Advanced

Quality assurance and the simulation of fires - A practical application for automated validation of user-generated input data for Fire Dynamics Simulator

Hermansson, Kristoffer LU (2015) In LUTVDG/TVBB VBRM10 20151
Division of Fire Safety Engineering
Abstract
In order to systematize and rationalize the quality control process for FDS (Fire Dynamics Simulator) input data a software application for this specific purpose has been developed. In order to study the occurrence of user related errors in FDS input data and the possibility of reducing such errors by using the developed application, a case study on archived FDS input data from actual building projects has also been performed. Controls, which should be part of the quality control process, were identified by performing a literature review—covering building code requirements as well as recommendations from different stakeholders—and implemented in the application. FDS input data was collected in cooperation with Swedish consultants and... (More)
In order to systematize and rationalize the quality control process for FDS (Fire Dynamics Simulator) input data a software application for this specific purpose has been developed. In order to study the occurrence of user related errors in FDS input data and the possibility of reducing such errors by using the developed application, a case study on archived FDS input data from actual building projects has also been performed. Controls, which should be part of the quality control process, were identified by performing a literature review—covering building code requirements as well as recommendations from different stakeholders—and implemented in the application. FDS input data was collected in cooperation with Swedish consultants and evaluated using the application. Results have been presented in the form of checklists containing controls which should be performed as part of the quality control process, in the form of the actual application, and in the form of a summary of the frequencies with which different errors occurred in the studied FDS input data. The most frequently occurring errors were either of the kind where an intended parameter value was not correctly specified or closely related to the model mesh. It has been concluded that the application is a useful tool in discovering errors in FDS input data, thereby providing a means of reducing their occurrence. (Less)
Popular Abstract
Fire simulations are often applied in performing different sorts of risk assessments, for example when evaluating the fire safety of buildings. The models used to perform these simulations are advanced and place high requirements on the users. A software application designed to aid the quality assurance process related to the use of one such model has been developed and evaluated by studying simulations from actual building projects.

FDS (Fire Dynamics Simulator) is a model for simulating fires, which is often used in the context of performing different sorts of risk assessments. One common application is in the verification of compliance with governmental requirements in the design of fire protection for buildings. Simulations of fires... (More)
Fire simulations are often applied in performing different sorts of risk assessments, for example when evaluating the fire safety of buildings. The models used to perform these simulations are advanced and place high requirements on the users. A software application designed to aid the quality assurance process related to the use of one such model has been developed and evaluated by studying simulations from actual building projects.

FDS (Fire Dynamics Simulator) is a model for simulating fires, which is often used in the context of performing different sorts of risk assessments. One common application is in the verification of compliance with governmental requirements in the design of fire protection for buildings. Simulations of fires in FDS are based on user-generated input data in the form of text files, which are crucial to the quality of the results obtained by running the simulations. For this reason, undiscovered errors in input data present a problem in that the risk assessments may be based on incorrect information.

In this project, for the purpose of providing better foundations for risk assessments where the results of FDS simulations constitute the decision basis, the process for performing quality controls of user-generated FDS input data has been further developed. A software application designed to automate parts of the quality control process has been developed and evaluated by studying the occurrence of user related errors in FDS input data.

Results have been presented in the form of checklists, one intended to be checked manually by the user and one intended to be checked using the developed software application. Further, the actual application has been produced, published (available at beta.kristofferhermansson.se), and evaluated. The evaluation was performed by studying a set of input data from actual building projects. Notably, the application was capable of identifying errors in all of the input files. Hence, it has been concluded that it is a useful tool in discovering such errors, thereby also providing a means of reducing their occurrence. The actual frequencies with which different errors occurred in the study have also been presented, and while these results might not be generally representative, they could serve as an indication of certain areas which should be handled with care in producing FDS input data. The errors most frequently occurring in the study were either of the kind where an intended parameter value was not correctly specified or closely related to the mesh.

The main objectives of the project were to determine to which extent the quality control process for FDS input data can be simplified by using an automated software application and to determine whether the occurrence of user related errors can be reduced by using such an application. The project was performed in three main steps. First, a literature review was performed to determine which checks should be part of the quality control process. Second, an evaluation of whether each respective identified check could be automated and performed the application was made and, in case deemed possible, implemented. Third, a case study where the application was used to evaluate input data from actual building projects was performed, in order to determine whether the application could be used to reduce the number of errors in input data and to determine which errors are most commonly occurring. (Less)
Please use this url to cite or link to this publication:
author
Hermansson, Kristoffer LU
supervisor
organization
course
VBRM10 20151
year
type
H2 - Master's Degree (Two Years)
subject
keywords
FDS, Fire Dynamics Simulator, fire safety, fire simulation, quality assurance, automation
publication/series
LUTVDG/TVBB
report number
5501
language
English
id
5463455
date added to LUP
2015-06-04 10:53:51
date last changed
2015-06-04 10:53:51
@misc{5463455,
  abstract     = {In order to systematize and rationalize the quality control process for FDS (Fire Dynamics Simulator) input data a software application for this specific purpose has been developed. In order to study the occurrence of user related errors in FDS input data and the possibility of reducing such errors by using the developed application, a case study on archived FDS input data from actual building projects has also been performed. Controls, which should be part of the quality control process, were identified by performing a literature review—covering building code requirements as well as recommendations from different stakeholders—and implemented in the application. FDS input data was collected in cooperation with Swedish consultants and evaluated using the application. Results have been presented in the form of checklists containing controls which should be performed as part of the quality control process, in the form of the actual application, and in the form of a summary of the frequencies with which different errors occurred in the studied FDS input data. The most frequently occurring errors were either of the kind where an intended parameter value was not correctly specified or closely related to the model mesh. It has been concluded that the application is a useful tool in discovering errors in FDS input data, thereby providing a means of reducing their occurrence.},
  author       = {Hermansson, Kristoffer},
  keyword      = {FDS,Fire Dynamics Simulator,fire safety,fire simulation,quality assurance,automation},
  language     = {eng},
  note         = {Student Paper},
  series       = {LUTVDG/TVBB},
  title        = {Quality assurance and the simulation of fires - A practical application for automated validation of user-generated input data for Fire Dynamics Simulator},
  year         = {2015},
}