Performance Evaluation of Smart Meters Based on Grey Relational Analysis

Based on the full performance test data of a large number of smart meters, a performance evaluation method for smart meters based on grey correlation analysis is proposed, aiming at the situation that lack of quality evaluation of smart meters in China. In this paper, six sets of representative test data of smart meters are selected as performance evaluation indexes. First, correlation analysis is used to demonstrate the independence between evaluation indexes. For each index, the weight value of different test points are obtained by principal component analysis, thus the evaluation index matrix is acquired. Then grey relational analysis is applied to construct the performance evaluation model on the basis of evaluation index matrix. Subsequently the performance evaluation results of samples can be aquired. At last, smart meter test data from three different random manufacturers were selected to prove this method. Considering the historical statistics of the field failure rate and expert opinion for each manufacturer, the evaluation results using this method are reasonable, thus this performance evaluation method for smart meters is reliable.