Research On Train Location Method In Urban Rail Traffic

With the rapid development of urban rail traffic, the train location technology has become one of the key technologies of the automatic train control system, its accuracy and reliability impact on the train safety and the traffic efficiency. In automatic train control system and train command system of the urban rail traffic, get the real-time and accurate information of train speed location is the protection of train safety and traffic efficiency. So it has great significance to study the train location method deeply.This paper aims at improving the accuracy and reliability of train location in the urban rail traffic. Because of the restriction of the location technology such as GPS in the urban rail traffic's special circumstances, the paper research the train location method based on the train speed measure. The train location based on the speed sensor is the common method in the rail traffic, but it has problem of wheel's slide and slip. The Doppler radar get more attention in train speed measure, but it still has some problem. Morever, location base on speed measure has the problem of error accumulated. Location with single sensor is difficult to achieve high accuracy and reliability. However, each sensor has its own advantages, the paper combine the speed sensor, Doppler radar, accelerometers and EUROBALISE for multi-sensor information fusion, and they will complement each other effectively by their advantages, as to achieve high accuracy and Reliability.The paper focus on the research of the application of the multi-sensor information fusion for train speed measure and location in urban rail traffic. In the paper according to the characteristics of various sensors the method of train location base on multi-sensor information fusion is researched, and the advantages and disadvantages of several fusion methods is analysed. In the paper the optimal information estimation methods is deeply researched, also the algorithm model of the train location. The classic information fusion algorithm is applied to the train location in urban rail traffic, and the train location method based on Kalman filtering algorithm and robust H_∞filtering algorithm are specifically researched, and also their improvement method. Reference federal Kalman optimal filtering information fusion algorithms, the train location base on the H_∞filtering information fusion is presented, which is applicable to the urban rail traffic. Finally, the feasibility of algorithm is proved through experimentation. The conclusion come out that, multi-sensor information fusion can improve the accuracy and reliability of the train location, and the train location method base on the federal Kalman optimal filtering information fusion has the higher accuracy, but it need the accurate mathematical description of the interference signal, and its robustness is not strong. The train location method base on the federal H_∞filtering information fusion don't need any hypothesis of the interference signal, it reduces the difficulties and improves the adaptability of the train location, it is more robust, and the accuracy of estimates is close to the train location method base on the federal Kalman optimal filtering information fusion. So it is more suitable for the urban rail traffic.