Prediction Of Degradation Trend Of Smart Electric Energy Meter Based On IGA-BP Neural Network

Smart electric energy meters play an indispensable and important role in the construction of smart grids,and their performance directly affects the accuracy of grid edge measurement and electricity metering.Degradation or failure of a smart electric energy meter will seriously affect its performance,especially in different environments.In this thesis,an improved genetic algorithm is used to optimize the BP neural network model(IGA-BP)to study the degradation trend of smart electric energy meters.The optimization performance of BP neural network is optimized to improve network performance.This thesis firstly clarifies the research background and significance of the degradation trend of smart electric energy meters,summarizes the research status of the degradation trend of instruments and equipment,and then selects the method of machine learning to study the degradation trend of smart electric energy meters.The IGA-BP model is used to study the degradation trend of smart electric energy meters.The basic error of the smart electric energy meters is selected as its degradation index,and the change trend of the basic error over time is the degradation trend of the smart electric energy meters.Sort and analyze the extracted data,and determine the environmental stress most related to the degradation trend of smart electric energy meters through correlation analysis method;excavate the basic information contained in the degradation data of smart electric energy meters,and use the function fitting interpolation method(FFI)to eliminate the influence of missing values in the original data on the degradation analysis.Then,a research model for the degradation trend of smart electric energy meters based on BP neural network is established.In order to improve the prediction performance of the model,a genetic algorithm is introduced to optimize the parame ters of the BP neural network model;Due to the inherent limitations of the genetic algorithm,an improved strategy is proposed to improve the genetic algorithm.The improved genetic algorithm is used to optimize the parameters of the BP neural network again,and the stability of the model is judged through the simulation test.Combined with the processed smart energy meter degradation data,the filtered environmental stress data and time stress data are used as the input,and the basic error of the smart energy meter is used as the output.The IGA-BP neural network model is trained and tested by the method of cross-training and testing,so as to realize the backward prediction of the degradation trend of the smart electric energy meter.Simulation experiments are carried out based on the Matlab platform,and a variety of similar prediction models are introduced for comparative analysis.The experimental results show that the IGA-BP neural network model has higher prediction accuracy than other models in the prediction of the degradation trend of smart electric energy meters,which verifies the validity and feasibility of the model in this thesis.