Research On Improvement Of Bionic Algorithm And Its Application In Fire Evacuation Path Planning

The occurrence of indoor fire often causes the loss of people and property,which is related to the complexity of indoor structure and the reasonable degree of evacuation route selection.Most modern buildings are high-rise complex structures which require higher requirements for internal evacuation and external fire rescue.In case of a fire,it is necessary to make real-time escape route planning for the personnel in the fire site by combining fire site information quickly.It is of great significance to scientifically plan evacuation paths for indoor personnel,guide their safe and rapid transfer,and reduce loss of life and property.The traditional path planning is mostly raster map with simple structure and small scale,and its application in practice is limited.At the same time,the traditional pathfinding algorithm is limited by the hardware level of the fire site factors involved in the evacuation process,and the path given lacks a certain degree of rationality.With the application of a variety of indoor sensors in newly built floors in recent years,these factors can be comprehensively considered in the fire escape system.Therefore,for indoor fire evacuation,it is necessary to realize dynamic path planning combined with fire site factors.In this paper,a variety of fire parameters are obtained by software simulation.Then,the path optimization is carried out by combining fire parameters with heuristic function.Finally,an improved bionic algorithm of dynamic path planning for fire evacuation is proposed by comparing many different classical path optimization algorithms.The experimental results show that the fire evacuation algorithm established in this paper can effectively help the people in the fire quickly evacuation.The research in this paper mainly includes the following three aspects:(1)Taking a commercial building in Changchun as the research object,pyrosim software was used to simulate and record various fire parameters generated during the fire spread process,providing data support for subsequent algorithms to consider environmental factors in the process of evacuation path planning.(2)Abstract the rooms,stairs and doors in each floor of the building as nodes in the dual space,establish an indoor road network structure for fire evacuation,and describe the geometric absolute position of the nodes through Cartesian coordinates.Considering the influence of fire factors and road factors on personnel evacuation,temperature,smoke,visibility and geometric distance between road nodes are selected as the key factors for quantitative calculation.The construction of a new heuristic function can avoid the situation that the algorithm only considers the immediate shortest path and ignores the change of fire site in the process of the simulation of path finding.(3)Three commonly used bionic algorithms(ant colony algorithm,A~* algorithm,genetic algorithm)are used for route planning of personnel evacuation in high-rise buildings,and the effectiveness of the improved algorithm in practical application is verified through comparative experiments under the same environmental conditions.The results show that the improved algorithm meets the emergency evacuation requirements under the background of fire,and improves the convergence speed to a certain extent.