Research Of Fault Diagnosis Expert System For ATO Based On Fault-Tree

The subsystem structure and working mechanism of automatic train operation (ATO) subsystem, a vital part of the rail traffic signal system, are typically characterized by complexity, multiple targets, delay and nonlinear dynamic system. With the rapid development of urban rail traffic, the safety and no-failure operation of the ATO system operation are increasingly becoming the focus of the public. The development and application of the modern fault diagnosis and testing technology bring new research subjects for the safety evaluation and fault diagnosis. By adopting the fault diagnosis expert system based on the fault tree analysis method, real-time online service can be provided for the drivers and the users so that the possible faults or fault sumptoms of the system can be timely and accurately judged to prevent accidents from happening or if not at least relieve the damages caused by the accidents by consulting the experts’ advice and maintenance methods offered by the system. In this way, the safety and reliability of the ATO subsystem can be guaranteed and improved.Taking the Trainguard MT system of SIEMENS Company as the research object, the thesis mainly studies the working principles, structure and the maintenance methods and fault processing methods of specific equipments. Then by adopting the fault tree analysis method, the thesis will establish and analyze the system fault tree. By introducing the simulation bottom event fuzzy failure rate method before the quantitative analysis of the fault tree, the thesis can solve the difficulties of obtaining failure rate and problems of lacking reliable data, and provide the data support for the qualitative and quantitative analysis of the fault tree. By changing the results of the fault tree analysis into the knowledge and the expert diagnosis opinions of the expert system base, and adopting the development environment based on Visual C++and SQL Server2000data base, the thesis finally completes the design of the fault diagnosis expert system based on the fault tree.ATO fault diagnosis expert system based on the fault tree can achieve the fault detection and real-time diagnosis of the ATO subsystem and other relative important equipments so as to provide experts’ advice and the fault reasoning interpretation within the field for the drivers and users. Meanwhile, the realization of the specific function of the expert system also proves the accuracy and feasibility of fault diagnosis in the field of ATO system by combining the fault tree with expert system. The achievements of the thesis will provide certain theoretical basis and technical support for the application and research of the train fault diagnosis system in the future.