AN INTELLIGENT SWITCHING OVERVOLTAGES ESTIMATOR FOR POWER SYSTEM RESTORATION USING ARTIFICIAL NEURAL NETWORK

摘要

One of the most important issues in power system restoration (PSR) is switching overvoltage caused by power equipment energization. This phenomenon may damage some equipment and delay power system restoration. This paper proposes an intelligent estimator which can be used to evaluate and control switching overvoltages. Transformer, shunt reactor, and transmission lines are important devices in PSR and their energization have been studied in this work. Artificial neural network (ANN) is adopted as an intelligent approach to deal with these overvoltages. Both Multilayer Perceptron (MLP) and Radial Basis Function (RBF) structures have been analyzed. Five learning algorithms, backpropagation (BP), delta-bar-delta (DBD), extended delta-bar-delta (EDBD), directed random search (DRS), and quick propagation (QP) were used to train the MLP. In the cases of transformer and shunt reactor energization, ANNs are trained with the worst case scenario of switching angle and remanent flux which reduce the number of required simulations for training ANN. Also, for achieving good generalization capability for developed ANN, equivalent parameters of the network are used as ANN inputs. The simulated results for a partial of 39-bus New England test system show that the proposed technique can estimate the peak values and duration of switching overvoltages with good accuracy and EDBD algorithm presents best performance.