People’s Evacuating Research Based On Modified Particle Swarm Optimization Algorithm

With the rapid development of economy in our country, urban construction follows, many buildings have sprung up on many places. While pursuing the large-scale and complication of the building, people are also undertake a few uncertain safeties. All kinds of emergencies have taken place frequently than ever before, the emergencies such as fire, earthquake, flood, have always been a human insuperable barrier, once they happened, they would not only lead to major national property loss but also pose a serious threat to public safety.It is important for us to study emergency evacuation in crowded places, because it takes not only the density, design of buildings, but also physical conditions of the people and the reasonability of the evacuation plan into consideration. Under such conditions, it will bring tremendous challenges to the evacuation. Throughout large casualties accident both at home and abroad in recent years, almost all are closely related to the staff behavior and evacuation time, so how to effectively prevent and reduce casualties under the emergency, especially prevent group death and group injury of accidents, has become a priority of the public safety. There are still inevitably problems falling into local minimum or stagnation in the course of evacuation study. What we should do is to further search the key factors which affect the emergency evacuation so as to improve the efficiency of the emergency evacuation and reduce evacuation time.In this thesis, the research background, significance and the evacuation problem research status both at home and abroad are introduced, according to a sudden and severe fire disaster. Moreover, some factors which influence the evacuation behaviors are analyzed, and the related algorithms which solve the evacuation problem are put forward. Finally, a related model is established based on the above analysis. Through deeply study the particle swarm optimization(PSO) and analyze the strengths and weaknesses between PSO and other optimization algorithms, a modified PSO algorithm based on the improvement of the parameters is proposed. Through many trials, the optimal solution is achieved, and the modified PSO model is established. The above modified model is used to simulate the evacuation in emergency, in order to make sure the more convenient and reasonable escape routes. The results show that the modified PSO has better stability and faster convergence speed.