Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

A COMPARATIVE STUDY OF TRAVELLING FIRE IN LARGE SPACES WITH MULTI-ZONE FIRE MODEL AND FDS

Bertoli, Paolo LU (2021) In LUTVDG/TVBB VBRM05 20211
Division of Fire Safety Engineering
Abstract
A knowledge of far field temperatures is valuable for Performance Based Design structural analysis and life safety analysis, where a parametric study is needed. A fire model using multiple zones called a multi zone (MZ) fire model has in previous studies shown good potential as an analytical tool for predicting far field temperatures in large compartments. This thesis aims to build upon the existing state of knowledge for application of MZ fire model, as a predictive tool for far field temperatures. The research approach primarily involves designing and testing several large scale travelling fire scenarios in MZ fire model and comparing these results to Fire Dynamic Simulator (FDS). As a further step, a comparison is made between MZ fire... (More)
A knowledge of far field temperatures is valuable for Performance Based Design structural analysis and life safety analysis, where a parametric study is needed. A fire model using multiple zones called a multi zone (MZ) fire model has in previous studies shown good potential as an analytical tool for predicting far field temperatures in large compartments. This thesis aims to build upon the existing state of knowledge for application of MZ fire model, as a predictive tool for far field temperatures. The research approach primarily involves designing and testing several large scale travelling fire scenarios in MZ fire model and comparing these results to Fire Dynamic Simulator (FDS). As a further step, a comparison is made between MZ fire model predictions for two well documented large scale travelling fire experiments: the Tisova full scale fire test and the Edinburgh Tall building Fire Test using MZ fire model. Results of this thesis are mixed. Comparison of far field temperature values show reasonably good agreement between MZ fire model and FDS computational fluid dynamics (CFD) models (on average 11
22% differences). However, comparison of MZ fire model temperatures to the measured temperatures of the two travelling fire experiments yielded significant differences (on average 7 35 % between the models, and 37 101% between MZ fire model and the experiments). The substantial difference between analytical and experimental results indicates that MZ fire model and FDS were not successful at predicting far field temperatures for the actual experimental fires. This can be caused by a combination of parameter and model uncertainties, and a lack of reliable data from the experiments. This research effort shows that MZ fire model has the potential to be a useful tool for predicting far field temperatures in large compartment fires; however, further research and experimentation are needed to refine modelling methods and techniques before MZ fire model is a viable engineering tool for fire safety analysis. (Less)
Popular Abstract
MULTI-ZONE FIRE MODEL FOR FIRES IN LARGE BUILDINGS

Fire safety engineering often relies on modelling to predict temperatures in a compartment. Fire safety engineers use this information to design fire safety measures for structural integrity and safe evacuation of occupants. For a small compartment, such as an apartment, there are several generally accepted methods for estimating temperatures. These methods range from simple hand calculations to well-developed and validated CFD models. For large open buildings, such as a supermarket or a large open-space office, these traditional methods become either impractical or computationally too expensive due to the large domain that needs to be simulated. It can take weeks (or even months) to... (More)
MULTI-ZONE FIRE MODEL FOR FIRES IN LARGE BUILDINGS

