Font Size: a A A

Short-term Urban Traffic Flow Prediction Based On The Chaos And Fuzzy Neural Network

Posted on:2013-06-21Degree:MasterType:Thesis
Country:ChinaCandidate:F QuFull Text:PDF
GTID:2232330392959135Subject:Intelligent Transportation Systems Engineering and Information
Abstract/Summary:PDF Full Text Request
With the improvement of the intelligent traffic technique, vehicle guidance system whichis an important part of the intelligent traffic system has become an effective way for trafficmanagement department to guide the urban traffic. The key technique of the vehicle guidancesystem is the prediction of the urban traffic state, i.e. use historical traffic data and real-timetraffic data to predict the traffic flow of the urban cross and road section.……First, The dissertation has made a study of AR, MA, ARMA Model and determines theorder of AR, MA, ARMA Model by regression function, moreover, the dissertation alsoverifies the accuracy of AR, MA, ARMA Model prediction by simulation. Second, thedissertation analyzes the neural network nonlinear model of time series and first discusses themodel and characteristics of BP, RBF, Wavelet Neural Network. The dissertation analyzesthe time series by use of BP, RBF, Wavelet Neural Network respectively on the basis of dataprocessing, and demonstrates the effectiveness of these three neural network in flowforecasting by simulation. The paper also research on Traffic flow forecasting method basedon chaotic systems and fuzzy neural network, many domestic and foreign research resultsshow that traffic flow has chaotic characteristics. In order to forecast the short-term trafficflow,we should do the following: first, calculate the chaotic characteristic parameters of thetime series of traffic flow; second, illustrate that time series has chaotic characteristics by useof the Lyapunov test; third, calculate the lag time of time series by mutual informationmethod; fourth, calculate the embedding dimension through false nearest neighbor algorithmand make it as the input layer nodes of fuzzy neural network; finally, establish ANFISnetwork.The Malab simulation graph and the performance index parameter present that theproposed methods have gotten high precision and is good enough for the real-time controllingand guiding of the traffic flow.
Keywords/Search Tags:Intelligent Transportation System, Short-term Traffic Flow Prediction, ARMA, BP, RBF, Wavelet Neural Network, Chaos, Fuzzy-neural Network
PDF Full Text Request
Related items