Travel Time Prediction Method for Urban Expressway Link Based on Artificial Neural Network

Abstract

According to the floating-car data measured from urban links, some data-processing techniques including data mending, wavelet de-noise and others are used to establish a time series of data to better reflect the original running characteristic of urban links. On this basis, the travel time forecasting researches are executed both by the BP neural network based on Bayesian Regularization algorithm and the genetic algorithm based on BP network. In this period, several prediction schemes are designed according to different network architecture and sample data. What’s more, the validity evaluation and the results contrast are performed. The experiments prove that the genetic algorithm based on BP artificial neural network is more practical and can improve the precision better.