Probabilistic Risk Assessment for Evacuation under Building Fire Based on Particle Swarm Optimization Method

摘要

The risk of people under building fire is expressed by particle swarm optimization (PSO) reliability index β and the probability of evacuation is obtained according to β. Random variables affecting the evacuation reliability are taken into account and are described into probabilistic distribution denoted uncertainty of evacuation. The mathematical model for evaluating evacuation based on PSO algorithm is set up. A case analysis is given in this paper to show that this method can perform the optimized calculation of complicated functions with many constraints. PSO can be used to calculate the reliability of occupant evacuation, and quantify the probabilistic risk exposed on the evacuees in the case of fire. Numerical results show that PSO can give a new way to study the evacuation reliability of occupants.