Multi-fidelity surrogate algorithm for fire origin determination in compartment fires

Abstract

With the significant growth of the world population, our cities are becoming more and more crowding. In this situation, any fire occurring would cause severe consequences, including property damage and human injuries or even deaths. In assessing the fire cause, the fire origin determination is a crucial step identifying the origin of fire outbreak and the sequential fire and smoke propagation. Traditionally, fire investigators relied upon the visible fire damages at the fire scene to determine the location of fire originated based on their own professional experience. The fire origin determination process is, however, subject to the expert interpretation inherently embedded in the qualitative analyses. Aiming to develop an alternative methodology assisting the fire investigation, we proposed a new Multi-fidelity Kriging algorithm to quantitatively determine the fire origin based on the soot deposition patterns predicted by the numerical simulations. The advantage of the Multi-fidelity Kriging is its capacity in maintaining a reliable accurate prediction with a very limited computational requirement in simulations. The algorithm is tested against a total of 41 different fire origins in a single compartment (i.e. 5 m width × 5 m length × 4 m height) with only 1 doorway for ventilation. The test results demonstrated that the Multi-fidelity Kriging algorithm is capable to predict the fire origin based on the simulated soot deposition pattern while posing significant saving to the computational cost by correcting low-fidelity samples based on knowledge extract on high-fidelity simulation results.