Short Term Passenger Flow Forecast of Urban Rail Transit Based on PSO-LSTM

Accurate short-term passenger flow prediction of urban rail transit can effectively alleviate pedestrian congestion, improve passenger comfort, and enable urban rail transit operators to allocate vehicles more reasonably, saving resources. In order to improve the prediction accuracy of LSTM model for short-term passenger flow of urban rail transit, and solve the defects of Long short-term memory (LSTM) model, such as slow Rate of convergence and easy to fall into local optimization, this paper proposes a LSTM model based on particle swarm optimization(PSO), and trains and optimizes the parameters of LSTM model. First, this paper preprocesses the original data by reducing dimensions, Outlier, denoising, normalization, and so on; Secondly, construct a prediction model and use particle swarm optimization algorithm to find the optimal parameter combination for the LSTM model; Finally, compare and verify the optimal model found. The simulation results show that the PSO-LSTM model is significantly superior to the traditional LSTM model. In terms of mean absolute error (MAE) and square root error (RMSE), the PSO-LSTM model is $41.16 \%$ and $50.19 \%$ higher than the traditional LSTM model. In terms of average absolute percentage error (MAPE), the PSO-LSTM model provides a reference for alleviating urban traffic congestion pressure and improving people’s travel safety index compared to the traditional LSTM model of $76.93 \%$. At the same time, it also provides necessary data support for intelligent transportation guidance and reasonable allocation.