Prediction of urban passenger transport based-on wavelet SVM with quantum-inspired evolutionary algorithm

Based on least squares wavelet support vector machines (LS-WSVM) with quantum-inspired evolutionary algorithm (QEA), the prediction model of urban passenger transport is proposed , that can provide the theoretical foundation of forecasting passenger volume of urban transport accurately. The prediction model of urban passenger transport is established by using LS-WSVM, whose regularization parameter and kernel parameter are adjusted using quantum-inspired evolutionary algorithm. QEA with quantum chromosome and quantum mutation has better global search capacity. The parameters of LS-WSVM can be adjusted using quantum-inspired evolutionary optimization. Combining with the data of the urban volume of passenger transport of Xipsilaan over years, the prediction model of urban passenger transport is validated, the simulation results indicate that the prediction model is effective, and based on LS-WSVM has more improvement than LS-SVM with Gaussian kernel in predicting precision, and then the improved LS-WSVM with QEA is efficient than with cross-validation method for tuning parameters.

I. Introduction

The prediction of urban passenger transport is an important aspect of intellectual traffic system (ITS), with the development of economy and the flowing amount of urban population becomes more and more, people bring forward the higher requirements about city transportation. So, predicting the urban passenger transport and its development trend can offer the important theoretical foundation for city transportation, the development plan of highway passenger transport and the station site of the urban transportation. The prediction of urban passenger transport is a non-linear question, and it is fuzzy, so it is difficult to predict. The question is often very complicated during practical prediction, and some study has already been done by the domestic and international scholar in this aspect. Shengchang Wang[1] predicted the urban passenger transport of highway of one city using the grey model, exponential smoothing model, regression analysis, elastic coefficient law and other kinds of prediction models as the analytic approach. Xuewei Li[2] forcasted the transporting amount of the intellectual traffic with the genetic algorithm. Chiyu Li[3] used the artificial neural network to carryon the prediction of transporting amount of the traffic.