Dynamic neural network for AGC in restructure power system

Abstract

In This paper, a new adaptive controller based on unsupervised learning approach, named feedback error learning (FEL), is proposed for automatic generation control (AGC) of power system. In the FEL strategy, both feedforward and feedback controller are used for control of process, simultaneously. Generally, the feedback controller contains the classic controller, i.e. PID controller, and the feedforward controller is a neural network based controller. In this paper dynamic neural network (DNN) is used for feedforward controller. The DNN have some memory in his structure and improved the overall performance. The proposed FEL controller has been compared with the conventional FEL (CFEL) controller and the PID controllers for two areas restructure power system.