Short-term passenger flow forecast of urban rail transit based on PSO-VMD-LSTM model
ZHANG Wanning
ZHENG Mingming
LIU Yan
Abstract：In order to reduce the interference of noise on the passenger flow prediction model,the particle swarm optimization(PSO)algorithm is used to determine the parameters of variational mode decomposition(VMD).The original passenger flow sequence is denoised using VMD,and the passenger flow data is decomposed into intrinsic mode functions(IMF)and residuals at different time scales.The Bayesian optimization(BO)algorithm is used to determine the hyperparameters of the long short term memory(LSTM)neural network,and the PSO-VMD-LSTM passenger flow prediction model is constructed.Taking the passenger flow data of Shapingba Station of Chongqing Metro Line 1 as an example,the prediction accuracy of the model is verified.The results show that compared with back propagation(BP)neural network,radial basis function(RBF)neural network,and LSTM neural network,the PSO-VMD-LSTM model reduces the root mean square error by 268.03,204.41,and 221.66,and reduces the mean absolute percentage error by 13.16%,10.21%,and 11.06%respectively.The PSO-VMD-LSTM model has high applicability and prediction accuracy for short-term passenger flow prediction in urban rail transit.
Keywords：passenger flow predictionPSO algorithmVMDBO algorithmLSTM neural network
Publication Date：2023-12-30
Online Publishing Date：2026-05-22(First online date of this platform, not the publication date of the document)
Pages：8( 43-50 )