Study On The Maintenance Scheme Of AC27.5kV Vacuum Circuit Breaker With Partial Observability

With the continuous construction and development of high-speed railway,the importance of maintenance for traction power supply equipment is increasingly prominent.At present,the traction power supply equipment usually adopts the maintenance mode combining periodic maintenance and accident maintenance.The maintenance personnel passively perform the maintenance work according to the predetermined plan.The flexibility and pertinence of this maintenance mode are poor,and the phenomenon of insufficient and excessive maintenance is common.Condition-based maintenance is a kind of maintenance mode that can make maintenance plan flexibly according to the state of equipment.Compared with traditional maintenance mode,condition-based maintenance has strong initiative.The maintenance plan is adapted to the actual needs of the equipment,which reduces the waste of human and material resources.At the same time,it can effectively solve the problems of insufficient maintenance and excessive maintenance.Condition-based maintenance has obvious advantages and has gradually become the new development direction of equipment maintenance mode.However,there are still some challenges in its application.Affected by the error of the equipment state data and the uncertainty of the state assessment method,there is uncertainty in the state assessment results,which may not be consistent with the true state of the equipment.Making a state maintenance decision without considering uncertainty may result in a decision error.Therefore,this paper takes AC27.5kV vacuum circuit breaker as the research object,and studies the maintenance plan considering the uncertainty of state assessment.The state assessment result is the basis for the condition-based maintenance decision.In order to improve the accuracy of the state assessment,this paper first studies the comprehensive assessment method of AC27.5kV vacuum circuit breaker.Gray correlation analysis and entropy weight method were used to determine the subjective weight and objective weight respectively,and the geometric average method was used to solve the combined weight.Subsequently,the ideal interval method was used to conduct a comprehensive assessment of the health status of the AC27.5kV vacuum circuit breaker.Finally,a health value is proposed to quantitatively describe the health status of the circuit breaker.In the degradation process of AC27.5kV vacuum circuit breaker,in addition to the performance degradation caused by natural aging,there are also sudden performance degradation caused by mechanical or traction load impact.Therefore,based on the GaussPoisson compound stochastic process,a comprehensive degradation process model of AC27.5kV vacuum circuit breaker is established.On this basis,considering the fact that there is uncertainty in the state assessment of AC27.5kV vacuum circuit breakers,the partially observable Markov decision process(POMDP)is introduced to study the maintenance decision method.The state observation probability is used to characterize the uncertainty of the state assessment and to correct the error of the state assessment result,and the belief state is used to represent the state of the circuit breaker in the form of probability distribution.Considering maintenance cost,maintenance cost and failure risk cost comprehensively,an objective function is established based on Bellman formula,and the lowest comprehensive cost per unit time is taken as the optimization goal to solve the optimal maintenance plan.Finally,in consideration of the demand for reliability and economy of the maintenance plan,a multi-objective optimization method for the maintenance plan of the AC27.5kV vacuum circuit breaker was studied.The method takes the average reliability,availability and comprehensive cost per unit time as the optimization goals.Use the degradation process model to derive the reliability function.Establish the availability function by comprehensively considering various types of shutdown events that may occur.The multi-objective function is established,and the NSGA-Ⅱ algorithm is used to solve it,so as to obtain the optimal maintenance plan that takes into account reliability,availability and economy.Theoretical research and analysis of calculation examples show that this paper realizes the decision of the maintenance plan of AC27.5kV vacuum circuit breaker when there is uncertainty in state assessment.This method can avoid the decision-making mistakes caused by ignoring the uncertainty,and make the decision-making result in line with the actual demands of the equipment.At the same time,it also realizes the multi-objective optimization of the maintenance plan.This paper can provides ideas and theoretical basis for traction power supply equipment maintenance decision.