Fire safety engineering often relies on modelling to predict temperatures in a compartment. Fire safety engineers use this information to design fire safety measures for structural integrity and safe evacuation of occupants. For a small compartment, such as an apartment, there are several generally accepted methods for estimating temperatures. These methods range from simple hand calculations to well-developed and validated CFD models. For large open buildings, such as a supermarket or a large open-space office, these traditional methods become either impractical or computationally too expensive due to the large domain that needs to be simulated. It can take weeks (or even months) to perform a calculation of temperatures for a large open space using a high-speed, multicore workstation or supercomputer. Consequently, it is highly desirable to find a reliable alternative method that can perform these calculations faster (on the order of minutes or hours) using a typical personal computer.
Lund University is researching a promising alternative method based on use of a Multi-Zone (MZ) fire model. As the name suggests, the model divides the space into the multiple zones to represent the smoke and heat movement in a large open building. MZ fire models can potentially become a viable tool for efficiently predicting far-field temperatures with reasonable accuracy. The MZ fire model method has been previously tested for smoke filling scenarios on a few large buildings and the method showed good results. Research undertaken by this project has extended the state of knowledge on application of MZ fire models by investigating model predictions for extensive fires in spaces significantly larger than those used in the previous fire experiments. The results showed promising temperature predictions for well-defined fire scenarios.
Live fire experiments for large spaces are scarce and difficult to model. Unfortunately, the limited information and data for these experiments does not allow accurate and complete fire model development. Therefore, two fictive scenarios were designed to represent a large open space office and a supermarket. These cases had no experimental support, so they were compared to another well-established and tested fire modelling software, Fire Dynamics Simulator (FDS). Additionally, two smaller experimental cases were also compared to MZ fire model and FDS.
Compared to previously reported results, this research for MZ fire models has shown higher differences in predicted temperatures compared to FDS. This outcome is attributed to the much larger scenarios and larger fires, as well as the fact that the fire was modelled as a travelling fire. The temperature results compared to the experiments showed large differences due to the overall difficulties in representing accurately the experiments in the models, along with the many uncertainties.
This research has shown that MZ fire models can potentially be used to predict far field temperatures in large buildings, but further research and experiments are needed to refine the model and its methods before it is ready for real world applications. Nonetheless, the potential benefits of using MZ fire models for large open spaces is readily apparent. (Less)
Please use this url to cite or link to this publication:
author
Bertoli, Paolo LU
supervisor
organization
course
VBRM05 20211
year
type
H2 - Master's Degree (Two Years)
subject
keywords
Multi-Zone fire model, FDS, Large enclosures, Travelling fire.
publication/series
LUTVDG/TVBB
report number
5643
other publication id
LUTVDG/TVBB—5643--SE
language
English
id
9049213
date added to LUP
2021-06-04 12:17:55
date last changed
2021-06-04 12:17:55
@misc{9049213,
  abstract     = {{A knowledge of far field temperatures is valuable for Performance Based Design structural analysis and life safety analysis, where a parametric study is needed. A fire model using multiple zones called a multi zone (MZ) fire model has in previous studies shown good potential as an analytical tool for predicting far field temperatures in large compartments. This thesis aims to build upon the existing state of knowledge for application of MZ fire model, as a predictive tool for far field temperatures. The research approach primarily involves designing and testing several large scale travelling fire scenarios in MZ fire model and comparing these results to Fire Dynamic Simulator (FDS). As a further step, a comparison is made between MZ fire model predictions for two well documented large scale travelling fire experiments: the Tisova full scale fire test and the Edinburgh Tall building Fire Test using MZ fire model. Results of this thesis are mixed. Comparison of far field temperature values show reasonably good agreement between MZ fire model and FDS computational fluid dynamics (CFD) models (on average 11
22% differences). However, comparison of MZ fire model temperatures to the measured temperatures of the two travelling fire experiments yielded significant differences (on average 7 35 % between the models, and 37 101% between MZ fire model and the experiments). The substantial difference between analytical and experimental results indicates that MZ fire model and FDS were not successful at predicting far field temperatures for the actual experimental fires. This can be caused by a combination of parameter and model uncertainties, and a lack of reliable data from the experiments. This research effort shows that MZ fire model has the potential to be a useful tool for predicting far field temperatures in large compartment fires; however, further research and experimentation are needed to refine modelling methods and techniques before MZ fire model is a viable engineering tool for fire safety analysis.}},
  author       = {{Bertoli, Paolo}},
  language     = {{eng}},
  note         = {{Student Paper}},
  series       = {{LUTVDG/TVBB}},
  title        = {{A COMPARATIVE STUDY OF TRAVELLING FIRE IN LARGE SPACES WITH MULTI-ZONE FIRE MODEL AND FDS}},
  year         = {{2021}},
}