatic flash fraction The consequence of this evaporation is that an instantaneous gas cloud will occur. The remaining liquid, which is cooled to its boiling point will form a pool on the ground. For this a suitable pool evaporation model have to be used.
Because the model uses the AMINAL approximation to calculate the total mass in the cloud, this total "airborne mass" (mass remaining in air: not rained out) is partly vapour (the adiabatic flash amount) and partly liquid droplets. When using a dispersion model based on the calculated total mass, and there is liquid in the cloud, it is suggested to use a "Dense dispersion model".
Linking of model data The following parameters will be linked to other models. Link to neutral gas dispersion - instantaneous release:
© 2013 TNO
142
RISKCURVES
–
mass of vapour evolved
Link to heavy gas dispersion - instantaneous gas release: –
mass of vapour evolved
–
temperature vapour/liquid (at boiling point)
Link to pool evaporation - boiling liquid: –
mass of liquid in pool
REMARK: The total mass evaporated calculated with the model for pool evaporation from land of a boiling liquid is considered as an instantaneous source and can be added to the mass of vapour evolved due to the flashing liquid instantaneous release. This total evaporated amount (flash, mixing with air, boiling pool) can be linked to the total mass released in the dispersion model (neutral and heavy gas) for an instantaneous source.
The calculated footprint of the local cloud fire is based on the shape of a half sphere, where the material is mixed to Upper Explosion Limit. For non flammable materials, expansion of the full airborne amount (including liquid droplets) to atmospheric vapour is considered.
Note that this half sphere (mixed with air and/or expanded to pure vapour) is not the typical situation that should be used as input for dispersion. For that reason, the calculated density of the airborn mass gives an impression on whether to use dense gas dispersion or neutral gas dispersion (for 2 phase mixtures usually much heavier than air: densegas) 5.2.2.6
Liquefied gas from long pipeline
The Morrow model (non stationary outflow from long pipeline) can be used to calculate the behaviour of expanding pressurized liquid in a pipeline after a rupture. Blocking of the pipeline is assumed, and the release is based on the contents of the pipeline itself. The model is valid until the distance to the interface (traveling pressure wave) is larger then the length of the pipeline. The time that is need for these model calculations is shown in the output box ‘Model valid until time’. After this the calculations continues with the predicted mass flow rate of the last time step until all mass is removed. Note that as of version 8.1, the model assumes a One sided outflow at the end of a pipe with the provided length, just like the similar Wilson model for gas pipelines. This is changed because the model is often linked to "spray release" which needs a consistent input with respect to flowrate, pressure and diameter. If a two sided outflow is to be evaluated, two separate pipelines need to be modeled. Although the model includes an output value "model valid until time" the determining outflow rate ("representative rate") is usually occurring within this time. The (extrapolated) outflow duration is also calculated, but this is only an estimation because it is larger than "model valid until time".
© 2013 TNO
Technical backgrounds
143
It is not advised to use values from the graphs at time t higher than "model valid until time".
5.2.3
Liquid release
Reference The model is described in the Yellow Book [1997], section 2.5.4.
Characteristics and use The model is suitable for the following situation: Vessel containing a liquid with a hole in the vessel or a hole in a pipe connected to a vessel below liquid level in the liquid phase. For this model a liquid may be: –
a non-boiling liquid (boiling point above ambient)
–
a gas cooled to liquid at or below its boiling point
– a pressurized liquefied gas at a pressure higher than the vapour pressure at the storage temperature. For this model it is assumed that the pressure above the liquid level inside the vessel remains constant.
The release rate is calculated using the Bernoulli equation. The release rate mainly depends on the leak size and by the pressure above the liquid plus the hydrostatic pressure of the column of liquid (from height of leak to filling height). Because of the release of liquid the hydrostatic pressure of the liquid column will decrease and so the release rate decreases in time. In total the amount of liquid between the height of the leak and the initial filling height will be released. Most times the releasing liquid will form a pool on the ground from which evaporation takes place. For this a suitable pool evaporation model must be used.
Releases from a hole in a pipeline connected to a vessel are in principal not different from the releases from a hole in the vessel itself, with the exception of the friction of the flow through the pipeline. This friction depends on the length, the diameter and the internal roughness of the pipe. Keep in mind that this model is valid for a pipeline connected to a vessel without a pump installed in this system. In the event that a pump is present upstream of the leak the release rate will be maximized by the maximum rate of the pump. Furthermore in this case it has to be identified whether the pump will still be running or will trip.
