Adaptive PID Tuning for DC Motors using Machine Learning

Direct Current (DC) motors are extensively used in industrial applications for their versatility in speed control and adaptability. This paper introduces a Reinforcement Learning (RL)-based approach to optimize Proportional-IntegralDerivative (PID) controller parameters, utilizing the Twin Delayed Deep Deterministic Policy Gradient (TD3) algorithm to address the non-linearity inherent in DC motor systems. The proposed method aims to improve energy efficiency, enhance speed regulation, and ensure stable performance across varying operational conditions. Practical validation of a DC motor is performed using MATLAB/Simulink simulations to analyze performance, focusing on speed control, torque, and efficiency.