Research On Optimization Method Of Maintenance Plan For Metrological Verification Line

With the accelerating process of power modernization and intelligent construction in China,SGCC(State Grid Corporation of China)has built a large number of electric energy measurement and verification lines in the past ten years to meet the growing demand for electric energy measurement and verification.Considering the importance of metrological verification line,carry out inspection and maintenance plan to ensure its operation quality is of great economic and social significance.Therefore,this thesis studies the optimization method of maintenance plan of metrological verification line.The main research contents are as follows:(1)This thesis comprehensively summarizes the research status of maintenance related problems at home and abroad,taken the metrological verification line of a power measurement center as the specific research object.This thesis deeply studies and analyzes the maintenance technical requirements of metrological verification line,and summarizes the shortcomings of the current maintenance plan.Aiming at reducing the cost of maintenance and false inspection,considering the influence of equipment degradation,time window of maintenance resource and other factors.Explores different optimization methods of maintenance plan,and successively improves the NSGAII(Non-dominated Sorting Genetic Algorithms-II)algorithm that used for model solution by using methods such as variable neighborhood search and adaptive hybrid mutation in the optimization process.(2)Consider the impact of equipment degradation on maintenance plan.According to the actual situation,an optimization method of maintenance plan considering equipment degradation is proposed.Where degradation is described by time delay model,the relationship between service age and failure probability is expressed by Weibull distribution model,and the maintenance effect is described by service age fallback.The maintenance plan optimization model aiming at minimizing maintenance cost and false inspection cost is established.An NSGA-II algorithm with mixed VNS(Variable Neighborhood Search)and disturbed population is designed to solve the problem.Effectiveness of the model and hybrid algorithm is verified by taking the verification line of low voltage current transformer in a metering center as an example.(3)Consider the influence of maintenance resource time window on maintenance plan optimization.The optimization model is further improved with limitation of maintenance resources’ available time window.To solve this problem,an improved NSGA-II algorithm with heuristic population initialization,adaptive hybrid mutation and VNS is proposed.Through the comparative analysis of actual cases,the effectiveness and feasibility of the optimization method and optimization model are verified respectively.(4)Combined with actual needs and relevant regulations of electric energy metering enterprises,a prototype system of intelligent monitoring platform for metrological verification line is designed and developed.This system covers functions of basic information management,inspection and maintenance plan management,inspection data management,analysis and discrimination of inspection data,provides managers with complete and accurate information,and improves the informatization level of operation quality supervision of measurement and verification line.