© 2013 TNO
144
RISKCURVES
Linking of model data The following parameters will be linked to other models. Link to pool evaporation model for non-boiling liquids: –
total mass released over total release duration
Link to pool evaporation model for boiling liquids, relevant for releases of gases cooled to liquid at boiling point: –
5.2.4
total mass released over total release duration
Pool evaporation The pool evaporation models calculate the amount of vapour evaporated from a liquid pool, which is formed on the surface after the release of liquid material. The evaporation model can either work with instantaneous and continuous supply of liquid, and will determine itself wether it is boiling or non-boiling liquid. Furthermore the type underground (land or water) and spreading conditions (bunds) will have to be provided by the user.
Spreading pool or Spreading in bunds Each choice has its own specific behaviour with respect to pool area, thickness of the layer and thus evaporation behaviour
Evaporation from land or water Differentiation is made between non-boiling liquids and boiling liquids by the program itself, based on temperature of the release, and the characteristics of the chemical. The model determines the conditions by itself.
Non-boiling liquid A non-boiling liquid is a liquid with a temperature below its boiling point. Normally the boiling temperature of the chemical will be above the ambient temperature. For a non-boiling liquid with a boiling temperature below the ambient temperature (= temperature subsoil) the liquid will rise in temperature because of the heat drawn from the subsoil by which the liquid will become a boiling liquid: Then the non-boiling evaporation model is not suitable.
© 2013 TNO
Technical backgrounds
145
Boiling liquid In EFFECTS it is assumed that for the evaporation of a boiling liquid the temperature of the liquid is equal to its boiling point, which is below the ambient temperature. This will be the case for: · ·
a gas cooled to liquid (refrigerated liquid) a gas compressed to liquid: After the immediate evaporation (flash-off and by entrainment of air) the possibly remaining liquid, which is cooled to its boiling point at ambient pressure will form a liquid pool on the surface. Then the same situation has occurred as for a refrigerated liquid.
Subsoil Type Determines the heat transfer behaviour, see Subsoil type
Subsoil Roughness The roughness description relates to the minimum pool thickness that can occur, see Subsoil roughness.
Results:representative values All "Purple book representative values" are being calculated on the base of the selected representative step
Representative density The density of the vapour that is released from the pool is calculated on the base of mixed with air from a 0.5 mtr top layer Source chemical: Surface area * source rate/m2 = source rate M evaporation [kg/s] at atmospheric pressure and representative temperature Input air: Wind speed [m/s] * width pool [m] * 0.5 m height * 1.2 kg/s = amount of air mixed M air in in kg/s The combination of these two rates and density gives the density of the mixture.
© 2013 TNO
146
RISKCURVES
Reference "An advanced model for spreading and evaporation of accidentally released hazardous liquids on land" by I.J.M. Trijssenaar-Buhre, R.P. Sterkenburg & S.I. Wijnant-Timmerman TNO, Utrecht, The Netherlands
5.2.5
Atmospheric dispersion The gas or vapour released will be dispersed in the surrounding area under the influence of the atmospheric turbulence. The concentrations of the gas or vapour released in the surrounding area can be calculated by means of the atmospheric dispersion models. These concentrations are important for determining whether, for example, an explosive gas cloud can form or whether injuries will occur in the case of toxic gases. Within EFFECTS a first differentiation is made between the following three types of dispersion models:
In the dispersion models account is taken of the atmospheric stability, the so-called Pasquill classes (A to F) and a certain wind velocity. The source dimensions are taken into account by means of an imaginary (virtual) point source wind upwards, for which the dispersed dimensions at the point of the actual source are equal to the actual source dimensions. The dispersion models apply only to open terrain. However allowance is made for the roughness of the terrain. The influence of trees, houses, etc. on the dispersion can be determined by means of a class of the roughness length.
Neutral gas dispersion The neutral gas dispersion model is based on the Gaussian plume model and no account is taken of the difference in density between the ambient air and the gas. Because of this, the model must only be used for gases with a density approximately the same as air, or if the gas concentration at the point of release is low. The direction of the release is always taken as horizontal to the wind direction. Both Neutral and Dense gas dispersion models are available for the following type of releases: Concentration, Explosive mass and Toxic dose. For the neutral gas and heavy gas dispersion models the following type of calculations can be carried out:
© 2013 TNO
Technical backgrounds
147
Concentration contour The model calculates: – –
– –
the dimensions of the contour (length and max. width) at given height the maximum concentration and corresponding distance at time t: only for neutral gas dispersion, instantaneous release and semi-continuous release when cloud has drifted away from its release point. graphical presentation of the contour in X-Y directions graphical presentation of concentration with distance.
