Rail transit short-time passenger flow volume prediction method based on W-BiLSTM

Abstract

The invention provides a rail transit short-time passenger flow volume prediction method based on W-BiLSTM. The method comprises the following steps: acquiring time sequence historical data of urban rail transit passenger flow volume as sample data; preprocessing and normalizing the sample data; performing wavelet decomposition and single-branch reconstruction on the sample data through a wavelet neural network to obtain training data and test data; initializing the BiLSTM neural network model, setting a mechanism and hyper-parameters of the BiLSTM neural network model, and inputting training data to construct and train a prediction model; when an expected error or a preset number of iterations is reached, selecting an optimal BiLSTM neural network model to predict the test data to obtain a predicted value; analyzing a predicted value error by taking a root-mean-square error and an average absolute percentage error as evaluation indexes; capturing the short-time passenger flow change rule of the rail transit to accurately predicting the speed of the urban road in the future. The invention can be applied to intelligent traffic and smart city construction. And data support is provided for avoiding travel congestion and guaranteeing the travel safety and efficiency of residents.

Description

technical field [0001] The present application relates to the field of rail transit short-term passenger flow forecasting, in particular to a W-BiLSTM-based rail transit short-term passenger flow forecasting method. Background technique [0002] Accurately grasping the demand for urban public transportation is the key point to realize smart city management. For cities with high population density, rail transit in urban public transportation is the main line of urban public transportation, which undertakes a large number of trips during peak hours. need. The short-term increase in passenger flow will cause excessive pressure on rail transit and bring great difficulties to rail transit operation scheduling and management. Therefore, short-term prediction of rail transit passenger flow is crucial to ensuring the rapid operation of intelligent transportation systems. important role. [0003] The existing methods used in short-term traffic flow prediction include historical ave...

AI Technical Summary

Problems solved by technology

[0004] This application provides a W-BiLSTM-based short-term passenger flow prediction method for rail transit to solve the technical problem that the abnormal passenger flow of rail transit is affected by many environments and it is difficult to accurately predict the passenger flow

Images