Testing for knowledge: Application of machine learning techniques for prediction of flashover in a 1/5 scale ISO 13784‐1 enclosure

Arjan Dexters, Rolff Ripke Leisted, Ruben Van Coile, Stephen Welch, Grunde Jomaas

Research output: Contribution to journalArticlepeer-review

Abstract

A machine learning algorithm was applied to predict the onset of flashover in archival experiments in a 1/5 scale ISO 13784-1 enclosure constructed with sandwich panels. The experiments were performed to assess whether a small-scale model could provide a better full-scale correlation than the single burning item test.

To predict the binary output a logistic regression model was chosen as machine learning environment. Because results indicated a high variance/low bias issue regularization was applied. It was found that lasso-regression significantly reduced the amount of variance at a negligible increase in bias.

With the regularized model, it was possible to discern the predictive variables and determine the decision boundary. In addition, a methodology was put forward on how to use the decision boundary to iteratively update the learning algorithm. As a result, it was shown how a learning algorithm can be used to facilitate ongoing experimentation. At first as a crude guideline, and in later stages, as an accurate prediction algorithm.

It is foreseen that, by iteratively updating the algorithm, by compiling existing and new experiments in databases, and by applying fire safety knowledge, the final learned algorithm will be able to make accurate predictions for unseen samples and test conditions.
Original languageEnglish
Pages (from-to)708-719
Number of pages12
JournalFire and Materials
Volume45
Issue number6
Early online date27 Jun 2020
DOIs
Publication statusPublished - Oct 2021

Keywords

  • machine learning
  • fire tests
  • sandwich panels
  • fire classification
  • flashover

Fingerprint

Dive into the research topics of 'Testing for knowledge: Application of machine learning techniques for prediction of flashover in a 1/5 scale ISO 13784‐1 enclosure'. Together they form a unique fingerprint.

Cite this