Font Size: a A A

Traffic Prediction Based On Neural Network Technology Research

Posted on:2007-08-03Degree:MasterType:Thesis
Country:ChinaCandidate:W W MengFull Text:PDF
GTID:2192360185991291Subject:Traffic Information Engineering & Control
Abstract/Summary:PDF Full Text Request
Along with the rapid development of the transportation infrastructure construction and intelligent transportation system (ITS) of our country, traffic planning and traffic inducing became the research focus of traffic field day by day, which realized with the important basis of the accurate traffic volume forecasting. Therefore people paid more and more attention to the traffic volume forecasting.The main work of this thesis was applying the advanced artificial intelligence technology such as artificial neural network and genetic arithmetic and so on to the traffic volume forecasting. It gave a useful attempt to enrich the methods of the traffic volume forecasting. The main content was composed with the following parts:(1) The development about artificial neural network was introduced, both the keystone and the correlation theories about artificial neural network were studid, and the keystone and arithmetic about the BP neural network were emphatically analyzed.(2) The forecasting methods for both the long-term traffic volume and the short-term traffic volume were summarized, and the forecasting mechanism and the shortages of them were analyzed; the arithmetic flow about BP network in traffic volume forecasting was presented, the idiographic forecasting methods for both the long-term traffic volume and the short-term traffic volume based on BP network were emphatically studied.(3) The modeling process of the traffic volume forecasting model based on BP network was analyzed at length, and the problems such as choice about the node number of hidden layer for BP network, data pretreatment, and so on were discussed; in allusion to the shortages of BP arithmetic such as easy to fall into local smallness and slow constringency speed, we brought forward the amelioration measures. After the above foundation, the traffic volume forecasting model for the long-term traffic volume and the short-term traffic volume based on BP network were constituted separately, they were applied by experiments and achieved perfect results.
Keywords/Search Tags:Long-term traffic volume, Short-term traffic volume, Forecasting, Neural network, Genetic algorithm, BP algorithm
PDF Full Text Request
Related items