Research In Traffic Date Fusion For Highway Tunnel

Tunnel monitoring system is an important guarantee for the safe operation of the tunnel, can be achieved by optimizing the control of tunnel that the traffic is smooth, effectively reduce the energy consumption goal. The various control measures are based on the detecting traffic data, but usually the accuracy of testing data is not high, poor detection systems integration, dramatically reducing the detection efficiency; There is also information can not be detected effective use, resulting in the value of monitoring system did not fully reflect.This paper studied the data highway tunnel traffic monitoring system acquisition and short-term traffic flow forecasting related issues based on the fusion method.The main duties of this paper include the following aspects: Firstly, from the perspective of the system studied the basic theory of data fusion technology, the main research is objective, classification and algorithms.of data fusion.Secondly, commonly used for tunnel traffic detectors: circular detector and video detector, analysis of principles and detection performances for this two kinds of traffic data collect, for two kinds of detectors work alone in traffic sensitive issues, put forward Neuro-Fuzzy Inference Based on heterogeneous traffic detection data fusion algorithm to improve the detection accuracy of the data traffic.Thirdly, based on the theory of short-term traffic flow forecasting, and the analysis of the traditional forecasting models,the paper introduced the multi-model integration forecasting methods and variable step-size multi-model fusion prediction method to predict, established of a prediction model ,and proposed to achieve predictable steps.Finally, based on the tested traffic data from a circular detectors and video detectors on the same surface in Beibei tunnel, using data fusion method proposed in this paper to amend the testing data, and comparative the survey data. The fusion method greatly improves the accuracy of testing data, and this integration of data as historical data, using data fusion methods and traditional forecasting models, respectively, realized on the short-term traffic forecasts.Through the analysis of this article points out the strengths and weaknesses of several data fusion model in the tunnel control system applications, and put forward the problem of the further studying, lay the foundation of the application of data fusion technology in tunnel traffic data processing.