Research And Application Of Ant Colony Optimization In Underground Commercial Street Fire Evacuation System

In recent years,the rapid development of commercial economy,more and more people choose to go to shopping malls for consumption and entertainment,at the same time,the safety problems of commercial buildings are also increasingly prominent.Due to the relatively closed underground space,especially the crowded underground commercial street,in the event of a fire,it is easy to cause serious accidents such as casualties if people cannot be evacuated in time.Therefore,in order to improve the escape efficiency of personnel and reduce property losses,this thesis designs and implements an underground commercial street fire evacuation system based on ant colony optimization.The main research contents are as follows:First of all,this thesis selects wireless sensor networks(WSN)technology to monitor the occurrence of fire in the underground commercial street,so as to establish a fire evacuation system to guide people to escape.It briefly introduces the basic knowledge of WSN and the heuristic function,pheromone update and state transition probability formula of ant colony algorithm.Secondly,this thesis proposes an ant colony optimization(ACO)for the problems that the convergence speed of ant colony algorithm is slow and easy to fall into local optimum.The algorithm takes into account various factors affecting the escape of people during the fire,such as the distance between the current location and the escape exit,carbon monoxide concentration to improve the heuristic function formula of ant colony algorithm;By combining the basic Max-Min Ant System(MMAS)model to optimize the pheromone update rules of the ant colony algorithm,allocate the amount of pheromone on each path according to a number of different factors,guide ants to choose the path that is better in comprehensive aspects,so as to improve the convergence speed;this thesis proposes dead zone impact factor to improve the state transition probability formula of ant colony algorithm,and accelerate the success rate of ants searching for the optimal path by reducing the probability of ants entering the dead zone.After improvement,the algorithm has achieved a certain improvement in convergence speed and prevented ants from entering local optima.Finally,the ant colony optimization proposed is applied to the underground commercial street fire evacuation system and designs functional modules in detail to achieve safer and more efficient evacuation effects.To better develop the fire evacuation system,this thesis selects Python as the development language of the system,and uses bootstrap framework,Django framework and My SQL database to develop the interface and functional modules of the web side.In addition,this thesis also developed the interface and functional modules for cell phone so that users can view the optimal escape route through cell phone.Through communication and system function testing,this system can quickly plan a reasonable escape route and has certain application value.