Reliability Analysis And Evaluation Of CTCS-3

As the core equipment of high-speed railway, train control system is extremely important for ensuring the safety and efficiency of train operation. In recent years, the train operation velocity has been increasing consecutively, which makes the requirements of safety and reliability for the system to be much higher in the future. Train operation control system is a complicated system that integrates four technologies into one including computer, communication, control(3C) and the traditional railway signalling technology. At the same time, it is also a typical safety-critical system. Once the failure occured, it will endanger the safety of train operation, and cause heavy casualties and huge property losses. This thesis took Chinese train control system at level 3(CTCS-3) as the research object with dynamic bayesian network(DBN) and cloud model were applied in its reliability analysis and evaluation, respectively.DBN was introduced in this thesis for overcoming the deficiency of traditional methods such as fault tree analysis(FTA), markov and bayesian network(BN) in reliability analysis of CTCS-3. Firstly, the fault tree(FT) model of CTCS-3 was constructed on the basis of its structure and function, and then it was converted into an equivalent DBN in accordance with the translation rules. Secondly, the reliability of CTCS-3 was analyzed by considering the effect of factors such as redundancy, maintenance, common cause failure(CCF) and polymorphism on system reliaility. The results show that the bidirectional inference of DBN can not only analyze whether it is satisfied with the requipment specifications of reliability, avaliability and maintainability(RAM) indicators of CTCS-3 but also identify the weak links of system effectively and simplify the calculating process. It is vital significant for improving system performance, reducing the costs associated with maintenance and ensuring operation safety of high-speed railway.In order to make the reliability evaluation results be more correctly and objectively, a new method which combined cloud model with combination weighting was proposed in the reliability assessment of CTCS-3. Firstly, the comprehensive evaluation model and hierarchy structure model of CTCS-3 were put forward based on the structure and function of system. Subsequently, the evaluation indexes weights were determinated through the method of combination weighting. Secondly, the ratio of actual and rated indicator of each equipment was expressed by using cloud model, and then the integrated indicators cloud model of the reliability of CTCS-3 was obtained according to the cloud operation rules and weighting model. Finally, the ultimate reliability evaluation result of CTCS-3 was gained through measuring the similarity of integrated indicators cloud and the cloud model of reliability evaluation set. The results show that the reliability evaluation of CTCS-3 based on cloud model and combination weighting can effectively describe the ability that CTCS-3 performs its prearranged functions, which provide a reliabiliable basis for the design and analisis of railway signalling system.