Research on power quality prediction for DG integrated smart grid based on neural network

In this paper, a new approach based on neural network is proposed to availably predict the power quality (PQ) for smart grid with distributed generation (DG) integration. We first analyze the various factors that have remarkable impact on the variation tendency of PQ, such as climatic variation, fluctuation of power loads, etc. Then we put forward a general prediction process for steady-state index of power quality. Three neural networks with different structures, including Back Propagation (BP), Radical Basis Function (RBF) and General Regression Neural Network (GRNN) whose feasibilities and features are contrastively analyzed are used to build the prediction models for PQ. Simulation study is carried out on the IEEE 13 node test feeder, and the results demonstrate that the proposed prediction approaches for the steady-state PQ indexes based on neural networks are practicable, in which the prediction model based on the GRNN has the best performance.