Radial Basis Function Neural Network Based Comprehensive Evaluation for Power Quality

The power quality comprehensive evaluation (PQCE) is the foundation of deciding the price according to the quality of power commodity as well as a part of the ancillary service in power market. A new method based on the radial basis function (RBF) neural network for PQCE is proposed in the paper. According to the national standards of P.R. China, this paper made the grades of each power quality index. Efficient samples based on the random-distribution theory were produced to train the network. The RBF neural network with high non-linearity-fitting capacity was applied to the PQCE, which overcame the shortages such as uncertainty and man-influenced in fuzzy mathematics, probability and mathematical statistics and analytic hierarchy process (AHP). The RBF neural network-based model possessed the objective and rationalization. The practical example for PQCE on the substations approved that the proposed approach is reasonable and feasible.