Prediction Of The Number Of Dynamic People In Urban Tourist Places Based On Safety Capacity

China is a country with a large population,so the most serious and easy to happen is the stampede accident caused by crowd crowding.In order to solve these problems,this paper takes Gongwangfu scenic spot in Beijing as an example to explore a real-time prediction method of the number of people in the park by using neural network,so that the managers of the scenic spot can clearly understand the congestion situation in the park,so as to carry out safety control in advance and avoid accidents.The distribution regularity of tourists in the scenic spot is poor,and the linear fitting effect is general.Therefore,this paper adopts the neural network with strong fitting ability for nonlinear problems as a solution.Firstly,the collection process of the original data and the data arrangement after the collection of the original data are described.Starting with the combination of the correlation analysis method and the MIV variable optimization method,the original data is screened.Secondly,the Elman neural network model and GA Elman neural network model are established to solve the real-time prediction of the number of tourists in Prince Gong's mansion.The model is optimized in structure and adjusted in parameters.The trained model is trained and used to predict the real-time problems.The errors of the two network structures are 4.58% and 2.41% respectively.Good results have been achieved.Finally,the prediction performance of the two neural network models is compared.GA Elman neural network with faster convergence speed and higher accuracy is selected as the final model.Finally,this paper starts with the maximum instantaneous capacity and the safety psychological capacity,and calculates the Gongwangfu respectively.Based on these two capacities and the prediction of the number of people in the previous paper,it puts forward the measures for the safety classification management and control of Gongwangfu scenic area,and simulates the number of people after the management and control through the neural network.It is found that the prediction of the change of the real-time number of people in the park is conducive to the early growth of the number of people The number of tourists entering the park can be well controlled to effectively avoid accidents caused by crowd congestion.