Evacuation Simulation In Crowded Region

The frequent turning up of high buildings, such as shopping malls, large entertainment places and large-scale stadium due to development of economy and society has satisfied multiple material and mental requirements of people, nevertheless these high buildings also posed safety threats. Therefore evacuation simulation, as an effective means to obtain dispersal route, is essential for these high buildings both in safety management and performance design.After comparing the advantages, disadvantages and applications of several chief crowd evacuation models, the thesis analyzed two dominant crowd evacuation models, which were the ones based on cellular automata (CA) and lattice gas automata (LGA).Subsequently the thesis proposed a new model on the basis of particle swarm optimization (PSO), and then analyzed the performance and the advantages of the model. Concerned about the complex external factors, the thesis introduced the definition of artificial potential field (APF), and proposed an innovative initialization method on the basis of vicinity search other than the usual approach of modeling external, which greatly simplified the modeling of external factors.Combined the advantages of particle swarm optimization and cellular automata, the thesis built a new composite crowd evacuation model, in which evacuees updates its velocity with respects to the density detected in the cover of vicinity Moor(er=2). Using model, the thesis analyzed the influence of exit width on evacuation time, the phenomenon of"bow"as well as the influence of obstructions and extra exits and so on .The results turned out to be in accordance with actual phenomena.