Research On Operational Risk Assessment Method Of High Speed Railway Train Control System

Train control system, track and electric multiple units are the critical technologies of high-speed railway. As the train control system is a complex system, with the increasing application of new technologies and equipment in the field of train control system, the complexity and correlation between systems has been deepen, which has intensified the urgency and complexity of response to the risk assessment of train control system. In the dissertation, the operation risk of Chinese train control system has been taken as the research objective. After the summary of the domestic and foreign achievements and the actual situation of the field, the problems of risk description, risk cause and risk evaluation of train control system have been deeply researched using the theories and techniques involving the emergence, cloud model, combination weight, fuzzy number, Bayesian network. The research work will be of certain theoretical and practical significance to ensure the high safety of the train control system and realize the sustainable development of China's high-speed railway.The main research contents are as follows.The operational risk of train control system was described mathematically, and the risk factors were analyzed from the angle of human, machine, environment and management. The emergence theory was used to describe the operational risk of train control system. The common methods of risk identification were introduced, and the radio block center was taken as an example to analyze the whole process of risk identification.Aiming at the defects of expert scoring method, an improved calculation method was presented for the severity and frequency calculation of risk matrix. Considering the randomness and uncertainty of risk severity, a new method of the severity of risk assessment based on the optimal combination weight and cloud model was proposed in the dissertation.For the frequency of occurrence, a fuzzy fault tree evaluation method was presented. To improve the existing evaluation method completely relying on expert, the real value and the expert fuzzy judgment were combined by the fuzzy fault tree. An example was given to demonstrate the effectiveness and correctness of the proposed method.The deficiency of the existing two-dimensional risk matrix was analyzed. On the basis of the two-dimensional risk matrix, the predictable degree of the risk was increased and a comprehensive evaluation method with G1 and fuzzy number was proposed. A three-dimensional matrix model for operational risk assessment of train control system was presented, and the improved radar chart method was used to integrate the risk severity,predictable degree of risk and frequency of risk occurrence and the three-dimensional matrix chromatogram was given. An example was given to demonstrate the effectiveness and correctness of the proposed method.The probability evaluation for risk was conducted. Taking the on-board equipment of train control system as an example, the Bayesian network method was used to analyze the probability safety of the on-board equipment of train control system. It not only verified the reliability of the vehicle system, but also illustrated the practicability of the Bayesian network in the safety probability evaluation.From the point of system, the operational risk was analyzed in the dissertation and a risk assessment method based on uncertain AHP and cloud model was proposed. The weight interval of safety assessment system was determined by uncertain AHP, which met the engineering practice and thinking of experts. The weighted interval was transformed into the exact weight value by set pair analysis, which made the calculation results more scientific and reasonable. Considering the cloud model's ability to deal with fuzziness and randomness,the evaluation cloud model, index and integrated cloud model were took out for the uncertainty and randomness in safety assessment. Finally, an example was given to verify the feasibility and effectiveness of the application of cloud model and uncertain AHP in the safety evaluation of train control operation.The risk assessment system for train control operation was designed. The physical and logical structure of the system and the function modules were designed in detail. The system was realized based on the hardware and software platform. The correctness and feasibility of the proposed theory and method are verified by the analysis of system function interface and operation results.In terms of risk description, risk cause, risk identification, risk rating evaluation and risk probability evaluation, the operational risk of train control system has been researched in the dissertation, which has certain significance to ensure the safety of the train control system operation.