Analysis And Evaluation Of Reliability For High-speed Train Transmission System

With the high speed development of our country's economy, the requirements for transportation grow day by day. To mitigate the strain on railway transportation capacity, China has carried out large-scale railway speed increase for 6 times since 1997. At the same time, by adhering to the principle of introduction, digestion, absorption and re-innovation, our country's high-speed train presents fast and large-scale development trend. At present, the total mileage of high-speed train operation for the design speed of 200 kilometers per hour and more grades is 23600 kilometers, and the total mileage of high-speed train operation with design speed over 200 kilometers per hour is 20000 kilometers, which is 60% of the total mileage of world's high-speed train operation. The development of high-speed train not only shorten the distance between cities in time and space, but also change the way people travel, meet people's demands for travel quality.The safety problems of high-speed train gradually manifests with the constant increase in its running speed. Since the running speed is very fast and the service environment is very complex, it will bring serious casualties and great economic loss once an accident occurs. Hence, ensuring safe and reliableoperation become the important foundation for the development of high-speed train, which require the components and system of high-speed train have higher reliability. As an important part of high-speed train, the safe, reliable and stable operation of transmission system will inevitably influence and determine the safety, reliability and economy of high-speed train operation. However, during the process of high-speed train operating, transmission system not only suffer the effect of track irregularities, but also suffer vibration and shock from the body, making the working environment worse. The reliability problem of components of transmission system is becoming more important, and the safe operation ability of transmission system during its service time has become an urgent concern for field operation and maintenance.This dissertation is under the guidance of system engineering, taking transmission system of CRH3 high-speed train as the research object. The thesis comprehensively analyzes the reliability of transmission system from two aspects: self-factors and external factors by utilizing the proportional hazards model(PHM), support vector regression(SVR), auto-regressive(AR) model, fuzzy influence diagram(FID) and extension theory. The main research contents are listed below.1. Analysis of operating reliability of critical componentsThe mathematic relationship between running state response of critical component and its operating reliability degree was studied based on PHM. The local tangent space alignment algorithm was utilized to realize the dimension reduction of multi-domain high-dimension feature vector set. The chaotic particle swarm optimization(CPSO) algorithm was used to estimate the parameters of PHM, and then the reliability evaluation model of transmission system critical parts under high-speed train operating state was established. Besides, the vibration transmission property of transmission system was studied from the perspectives of time domain, frequency domain and vibration acceleration frequency spectrum, and the main vibration source of transmission system during line operation was analyzed.2. Forecasting the residual life of critical componentsTo improve the forecasting performance of vibration signal, a hybrid prediction model which combined SVR and AR model was proposed based on CPSO althogrim. The effectiveness of the proposed hybrid prediction model was verified by comparing the one-step prediction of vibration signal with three different models. The forecasting performance of the proposed model was further analyzed, and then it was used to predict the vibration signal of critical components. Combined with the reliability evaluation model given in the previous chapter, the residual life of critical component was predicted based on the forecasted vibration signal.3. Analyze the external factors influencing the reliability of transmission systemThe external factors which influenced the reliability of transmission system were analyzed under the guidance of system engineering. The influence diagram of external influencing factors which affected the reliability of transmission system was established by analyzing the relationship between each factor. The FID theory was adopted to evaluate the external factors to find out the factor which has the most important influence on the reliability of transmission system, and the corresponding control measures were proposed.4. Multi-level evaluation of reliability of transmission system based on the extension theoryThe evaluation index system of reliability of transmission system was constructed from the aspects of personnel factor, self-condition factor of transmission system, infrastructure construction factor, environment factor, management factor and maintenance factor on the basis of system engineering. To determine the weight coefficient of each index, the subjective evaluation method and objective analysis method were combined on the basis of the game theory. The extension theory was introduced as an assessment method to evaluate the reliability of high-speed train transmission system by using its characteristics of qualitative analysis and quantitative evaluation. By establishing the multi-level matter-element model and combined with the correlation function, qualitative and quantitative reliability evaluation of transmission system was realized.5. Operational condition monitoring of critical components of transmission system based on the extension theoryThe matter-elements of operational condition of critical components were constructed based on the extension theory, and a determination method of classical domains of characteristic indexes based on CPSO algorithm was proposed. The results of classical domains were compared to that determined by mathematical statistics method. The accuracy and effectiveness of the classical domains gotten by CPSO algorithm were verified through a study case. On the basis of classical domains, the operational condition of critical component was monitored qualitatively and quantitatively with the correlation function. Furthermore, an operational condition monitoring system of critical components was designed with the correlation function as the qualitative and quantitative analysis tool, and together with Matlab GUI as the development platform.