Combination of Continuous Action Reinforcement Learning Automata and PSO toDesign a PID Controller for AVR System

This paper presents a hybrid approach involving Continuous Action Reinforcement Learning Automata(CARLA) and particle swarm optimization (PSO) to design a optimal and intelligent proportionalintegral-derivative (PID) controller of an automatic voltage regulator (AVR) system. The proposedmethod is CARLA which is able to explore and learn to improve control performance without theknowledge of the analytical system model. The role of an AVR is to hold the terminal voltagemagnitude of a synchronous generator at a specified level. Hence, the stability of the AVR systemwould seriously affect the security of the power system. CARLA-PSO is a method that combines thefeatures of PSO and CARLA in order to improve the optimize operation. The proposed method wasindeed more efficient and robust in improving the step response of an AVR system. Numericalsimulations are also provided to verify the effectiveness and feasibility of PID controller of AVR basedon CARLA-PSO algorithm.