Adaptive optimal wide-area controller for multi-area power system using phasor measurement unit measurements

A neural network-based adaptive optimal controller for the damping of inter-area low-frequency electromechanical oscillations inherent to multi-area power systems is proposed. The controller is designed based on the linear quadratic regulator, which is usually designed based on the linear model obtained about an operating point (OP). As the power system is non-linear and dynamic, the optimal condition will only be satisfied at the designed OP. This proposed controller utilises an adaptive structure so that the controller can effectively act in all operating conditions. The phasor measurement unit measurements provide the remote feedback input to the controller. The generator states are estimated using the unbiased minimum variance filter so that the estimation can be done in the absence of generator input measurements. The performance of the proposed structure under various operating conditions and dynamic conditions are analysed in this study.