Application Of Data Fusion Technology In Fault Diagnosis Of Maglev Train Operation Control System

In maglev train transportation system, Operation Control System (OCS) plays a core role of automatic control and safety protection on the train, and it is a key equipment to ensure that the train operates safely. Because of the aging of all the components and impact of environmental factors, the occurrence of failure is inevitable. Therefore, the study on the technology of OCS fault diagnosis is very important to maintain the stability of maglev train transportation system. After researching both the domestic and foreign fault diagnosis methods, a new method combining data fusion technology is proposed and applied in the project of OCS equipments fault diagnosis. The project is subordinate to the project of "Innovation and Industrialization of High-speed maglev train" which is supported by National Key Technology R&D Program of China’s Eleventh Five-year Plan. The results indicate that, this method achieved satisfactory effect of diagnosis.The main contributions of this thesis are listed as follows:1. Analyzed the maglev train OCS fault diagnosis system specifically, and clarified the types and fault states of all the components that need to be diagnosed in the OCS.2. After comparing the applicable conditions of various data fusion methods in fault diagnosis area and based on its own characteristics of the maglev train OCS, a new fault diagnosis method is proposed and the effectiveness is analyzed. This method uses a pairwisely fused model with multi-level diagnose, which improves the original D-S evidence theory data fusion method.3. At last, the new method is applied in the OCS equipments diagnosis which is subordinate to the project of "Innovation and Industrialization of High-speed maglev train". The project is supported by National Key Technology R&D Program of China’s Eleventh Five-year Plan. Via C++programming tool, a complete fault diagnosis system has been constituted. The results indicate that, the new method not only can improve the validity and reliability of the fault diagnosis results, but also can solve the problem of the existing method that when there is too much evidence the fusion process is very complex and it is difficult to find the intersection of all the evidence.