The Reliability Prediction of Smart Meters Using Weibull Distribution Based on Fusion of Least Square Estimation and Maximum Likelihood Estimation

摘要

There is a big cost to using on-site inspection methods to test the performance of smart meters. However, using some algorithms and failure information to predict the reliability of smart meters is a useful way to evaluate the state of smart meters. A reliability prediction model based on Weibull distribution is proposed. And a fusion way of least square estimation and maximum likelihood estimation is used to calculate the parameters of Weibull model. Then the test results show the proposal method has a good prediction accuracy and a low RMSE.