| As people’s requirements for transportation are getting higher and higher,and the high-speed railway industry is booming,the reliability of high-speed train systems is receiving more and more attention.Train speed and position information is critical to the train control system.Therefore,in order to reduce the impact of speed positioning error accumulation on system reliability degradation,the paper fully considers the complexity and variability of high-speed train operating environment,taking the wheel speed measurement positioning system as the research object,using the combination of similarity propagation clustering(AP)and Hidden Markov Model(HMM)to identify the degraded state of the odometer-based train speed and distance measurement system,and then uses the non-structural triangular fuzzy number weighting method to assign weights to establish the system real-time operational reliability assessment and prediction model,and implement conditional maintenance decisions based on real-time evaluations of operational reliability.This paper takes the train’s key basic component odometer-based train speed and distance measurement system as the object,studies the overall scheme of the the reliability and maintenance decision of the basic components,and provides the preliminary basis for the real-time reliability assessment and condition-based maintenance of other components of the high-speed train and even the entire vehicle-mounted equipment.The research work completed in this thesis mainly includes:(1)For the odometer-based train speed and distance measurement system,considering the unique influencing factors and mechanisms of the degradation state,the system degradation state model is established,which can evaluate the current degraded state and predict the future degradation state of the system.(2)It presents a real-time reliability evaluation method for the odometer-based train speed and distance measurement system based on AP-HMM and unstructured trigonometric fuzzy weighting method.It establishes a real-time reliability evaluation model based on the degradation model.The real-time operational reliability evaluation framework for odometer-based train speed and distance measurement system,including sensor failure,slip/slide,self-aging and wheel diameter wear,evaluates the reliability of current operating conditions and predicts future operational reliability.(3)Based on the real-time reliability evaluation results of the system,the maintenance cost rate and availability are optimized for multi-objectives.The preventive maintenance decision-making model is established and the bacterial foraging optimization algorithm is used to solve the problem.Realize the maintenance of the axle speed positioning system.In order to verify the real-time reliability evaluation method of the odometer-based train speed and distance measurement system proposed in this thesis,the reliability of the single influencing factor is established based on the Wuhan-Guangzhou line measured data.The real-time evaluation model of the operational reliability of the odometer-based train speed and distance measurement system is established according to the weight of the dynamic distribution of the influencing factors.Finally,the preventive maintenance optimal decision-making model is established,and the feasibility of the real-time reliability evaluation method of the the odometer-based train speed and distance measurement system based on AP-HMM and unstructured triangular fuzzy number weighting method is verified.The results obtained by the thesis provide an important way to solve the real-time evaluation of the operation reliability of the high-speed train operation control system under the influence of multiple factors,and provide important support for real-time evaluation and maintenance of high-speed train basic components and even the entire vehicle equipment reliability.Figure 71,Table 28,references 111. |