An intelligent wide area system-centric controller and observer for power system stabilization using optimal Dual Heuristic Programming (DHP) architecture

In this paper, we propose an intelligent approach to power system stabilization using a Wide Area System Centric Controller and Observer (WASCCO). This architecture augments local controllers for stabilization connected to generators such as Power System Stabilizers (PSS) based on Intelligent Supervisory Loop (ISL) concept implemented with a wide area controller and Observer. For the monitoring and controller of multiple generators a Dual Heuristic Programming (DHP) action-critic neural network architecture is utilized. This coupled with a predictive Wide Area Neural Network Identifier (WANNID) generates a control signal to augment local controller. The controller performance is tested on a five generator, eight buses, two area power network to damp inter-area model oscillations. Simulation studies indicate that the proposed controller is capable of improving the stabilization of the generator over the use of PSS alone while continuously improving its performance through the use of online learning.