Reliability Assessment And Prediction Method For Smart Electricity Meters Under Spot Trading Mode

Under the spot trading mode of the electricity market,real-time settlement and collection of users’ electricity consumption and electricity information are carried out.The reliability research of electricity meters should not only consider historical data,but also the field data information generated during real-time operation.However,the traditional reliability research methods rely on historical data and cannot conduct real-time research and judgment on the operating state of in-service smart electricity meters.Based on the study of the failure process and operation process of smart meters,this paper preprocesses the field data,and carries out the reliability assessment and prediction of intelligent electricity meters by means of post-analysis and in-process analysis respectively.This paper evaluated the running state of the remaining in-service tables of the same batch with the data information of the batch fault meters,and evaluated its future running state with the data information generated by the single in-service meter itself.The specific research contents are as follows:First of all,the paper discusses the separation and sorting process of the electricity meter and the electricity information collection process.According to the characteristics of the disassembled fault data,the field data preprocessing method of left deletion and right truncation is presented.At the same time,the abnormal event data in the electricity information acquisition system is preliminarily statistically inferred,and the typical abnormal event with high probability is obtained.Then,for deleted data structure of asynchronous running,the reliability evaluation model was constructed based on Weibull distribution,and the point estimation and confidence limit estimation of the reliability measure were obtained.For the case of prior information acquisition,the Bayes reliability evaluation method is carried out by referring to the latest standard of State Grid.According to the pretreated fault data structure,the prediction method of the future interval fault number of smart meters is given to solve the problem of storage and replacement of batch smart meters,the reliability evaluation and prediction of batch remaining in-service electricity meters were completed,and the reliability post-analysis of smart meters under spot trading mode was realized.In addition,the Bayes method combined with Beta distribution is used to predict the probability and cumulative number of abnormal events triggered by the fault of smart meters to solve the problem of whether the typical abnormal events are caused by the failure of smart meters under the spot trading mode.For the running state of a single electricity meter,the occurrence time of abnormal events is predicted based on Poisson process,and the failure time of the smart meter is predicted based on the prediction results of the cumulative number of abnormal events triggered by the failure of the smart meter,so as to predict the remaining life of the in-service smart meter,and realize the operational analysis of the reliability of the smart meter in the spot trading mode.Finally,the verification research was carried out in combination with engineering examples,and the effectiveness of the method was proved by using on-site fault data and abnormal event data,and the reliability analysis of the smart electricity meter was completed after the event,in the process,in batches,and in single and dynamic synthesis.