ON DECENTRALIZED SECONDARY VOLTAGE CONTROL VIA ARTIFICIAL NEURAL NETWORKS

In the paper, an implementation of the artificial neural network (ANN) in a decentralized secondary voltage control of the power system is presented. A fully diaributed modular secondary voltage controllers at generator busses is proposed. The ANN is trained on series of simulated power system states, obtained with a help of a computer program for optimal power flow calculation. Performance of the trained ANN controller has been evaluated during normal power system operation and for some power line outages. Its coordinating role in case of two adjacent generators is discussed.