Robust decentralized L.F.C. with online iteraction trajectory improvement

Abstract

A Model-following decentralized algorithm to design near-optimal load frequency regulator for large interconnected dynamic power systems is considered. By using a simple reduced-order model for the interaction between subsystems, a decentralized controller whose parameters are independent of initial states and which can accommodate constant disturbances, is designed. No information transfer between different sub-systems is necessary. The choice of the crude model of the interaction is scrutinized in order to develop a feedback controller which fits the load frequency problem. The decentralized algorithm is applied to a multi-area power system to control the load and frequency. The performance of the resulting controller is compared to that of the known integrated optimal controller. The algorithm is also tested on different system configurations to prove its efficiency.