Dynamic Error Estimation for Power Energy Meters Based on Wavelet Decomposition and Outlier Robust ELM
Zhehao Lin, Peng Xiangang, Kaidong Lin, Yi Liu · 2018
In order to evaluate the status of power energy metering, its dynamic error level needs dynamic estimation._The wavelet decomposition and the optimized extreme learning machine(ORELM) is introduced to the dynamic error estimation model. First, the original dynamic error sequence is decomposed into a set of components with the wavelet decomposition. And after predicting the components respectively with the ORELMs-, the final prediction results are obtained by reconstruction and the dynamic error estimation would be completed. The results shows that the proposed method in this paper not only can estimate the dynamic error quantitatively, but also has high accuracy in estimation, and can provide a new way for the calibration of power energy meters and its state detection.