Research on Short-term OD Passenger Flow Forecast of Rail Transit Based on Optimized PSO-BP Algorithm
To achieve better management of urban transportation systems,it is necessary to accurately predict the passenger flow of ur-ban rail transit inbound.In order to achieve the purpose of improving rail transit efficiency and improving the quality of operation serv-ices,a back propagation(BP)neural network is constructed to predict the subway passenger flow;using the particle swarm optimiza-tion algorithm(PSO)to further optimize the BP neural network to form a subway passenger flow prediction system that considers the influence of complex factors.Based on the data of Xi'an Metro Line 1,this paper qualitatively analyzes the time-space correlation of station OD passenger flow,uses regression analysis to quantitatively analyze the factors affecting passenger flow,and screens out the three time characteristics of weather,holidays,and operating hours.In order to improve the prediction accuracy,this paper constructs the BP neural network model under different time periods,optimizes the prediction error of the PSO-BP neural network model,and forms a rail transit short-term OD passenger flow prediction model based on the PSO-BP neural network,adding time characteristics of the short-term OD passenger flow forecasting model.The PSO-BP neural network model prediction value M1 decreased by 48.2%on average,M2 decreased by 37.6%,and M3 decreased by 21.9%.This method and the model provides more accurate data for rail tran-sit operation departments to formulate train operation plans.
万方数据
维普
