A multi-objective AVR-LFC optimization scheme for multi-area power systems

Abstract

The AVR and LFC schemes are essential power system tools, which are used to support the operation, stability, and security of the generation systems in the main power networks. These control units are designed using numerical as well as analytical approaches. The numerical solutions that are based on heuristics for instance, vary in quality and procedure of the underlying optimization processes leading to the desired outcomes. These processes are mostly used to optimize a single objective criterion that compromises between different performance characteristics like, steady state and transient responses. This work introduces a nonlinear threshold accepting heuristic with an innovative multi-objective optimization architecture. This method is employed to select the control gains of a coupled AVR-LFC scheme for a two-area power system to ensure stable and fast AC generator bus voltage-frequency regulation in react to dynamic disturbances. This approach decides a set of control laws that guarantee fast dynamic response while optimizing conflicting objective criteria. It is validated against other standard heuristics, namely artificial bee colony, differential evolution, and particle swarm optimization in addition to other analytical approaches like internal model-control schemes.