Application of Expected Sarsa-Learning for Load Frequency Control of Multi-Area Power System

Abstract

This paper presents expected sarsa-Iearning algorithm-based control strategy for load frequency control of two-area power system. This algorithm comes under value- teration-based model-free reinforcement learning algorithm. The pre-learning for this algorithm is carried out on two area power system. Then the effectiveness of the proposed control strategy is evaluated on the modified two-area power system. DFIG-based wind power generation is incorporated into each area of the modified power system. The simulation study is demonstrated to analyse the performance of the proposed controller. © 2023 IEEE.