Research On The Stability And Prediction Method Of Electric Energy Meter Measurement Error In Field Environment

Smart energy meter is an important measurement unit in the national grid,and its measurement accuracy is related to the interests of users and the reputation of the enterprise.However,the smart electric energy meter has large measurement errors in the field environment,and the operation stability is poor,and sometimes it even threatens the normal operation of the entire power system.Therefore,the study of the stability and prediction method of electric energy meter measurement error in this paper has important practical significance.First of all,this article used python software to make basic statistics on the collected test data.On this basis,in view of the influence of field environmental and electrical stress on the measurement error of electric energy meters,linear regression and polynomial regression algorithms are used to construct regression models of environmental single stress,electrical single stress and measurement error respectively.The result was that the use of these two regression models can well show that the measurement error of the electric energy meter is affected by the environmental and electrical single stress.Secondly,in view of the influence of severe cold environment on the measurement error stability of electric energy meters,the ARIMA algorithm is used to fit the change trend of the upper and lower limits of measurement error of 20 electric energy meters in a year,so as to study the change trend of measurement error stability.Aiming at the important stress problem of electric energy meter measurement error in field environment,the integrated algorithm XGBoost is used to study,and it is concluded that the measurement error of electric energy meter in field environment is mainly affected by the superimposed stress of temperature and absolute humidity.Finally,by comparing the change trend of the measurement error stability of the electric energy meter at the same base under different test points,it is concluded that the test point conditions have a certain influence on the measurement error stability.Therefore,the XGBoost model is applied to each base,and it is concluded that the measurement error of the electric energy meters of the two bases is more affected by electrical stress than by environmental stress.Aiming at the problem of electric energy meter measurement error prediction,this paper used a big data algorithm Light GBM to construct the metering error prediction models of the two bases respectively,and verify the prediction accuracy of the model on the test set.