Research On Prediction Mode Of Urban Short-term Traffic Flow

The traffic flow forecasting is a branch of prognostics, which is an important research area of intelligent transportation system. Fast and accurate prediction of traffic flow is an important premise and foundation of applications in real-time traffic signal control, traffic distribution, route guidance, and etc. However transportation system is a complex system composed of human, vehicles, roads and many other objects, which with characteristics of highly complexity, nonlinearity, uncertainty. So accurate, real-time, reliably traffic flow prediction has become an important research.According to the theory of chaotic time series analysis and chaotic properties of traffic flow, this paper analyzed the real traiffic flow data collected from intersection of Beijing west road and Xikang road in Nanjing, China. The phase space reconstruction technique is used in the traffic flow prediction, expected to explore the laws embedded in the traffic flow to enhance traffic flow prediction accuracy. The multiple-autocorrelation method is used to access the phase-space reconstruction time delay r and the G-P method is adoped to get embedding dimension m, and Lyapunov exponent is calculated to verified the existence of traffic chaos.Traffic flow data has characteristics with highly complexity and nonlinearity, and artificial neural network has a very strong ability with non-linear processing, self-organizing, self-adaptively and self-learning, which is an effective way in traffic flow prediction. Wavelet transform has good time-frequency eharacteristics, that the wavelet neural network is applied in the paper. The supporting, symmetrical and cosine-modulated gaussian wave Morlet wavelet network is adopted as the activation function, its good good time-frequency regularity has prominent advantages at the detection of discontinuous and singular of time sequence.Finally, short-term traffic flow prediction model based on chaos phase space reconstruction and wavelet network has been set up. Research using authentic traffic flow data collected from intersection of Beijing west road and Xikang road has been divided into training data and testing data for prediction studies. The results fully validate the effectivity of improved chaos-wavelet network traffic flow prediction model.