A Study On Urban Expressways’ Traffic Information Extraction And Coordinated Optimization Method

With the rapid development of domestic economy, vehicle ownership rises dramatically, traditional infrastructure investment alone cannot cope with growing travel demand. Many metropolis, such as Beijing, Shanghai, Guangzhou and Chengdu has witnessed severe traffic jam on urban expressways. Traffic on urban expressway is facing with problems like low transportation efficiency, frequent traffic accident, severe environmental pollution, etc. Urban expressways serve as the backbone of city’s road network, the main arterials of regional and external connection, which carries most of city’s daily traffic volume. Therefore, raising expressway’s transportation efficiency and alleviating traffic jam is the main priority of urban traffic environment improving.The technical progress of data collection lays solid foundation for the research of traffic information extraction. Traffic data has following characteristics:large-scale data volume, wide fluctuation range of application load, high dependence for dynamic information processing, high demand for data sharing, high demand for applicability and stability. However, different data collection system are controlled by different administration, the administrative barrier makes data hard to share. Moreover, data collectors’ layout shows a lack of coordination, data collection cycle difference and data format difference makes traffic data fusion even harder.At present, many cities have active Intelligent Transport System, which makes traffic infrastructure works at higher efficiency. However, due to the complexity of traffic flow, the correlation between traffic control system and traffic guidance system is not significant enough to jointly produce ITS benefit.With traffic data collection system and its multi-source, high volume, varied format, dynamic urban expressways’traffic data, based on the research of expressways’traffic information extraction and traffic coordinate control technology, this paper studies on expressways’ traffic data fusion, traffic flow parameter prediction based on granule computing, and expressways’traffic coordinate control. Main research findings include:(1) This paper introduces float car data collection, video data collection and microwave data collection principles, then practices a statistical analysis on the joint traffic data collected by three method, and finally conclude data cleaning method, lays data foundation for further research.(2) Various detectors’data cannot fully reflect actual traffic conditions, each one of them contains limited traffic information, therefore a multi-source traffic data fusion technology of urban expressway based on AF-SVR model is proposed in this paper. Firstly, the traffic data collected by different detectors is used to train the Support Vector Regression (SVR) model. Secondly, the parameters of Support Vector Regression model are calibrated by Artificial Fish (AF) to build optimal model for multi-source traffic data fusion, the output can reflect actual traffic conditions, which is verified by field velocity data. Finally, this multi-source traffic data fusion technology is applied on 3rd Ring Road Expressway in Chengdu.(3) This paper takes information granule as basic analysis unit, provides a brand new approach to build fuzzy time series model aiming to neutralize original model’s defects. Based on digging internal correlations of the data, this new approach analyze dynamic regional interval length under the influence of time variables. This approach is characterized as Gath-Geva fuzzy clustering algorithm based on the segmentation of time series, where the information granule serves as data units analyzed by granular computing, from which the traffic flow parameters’fluctuation trends are captured. Experimental results shows that traffic flow parameter prediction based on granular computing gives reasonable confidence interval and is more reliable than original parameter prediction approach.(4) In the research of control sub-area’s dynamic partitioning, road network are analyzed under normal state and abnormal state environments. Based on this paper’s integrated analysis of traffic information, under normal state, traffic state information are adopted as the main decision parameter, this paper presents clustering degree of correlation, and devises traffic sub-area dynamic partitioning procedures. Under abnormal state, based on traffic wave theory, from traffic incidents’alert level (or incidents’type), incidents’ respond time, incidents’upstream traffic parameters, this paper builds an urban expressways’ traffic incident influence area detection model in concern of time span and space span.(5) This paper presents a regional traffic coordinate control method based on modified MPC strategy. This method takes ramp control, variable speed limits control, VMS guidance control into consideration, realizes urban expressways’ dynamic control, the model has attributes like closed loop, optimization and coordination. The case study results shows that this model performs better in solving precision and solving speed, compared to former models.(6) This paper developed a software demo named "Urban Traffic Management and Control Oriented Information Extraction and Coordinate Optimization System". In the background of Chengdu 3rd Ring Road, the software utilizes above chapter’s data process technology and application model, realizes functions like traffic information checkout, traffic state prediction, dynamic sub-area partitioning and regional traffic control, which can be referred during Chengdu’s ITS construction session.Finally, this paper concludes main research findings and innovations, explains drawbacks and specifications that calls for further research.