Simple neuro-controller with a modified error function for a synchronous generator

A simple neuro-controller for a synchronous generator is presented in this paper. The controller consists of a single neuron trained on-line using the back propagation algorithm based on a modified error function. The modified error function uses the controlled variable and its derivatives. Based on the variables used in the modified error function, the controller performs the functions of terminal voltage control and power system stabilizer. Simulation studies for a single-machine and multi-machine power system environment are presented to demonstrate the effectiveness of using the simple neuro-controller. (C) 2003 Elsevier Science Ltd. All rights reserved.