Research On The Reliability Modeling And Evaluation Method About The CTC Of High-speed Railway

With the development of social economy and the progress of science,the computer system becomes more and more complex, and its reliability is becoming more and more serious. As a complex computer system, the high-speed railway centralized traffic control (CTC) system is not only an necessary equipment for the high-speed railway dispatching command, but also an important management facility to ensure the punctual and orderly operation of trains. Compared with ordinary railway, high-speed railway has the characteristics of high density, high speed and large capacity, at the same time, the high-speed railway requires high reliability for once it breaks down, the harmfulness and social influence are extremely far-reaching.At present, the theoretical system of system reliability has already completed and entered the stage of popularization and application. Based on the theory and method of the system reliability, surrounding the main aspects that include the hardware system, the sofeware system, system data, dispatch and so on, this dissertation researched the system reliability modeling and evaluation method using the system engineering method, the results can make references for the reliability of the CTC system.The main research contents and results are as follows:(1)The structure, function, key factors that affect the reliability and failure mechanism of CTC system are analyzed systematically, and the research object of this dissertation is selected from the aspect of system elements.(2) Based on the analyzing of the system structure and working principle of the two different structures of the station autonomous machine system and considering the factors such as the failure rate and the detection rate of the extension,the system reliability and safety model has been constructed based on the continuous Markov(Markov) process. The results show that double hot stand by autonomous system has higher reliability than double 2-vote-2 autonomous system, but double 2-vote-2 autonomous system has higher security than double hot stand by autonomous system.(3) For the dynamic and fuzzy analysis of the reliability of hardware system, the two kinds of reliability models of autonomous machine based on Dynamic Fault Tree(DFT) have been established firstly. Then, the triangular fuzzy number of failure rate and detection rate of the extension and obtaining fuzzy state transition matrix has been calculated according to Markov model. Lastly, the fuzzy reliability of the system is obtained by Laplasse transform. Based on the similar cases (2), it is proved that the reliability of the double hot standby system is still higher than that of the system with double 2-vote-2 system.(4)For the real-time,function and structure of the CTC system's software,considering the execution time ratio, failure rate, importance and other factors of each task module,the reliability evaluation model of software system is established based on real time multi task. The software of high-speed railway CTC system with 6 dispatching functional modules is used as the example,and the results show that the scheduling task module has a greater impact on the reliability of the system compared to other task modules, thus the reliability of CTC software can be greatly improved by reducing the failure rate of the task scheduling module.(5)According to the data classification and flow relationship of CTC system, a comprehensive evaluation index system is established, and then, the reliability evaluation model is constructed based on analytic hierarchy process (AHP) and information entropy. The model is used to evaluate the actual data of a certain CTC system and the results show that the lower limit values of reliability for basic plan data,implementation plan data, dispatching data are 0.8687, 0.9229, 0.8661 and the comprehensive data ((including the three data above) of system is 0.8802.(6)The dispatching personnel in CTC system are divided into decision makers and operators, and in the framework of SRK human error database, a fuzzy comprehensive evaluation model of human reliability is constructed based on scenario analysis and set pair theory. The results show that the reliability of the operator is reduced by 0.04% and the reliability of the operator does not decrease with the increase of working time, but the reliability of the decision maker was reduced by 2.29% and the reliability of the decision maker decreased significantly with the increase of working time.