Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Investigating machine learning for fire sciences: literature review and examples
RISE Research Institutes of Sweden, Safety and Transport, Fire Technology.ORCID iD: 0000-0001-7524-0314
Bengt Dahlgren, Sweden.
Brandskyddslaget, Sweden.
RISE Research Institutes of Sweden, Safety and Transport, Fire Technology. (Brandforskning)ORCID iD: 0000-0002-0380-9548
2021 (English)Report (Other academic)
Abstract [en]

In this work, a review of current literature on artificial intelligence (AI) and more specifically machine learning (ML) is presented. ML is illustrated by two case studies where artificial neural networks are used for regression analysis of 110 spalling experiments and 81 Fire Dynamics Simulator (FDS) models of tunnel fires. Tunnel fires are often assessed by fire safety engineers using time-consuming simulation tools where a trained model has the potential to significantly reduce time and cost of these assessments.

A regression model based on a neural net is used to study small scale spalling experiments and similar accuracy compared to least-square fits are obtained. The result is a function based on 14 determining experimental parameters of spalling and result in, spalling times and depths. It is a relatively small effort to get started and set up models, comparably to regular curve fitting. In this first case study the training times are short, it is thus possible to establish how the model performs on average.

The 81 tunnel fire simulations are trained using a similar neural net however it takes considerable time to organize data, creating input, target data of the desired format and training. Here, it is also crucial to normalize the data in order to have it in a suitable format when training. 

It should be noted that ML is often an iterative process in such a way that it may be difficult to know what settings will work before starting the process. It is equally important to illustrate and get to know the data, e.g., if there are large differences or orders of magnitude differences in the data. A normalization procedure is most often practical and will give better predictions.

Place, publisher, year, edition, pages
Borås, 2021. , p. 36
Series
RISE Rapport ; 2021:59
Keywords [en]
Machine learning, Fire Dynamics Simulator, Fire spalling
National Category
Other Civil Engineering Probability Theory and Statistics Other Computer and Information Science
Identifiers
URN: urn:nbn:se:ri:diva-53512ISBN: 978-91-89385-49-8 (electronic)OAI: oai:DiVA.org:ri-53512DiVA, id: diva2:1566827
Funder
Brandforsk, 320-006Available from: 2021-06-15 Created: 2021-06-15 Last updated: 2023-06-07Bibliographically approved

Open Access in DiVA

fulltext(1216 kB)554 downloads
File information
File name FULLTEXT02.pdfFile size 1216 kBChecksum SHA-512
cc971ec0ba398d43d82bbde370349e3448a55c8c1f607be8e74b0427fb1cd818e7eb63225ffc126aa795cdda8457447ff3f6b1b01d2754ed53e82bc538b21c01
Type fulltextMimetype application/pdf

Authority records

Anderson, JohanMcNamee, Robert

Search in DiVA

By author/editor
Anderson, JohanMcNamee, Robert
By organisation
Fire Technology
Other Civil EngineeringProbability Theory and StatisticsOther Computer and Information Science

Search outside of DiVA

GoogleGoogle Scholar
Total: 582 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

isbn
urn-nbn

Altmetric score

isbn
urn-nbn
Total: 2185 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf