Study On Emergency Evacuation Algorithm In Fire Environment

Emergency evacuation in fire environment has been a hot topic in the field of public safety,since the emergency evacuation problem has been put forward,researchers at home and abroad have made some achievements and progress.In the initial stage,researchers have proposed many models including the social force model,the potential field model and the queuing theory model.However,these models focus on describing the micro effects between individuals,ignoring the dynamic process and mechanism of evacuation as a whole.In order to make a better analysis from the global point of view,the ant colony algorithm is proposed to solve the path planning problem in fire environment.Based on the traditional social force model and the basic ant colony algorithm theory,the same characteristics of all individual change rules in social force model and the disadvantage that ant colony algorithm is easy to fall into local optimum are improved.Building emotional model for all individuals in the group,the improved social force model is used to describe the micro behavior between individuals,ant colony algorithm is used to describe the overall mechanism of evacuation,the combination strategy of micro model and macro model is studied qualitatively and quantitatively.Main research work in this paper including following two aspects:(1)Aiming at the problem of high density crowd evacuation,a crowd evacuation algorithm based on the improved social force was proposed.Firstly,in the traditional social force model,all individual change rules are the same,and problems are improved,on the basis of the group form,the optimal speed was set to constrain the speed of the members in the group.Secondly,according to the theory of personality,emotion model was established for all individuals,the individual's emotional state was divided into positive emotions and negative emotions,establishing the mechanism of emotion influence through individual influence.Finally,the command strategy was obtained by the individual emotion distribution in the process of crowd evacuation.The simulation results showed that the improved social force model eliminates the turbulence in the narrow channel,individuals accelerating or keeping the critical speed was avoided,and provide a basis for when to join the command strategy.(2)In the fire environment,to solve the problem of easily falling into local optimum in ant colony algorithm,an improved ant colony algorithm is proposed for solving the path planning problem of crowd evacuation in this paper.The improvement of ant colony algorithm is divided into two aspects: first,in the heuristic function of ant colony algorithm,not only the fire product,but also the personnel density factor are considered;second,dynamically adaptively adjust the pheromone intensity,local and global pheromone updating strategy is adopted to update the pheromone on the path,and introduce crossover operation,enhance algorithm escape ability.In the fire environment,the individual emotional difference has a great influence on the path choice,mathematical model of emotion for the individual is established in this paper,individuals with different emotions have different choices of paths.Simulation results show that the planning method proposed in this paper can be used to plan the optimal escape route for individuals with different emotional types,it avoids the local optimum and convergence speed is faster.