Study On The Method And Application Of Short Term Passenger Flow Forecast Of Urban Rail Transit

Since the reform and opening up 30 years,the domestic economy is growing rapidly,the urban construction speed,the size of the city,the number of motor vehicles,traffic congestion,environmental pollution and high energy consumption,which has become the important means to solve the problems of urban rail transit,and has a very important and far-reaching impact on the development of urban rail transit network planning,investment,construction and operation management.Short-term passenger flow forecast is an important part of urban rail transit operation and management,and is also an important basis for urban rail transportation planning and organization.In order to carry out short-term passenger flow forecast work,this paper firstly analyzes the impact factors,which affect passenger flow in urban rail transit,the spatial distribution of urban rail transit,the time distribution of urban rail transit,the supply capacity of urban rail transit,and the supply capacity of urban rail transit.The traffic demand of urban rail transit passenger flow,as well as the supply capacity of urban rail transit,provide the basis for the exploration of short-term passenger flow forecast.Subsequently,this paper takes the Chongqing light rail as the research object,and analyzes the characteristics of Chongqing’s light rail passenger flow,such as the trend,the periodicity and the abrupt change.Subsequently,this paper analyzes the characteristics of Chongqing light rail transit,such as the trend of Chongqing light rail transit,the periodicity and the characteristics of the mutation,and then uses the regression,time series,support vector machine and neural network analysis method to compare the fitting results of these data.Based on the above analysis results,weighted method,BP neural network method and random forest method were used to predict the results of the above analysis.It is found that the combined forecasting method is better than the single forecasting method,and the combination forecasting is better than other forecasting methods.