For semi-continuous and instantaneous releases the concentration contour is calculated for one specified time t.
Explosive mass The model calculates: – – – –
the explosive mass, for concentrations higher than LEL or between LEL and UEL the dimensions of the LEL contour the dimensions of the UEL contour (not for heavy gas dispersion) whether the source is at ground level or the plume touches the ground level or it is a free plume. For semi-continuous and instantaneous releases the explosive mass is calculated for one specified time t. For neutral gas dispersion the explosive mass and the dimensions of the LEL and UEL contour are calculated for a height equal to the source height. For heavy gas dispersion these parameters are calculated for a height equal to zero (ground level).
Toxic load The model calculates: the toxic load, Cn.t, with C = concentration in mg/m 3 and t = duration in min at position (x, y, z) for neutral gas dispersion · at position (x, 0, 0) for heavy gas dispersion for a certain exponent n and for a given maximum exposure time after arrival cloud · the maximum concentration at position (x, y, z), and for semi-continuous and instantaneous releases the corresponding time at position (x, y, z) · the arrival time and departure time of the cloud at position (x, y, z) (not for continuous releases) For these parameters a concentration equal to 1% of the maximum concentration is assumed. ·
Continuous, Semi-continuous or Instantaneous
© 2013 TNO
148
RISKCURVES
Within the model itself, the user has to choose for continuous, Semi-continuous or Instantaneous.
For rather long releases the continuous release dispersion model has to be used and for very short releases the instantaneous release dispersion model. In general the following is used to judge whether the source can be considered as continuous or instantaneous [Yellow Book]: Continuous: at distances < 1.8 * wind velocity * duration of release Instantaneous: at distances > 18 * wind velocity * duration of release Semi-continuous: intermediate cases. The dispersion calculations for the semi-continuous releases could be rather time consuming.
Dense gas dispersion If the gas has a higher density than air (because of a high molecular weight or of a low temperature) it will tend to spread in a radial direction because of gravity, resulting in a ‘gas pool’. As a result of this, in contrast to a neutral gas, the gas released may spread against the direction of the wind. In combined models, the selection criteria for using dense gas is: Dense gas is situation where density of mixture (possibly including liquid droplets with very high density is 10% heavier than air.
Heavy gas dispersion models are available for the following type of releases: Instantaneous gas release: instantaneous release of gas, vapour or flashing liquid. Pool evaporation:
vapour source is formed by evaporation from a pool
Jet, horizontal or vertical: (semi-)continuous release of gas, vapour or spray release in vertical or horizontal direction
Turbulent free jet When a gas or vapour releases and the Reynolds number under the release conditions is greater than about 2.5.104 (e.g. high release velocity) a jet occurs. Another condition is the absence of obstacles in the jet. Turbulent free jet dispersion occurs when the gas velocity at the release equals or is close to the velocity of sound.
© 2013 TNO
Technical backgrounds
5.3
149
Model input parameters Each model inside EFFECTS uses it's own set of required input fields. To obtain additional information about a specific input box, simply press while the cursor in the input box, and specific information about the field will pop up.
