基于马尔科夫模型的短时交通流预测分析-prediction and analysis of short-term traffic flow based on markov model.docxVIP

AbstractShort term traffic flow prediction plays a significant role in urban traffic volume control and guidance system. The quality of prediction influences the effect of urban traffic flow control and guidance system directly. Hence, short term traffic flow prediction is very essential to intelligent transportation system (ITS).This paper, aims at improving the accuracy of the prediction scheme, builds the Markov model first. Considering the non-linear and non-stationary characteristic of the traffic data time series, this paper proposes Markov-BP neural network model and Wavelet-Markov-BP neural network model, respectively.The main contents of this paper list as below:Since the data gained from inductive loop detector has errors, redundancy and other quality problems, this paper utilizes outlier mining technology to detect abnormal data and makes use of ‘adjacent time period data average’ method to repair abnormal data, in order to resolve data quality problems. Then, an improved wavelet denoising method is proposed. The traffic flow data after noise reduction processing can reflects the characteristic of the traffic flow better.Considering the non-linear characteristic of the traffic data time series, this paper proposes Markov-BP neural network compositional model. By using the error correction thought and dynamic rolling prediction scheme, this compositional model can achieve better results in comparison with Markov model.Considering the non-stationary of the traffic data time series, this paper proposes Wavelet-Markov-BP neural network compositional model (W-M-BPNN). It uses multi-resolution analysis to decompose the raw traffic data time series. Therefore, the time series will be decomposed on different scales and those decomposed time series can present more properties of traffic flow. The experiment results show that the three models can achieve acceptable accuracy, by contrast, the W-M-BPNN model is superior to the other two models.Key words: Intelligent