Research On Emergency Evacuation Modeling And Path Planning Based On Artificial Bee Colony Algorithm

With the improvement of science and technology and the development of social economy,the demand of people's lives is growing.To meet the demand of people's lives,the number of large public places is increasing.Although this situation provides great convenience to people's life,at the same time,there are also great risks and hidden dangers in the accumulation of a large numbers of people.In recent years,people have conducted a lot of in-depth research on the subject of emergency evacuation to solve this problem.How to choose an appropriate evacuation route is the focus of this paper.In this paper,the cellular automata model was used to construct the evacuation environment,and the evacuation model based on artificial swarm algorithm was studied in order to optimize the evacuation route of personnel,reduce the total evacuation time and minimize the risk of personnel.The main job of this paper is as follows:(1)In order to establish the evacuation model quickly and efficiently,this paper constructed an evacuation simulation model based on cellular automata,and used the artificial bee colony algorithm to simulate the evacuation process.The comparison experiment with the traditional particle swarm algorithm proves that the evacuation simulation model based on cellular automata has shorter evacuation time and higher evacuation efficiency.In addition,the results of obstacle setting experiment show that evacuation of the model is susceptible to the adverse effects of multi-obstacle environment.(2)Considering the blindness of selection of onlooker bees to employed bees and the adverse effects of multi-obstacle evacuation environment on the evacuation,this paper established a vision-guided evacuation model.This model was based on the basic artificial bee colony algorithm.By introducing the onlooker bee's field of vision,we selected the best individual as the employed bee in the onlooker bee's field of vision,so as to improve the selection of employed bee.Moreover,this model introduced the number of people causing congestion,repulsion of obstacles and repulsion of other individuals into the fitness of the model to reduce the dynamic impact of evacuation environment.According to the experiment of parameter setting and the experiment of algorithm comparison,it can be concluded that the vision-guided evacuation model has stronger capability to gather towards the exit in the later stage of evacuation process than particle swarm optimization and artificial bee colony algorithm,under the condition of proper parameters' setting.(3)For improving the evacuation path of individuals,we proposed an optimization model of artificial bee colony algorithm under the influence of multiple factors.On the basis of the vision-guided evacuation model,this model was combined with the foraging formula of particle swarm optimization.The flight of the onlooker bee was influenced by the best bee in the onlooker bee's field of vision,the best historical bee,the best global bee and the inertial employed bee.Through the experiment of exit setting and multi-algorithm multi-level experiment,it is found that the optimization model based on multi-factor is sensitive to the exit setting.Compared with particle swarm optimization,basic swarm optimization and other algorithms,this optimization model has great advantages in evacuation path and evacuation time.