© 2013 TNO
150
5.4
RISKCURVES
Combined models New in version 9 is the possibility of using combined (or automated) models. Basically they consist of a pre-described chain of models, linked together into one combined model. Because a lot of the input parameters of a model can be taken from output of the preceding model, the required input of the combined model is not the same as "all inputs of all models together". Although they are referred as being a model-chain, it is better to think of a combined model as being a tree, because it may consist of several branches. The figure below illustrates that outflow model 1 can be followed by a spray release model 2 and a pool evaporation model 3. Note that models can share equal inputs, such as the selected chemical, the wind speed and and the ambient temperature, and some inputs are taken from a preceding model, like a rain out mass or liquid (droplet) fraction of airborne mass. Furthermore, depending of the conditions of the cumulated source, the appropriate dispersion model has to be run. In the occasion of a two phase release, the source rates from for instance airborne mass rate from spray release and evaporation rate from pool-evaporation, are to be cumulated, to create the combined source input for the dispersion model. The cumulation procedure as used in the models is described in detail in cumulation of sources. Note that the combined model often incorporates 4 types of dispersion models which will be abbreviated in the model log: HGDE: Heavy Gas Dispersion Explosive mass model, (Inst represents Instantaneous mode, Pool represents Pool evaporation mode) HGDT: Heavy Gas Dispersion Toxic model NGDE: Neutral Gas Dispersion Explosive mass model NGDT: Neutral Gas Dispersion Toxic model
© 2013 TNO
Technical backgrounds
151
Furthermore, depending on the properties of the chemical material incorporated (and the selection of phenomena to evaluate), the model chain will decide which mode (toxic of flammable) or typical model type (heavy gas dispersion or neutral gas dispersion) to activate. Usually, the branching starts with selection of the typical phase of the chemical (gas, liquefied gas or liquid). Apart from material phase, the type of scenario (instantaneous or semi continuous) will also play an important role in selection of the release model.Typical LOC (Loss Of Containment) scenarios will consist of an outflow (release) model (possibly followed by spray release or turbulent free jet), followed by a possible pool evaporation model (if there is a rain-out mass) and finally a dispersion model where the density of the gas determines whether to use the heavy gas or the neutral gas model. In order to be able to determine the correct chain, preconditions have been defined for all branches of the model tree. For example: pool evaporation and pool fire models will require the existence of a pool (pure liquid phase) or rain out mass from the two phase release. The type of decisions that have been made along the route of calculating, are being presented in the model log. All models that have been run will be listed here. Models that were skipped from calculation will provide a reason why: "neutral gas dispersion did not run because precondition not fulfilled: not a neutral gas" or "material is not toxic" Apart from the chain of events as illustrated above, some specific phenomena can also be incorporated within combined models: depending on the type of ignition: a BLEVE, Pool fire or Jet fire can happen. The occurrence of these phenomena also depends on the input parameters: type of release (instantaneous, continuous) and state of the chemical.
© 2013 TNO
152
RISKCURVES
The combined models will automatically incorporate all possible phenomena according to the schedule below.
5.5
Cumulation of sources In the occasion of a release from a two phase chemical the amount of mass that is thrown into the air and is to be used as the dispersion source rate can be determined by two processes: · Material that remains in air after the release (flash or spray) · Material that rains-out, but eventually will evaporate from a pool The way the two sources are combined depends on the kind of release that is occurring. Instantaneous (G1) scenario In case of an instantaneous G1 scenario, the dispersion models will have to run in "instantaneous mode", whereas the source rate from the pool is an continuous source. For that reason, two dispersion models will need to run. Note that it can occur that the instantaneous flash will be a heavy gas (due to the liquid fraction and temperature), whereas the pool evaporation source may be "neutral". Note the density of the evaporated mass is based on mixing with air of a 0.5 meter window height. After calculation of (toxic or explosive) result, the dispersion results itself are cumulated. For this cumulation of an instantaneous (G1) dispersion result, distinction has been made between cumulation of dispersion-explosive models and cumulation of dispersion-toxics models.
© 2013 TNO
Technical backgrounds
153
In normal mode, the presented contours and lethality grids will only present the corrected dominating source, if the "expert mode"has been chosen, the secondary dispersion results are also visualized.
© 2013 TNO
154
RISKCURVES
© 2013 TNO
Technical backgrounds
155
Continuous release In case of the continuous release, the source rate is determined by: 1. The 2 phase Bottom Discharge (TPDIS) model, followed by spray release, which calculates rainout mass flowrate and a airborn flowrate 2. The pool evaporation, fed by the rain out massrate will also create a continuous sourcerate. The dispersion model has to be fed with important cumulated parameters: combined mass flowrate, representative release duration, and liquid fraction of the mixture. Not that if the input chemical is a pure liquid, the dispersion model will run in "poolevaporation mode" and input will be purely the pool evaporation mass rate (release height 0). If the chemical is a gas, the cumulation routine will skip the pool-input, and the following dispersion model will run in "horizontal jet" mode, with dimensions taken from the jet diameter.
