Research On The Evacuation Model And Algorithm Based On The Group-based Post-earthquake

Recently,earthquake disasters have occurred frequently around the world,representing serious threats to human life and property safety.After the earthquake,what is essential to reduce the damage caused by earthquakes is to ensure the safe evacuation of people in the building.Therefore the corresponding emergency evacuation schemes that are effective measures to implement crowd evacuation after earthquake are carried out in advance by analyzing the key factors of the behavior characteristics of the evacuated people,the force of evacuated personnel and the optimal route of evacuation,which has practical significance.A large amount of research has been conducted on emergency evacuation by domestic and foreign scholars.But regarding to the post-earthquake evacuation,the following problems still require to be further solved and improved:(1)lacking of in-depth exploration of the group behavior in group evacuation;(2)having few related studies on crowd evacuation after earthquake in multi-story and high-rise buildings;(3)further improving the planning of optimal evacuation paths in group evacuation.Targeting at the post-earthquake safe crowd evacuation in multi-story and high-rise building and the above existing problems of safety evacuation,this paper gives its contributions mainly from the following three aspects:(1)Aiming at the traits of temporary group evacuation,"herd behavior" is put forward in the post-earthquake evacuation."Herd behavior" refers to a movement phenomenon in which individual behavior is affected by the behavior of the surrounding people in an emergency,and thus follows others' behaviors.In the post-earthquake group evacuation,family,friends,neighbors and other closely related people will form a group phenomenon in accordance with the degree of intimacy,which is called "herd behavior".The closer ties among people will have higher degree of aggregation within the group.Moreover,herd behavior in the evacuated crowd is presented throughout the whole evacuation.Therefore,in order to reflect the grouping phenomenon of the crowd more accurately during the evacuation,a special herd behavior existed in the post-earthquake evacuation is proposed in this paper.What's more,the interaction effects within pedestrians and between pedestrians and the leader are fully considered and two kinds of simulation behavior models-chasing and following are employed to describe typical behavioral characteristics and behavioral rules in the post-earthquake evacuation.(2)Social force model fusing herd behavior is put forward aiming at the problem of post-earthquake evacuation in multi-story and high-rise buildings.Based on the fact that the complicated construction of the multi-and high-rise buildings and the intensive personnel within them give rise to more complex crowd evacuation,social force model fusing herd behavior is put forward,which is proposed to simulate the crowd evacuation in the multi-story and high-rise buildings after the earthquake.In this model,"herd" forces,composed of the visual,cohesive,and repulsive forces among the evacuated pedestrians,are added to simulate interbehaviors that are produced between pedestrians and other pedestrians in the surrounding environment,which can more realistically reflect the herd behavior in the post-earthquake crowd evacuation in the multi-story and high-rise buildings.The experimental results show that this social force model fusing herd behavior can effectively evaluate the impact of factors such as group size,group relationship,export volume and export size on the evacuation of post-earthquake,thus can more efficiently and authentically simulate the evacuation movement of the crowd after the earthquake in the multi-story and high-rise buildings.(3)Optimization algorithm of group evacuation path fusing herd behavior and firefly algorithm is introduced according to the planning of the optimal evacuation path in post-earthquake evacuation.The planning of the optimal evacuation path refers to finding an optimal path from the starting point to the evacuation exit in a complex environment.The traditional firefly algorithm has the disadvantages of slow convergence speed and low solution precision when solving problems of the optimization path.In view of the above deficiencies,improved optimization algorithm of group evacuation path is proposed in combine with the characteristics of post-earthquake evacuation.First,a relative attraction mechanism is presented to form multiple small-scale evacuation teams in the initial stage of evacuation.Then,the information sharing mechanism among the evacuation teams is established in the evacuation,and each evacuation group is continuously adjusted or reorganized according to the relative attractiveness,so as to more veritably simulate the potential impact of the evacuation group behavior on the evacuation route selection in the post-earthquake.The experimental results show that the improved firefly algorithm fusing herd behavior can effectively reduce the overall evacuation time,improve the authenticity of the group evacuation simulation,and solve the planning of the optimal path problem in the post-earthquake evacuation.