Abstract: This article proposes the use of a soft actor–critic (SAC) algorithm-based reinforcement learning (RL) controller as the only primary controller to improve the dynamic performance of the ...