© 2013 TNO
Appendices
6
Appendices
6.1
List of chemicals
157
The full list of chemicals in the extended (DIPPR) database can be found in a separate document. The following chemicals are included in the standard YAWS database of EFFECTS: Acetic acid Acetonitrile Acrolein Acrylonitrile Air Allylamine Allylchloride Ammonia Arsine Benzene Bromine Butadiene (1,3,) Butane (n-) Butene (1,) Butylamine (n-) Butylamine (s-) Butylamine (t-) Carbon monoxide Carbondioxide Carbondisulfide Carbonic dichloride (phosgene) Carbonyl fluoride Chlorine Chloroacetaldehyde Chloroacetyl chloride Chloroform Chloroprene Cumene Cyanogen Cyclohexane Diborane Dichloroethene (1,1,) Dichloromethane Dimethyl amine Ethane Ethene Ethyl acrylate Ethyl amine Ethyl mercaptan Ethylbenzene Ethylene dichloride Ethylene oxide Ethyleneimine Fluorine
© 2013 TNO
158
RISKCURVES
Formaldehyde Formic acid Gasoline Hexafluoroacetone Hydrazine Hydrogen Hydrogen bromide Hydrogen chloride Hydrogen cyanide Hydrogen fluoride Hydrogen peroxide Hydrogen sulfide Isobutyl amine Isobutylene Isoprene Isopropyl amine Ketene Methane Methanol Methyl acrylate Methyl amine Methyl bromide Methyl chloride Methyl iodide Methyl mercaptan Methyl methacrylate Nitro propane (2,) Nitrogen Nitrogen dioxide Nitrous oxide Oxygen Pentane (n-) Perchloryl fluoride Phosphine Propane Propene Propyl mercaptan Propylamine (n-) Propylene oxide Styrene Sulphur dioxide Sulphur trioxide Tetrachloroethylene Tetrahydrofuran Toluene Trimethyl amine Water Xylene (m-)
6.2
Low level error messages In some cases an error happens at such a low level that the software can not link this to an explanation string and it only displays a cryptic dialog.
© 2013 TNO
Appendices
159
In case of reproducible situation, you might send an email to the helpdesk, describing the actions you performed that raised this error.
Note that the helpdesk will not provide support for internal Windows errors, or errors that cannot be reproduced on a standard, clean, system.
Apart from the details information provided in the message, it is very often useful to include the .Effects projects file that has triggered the error as an attachment to the email (preferably ZIP the project file to reduce file size) .
It is strongly advised to describe the circumstances under which the error happened.
© 2013 TNO
160
6.3
RISKCURVES
Known limitations Ignition points RISKCURVES is not yet capable of working with location dependant ignition propabilities or ignition points. This implies that the so-called "free field" method is used for both individual risk and societal risk calculations. The free field method assumes explosion occurring at the point where the LEL cloud has reached its maximum size. (Time is tMac as reported by dispersion explosives model).
Mixtures of chemicals Our chemical database currently does not support the definition of mixtures. For well known "real world" mixtures such as natural gas and LPG are usually modelled as Methane and Propane, but for other mixtures the following table is often used to derive the closest chemical.
Spray release model 1) The model is not feasible for use with e.g. CO2, which has a solid-vapor equilibrium at atmospheric conditions. (normal boiling point unknown or lower than melting point. 2) The model does not take solidification (snow) into account: the energy balance doesn't include solid phase transition enthalpy changes. Neutral gas dispersion For instantaneous or semi-continuous release the passage time of the cloud is not reported in the results. (artifact 35647)
© 2013 TNO
Appendices
161
Dense gas dispersion The dense gas dispersion calculation model, currently used in EFFECTS, is based upon the SLAB code. At the moment we are aware of a number of issues with SLAB, which may result in unreliable results. We are working on improving the dense gas calculation engine. Some situations, potentially providing unreliable results, are listed below. 1) In SLAB, there is a transition between different models at the moment when the release stops (i.e. after the “duration of the release”). This may cause differences between results just before and just after the end of the release. In a “concentration vs. time”-graph, there may be a discontinuity at this moment. (artifact 30165) 2) For evaporating pool sources, the model sometimes cannot do a computation for certain combinations of mass flow rate, pool area, roughness length, stability class and wind speed. (artifact 37421) 3) The input parameter “diameter of expanded jet” has a substantial influence on the results, but can be hard to estimate by the user. Some release models, such as “spray release” can be used to compute this diameter. (artifact 30682) Pool evaporation 1) When evaporation is higher than the mass flow into the pool, the pool thickness decreases until the minimum pool thickness (surface roughness) is reached. After that, the pool will shrink in size. For pools in bunds on water, that model has not been implemented completely; for those pools, after the maximum area (bund area) has been reached, the pool thickness keeps decreasing until it is zero, while the pool area is constant. (artifact 35366) 2) The transition from non-boiling to boiling for cryogenic substances works for spills on land, but not on water. The calculation stops when the boiling temperature is reached. Workaround: set the initial temperature of the spilled substance equal (or slightly higher than) to the boiling temperature. (artifact 32938) 3) Volatile liquids on water. When a volatile liquid is released, its temperature usually decreases. This causes the water to also cool down. In the model for heat transfer that is implemented in EFFECTS, there is no heat transfer by convection (i.e. the water behaves like a rigid body). Therefore, the user may choose to set a maximum temperature difference between the pool and the water to limit the temperature decrease of the pool. When the maximum temperature difference is reached, the chart of “heat flux from subsoil versus time” is not correctly presented: the heat flux from subsoil as calculated is not sufficient to keep the temperature constant. An additional heat flux term, stemming from heat transfer by convection of the water, would keep the pool at a constant temperature. Combined models 1) For continuous release of liquefied gases, the heavy gas dispersion model always uses the ‘horizontal jet’ as source mode, even if the pool evaporation is dominant relative to the spray release. 2) Combined models do not handle vertical jet releases correctly. For this to work, the neutral gas dispersion model should be able to handle vertical jet releases. Also, the heavy gas dispersion model would need to be able to interpret an outflow angle and decide if it is a horizontal or vertical outflow.
© 2013 TNO
162
RISKCURVES
© 2013 TNO
Index
Formula 99 Formula ID 99 Fraction confined mass in Multi energy explosion method 89 Fraction of CO2 in Atmosphere 120
Index -A-
-G-
A new GUI 21 Accuracy parameters 113 Accuracy settings 94 Air relative humidity 120 Ambient temperature 120 Atmospheric pressure 120
Gas release from a long pipeline 135 Gas release from a vessel or pipe 134 Geo-referencing 107 Graph display panel 60 Graph selection box 67 Ground / Surface/ Bund temperature 120 Group Risk 125
-CCalculation Set 22, 54, 113 Calculation Settings 22, 113 CalculationSet definition 54 Cell size Risk grids 113 Chemical database editor 96 Chemical Databases 95 Command button panel 65 Comparison set 22, 123 Concentrating averaging time flammables 89 Concentrating averaging time toxics 89 Cumulation set 22, 123 CurveNumber for Multi energy explosion method
-DDamage definition 124 Default mixingheight 89 DIERS top venting (vessel only) Display units 85
-HHeat radiation damage probits 116 Heat Radiation Exposure Duration 116 Heat radiation level total destruction 116 Hole contraction coefficient 89 How to use the built-in GIS system 71
-I89 Individual Risk 125 Indoor Ventilation ratio 116 Installation of the software drivers for the dongle Inter accident distance 113 Inter accident distance FN 113 Introduction 7
137
-K-
-E-
Known limitations
Editing constant properties 99 Editing properties of chemicals 99 Editing temperature-dependent properties Environment parameters 120 Environment settings 93 Equipment 22, 122 Error messages 158 Expert Parameter settings 89
-L-
-FFixed indoors outdoors toxic ratio FN subsectors 113 © 2013 TNO
163
116
99
160
Lethal fraction flame contour 116 Lethal fraction flash fire 116 Liquefied gas from long pipeline 142 Liquefied gas release 136 List of chemicals 157 LOC scenario 22, 123 Log Panel 78 Lowest_significant_frequency 113
8
164
RISKCURVES
Scenario panel 64 Selecting a chemical 97 Societal Risk 125 Societal Risk Map 13, 128 SR map 13, 128 StandardPipeRoughness 89 Stationary equipment 22, 122 System requirements 8
-MMap display panel 69 Mass and volume calculator 105 Maxium number of accident points 113 menu bar 49 Meteorological data 22, 121 Meteorological distribution 91 Minimum valid and maximum valid temperatures Modelset 22, 123 Mortality Probit calculator 106
-NNode input panel 66 Number subsectors FX
113
-OOptions
99
-TThe user interface in detail 47 TNO software products 7 toolbar 50 Toxic Exposure duration 116 Toxic Inhalation Heigth 89 Transport equipment 22, 122 Tree nodes 22
-U83
Uninstalling the software Upgrading 8
-P-
-V-
Peak pressure inside damage 116 Perform toxic indoors calculation 116 Pipe contraction coefficient 89 Population 22, 121 Presentation settings 87 Pressure damage based on 116 Pressure level total destruction 116 Probabilty FlashAndExplosion 89 Project file 128 Project tree 52 Protection factor clothing 116
Vapour release from vessel or pipe 138 Viewing graphs of the toxicity parameters Vulnerability parameters 116 Vulnerability settings 92
99
-WWater temperature 120 What is RISKCURVES 12 Which input 13 Which results 13 Which tasks 12 World file 107
-QQuick start
8
27
-RReport panel Result panel
76 59
-SSaving your data and checking the function Scenario 22, 123
99
© 2013 TNO
