Modeling and Design of a PID Controller for Voltage Control of Microgrids Using Hybrid Q-Learning and Particle Swarm Optimization

This paper focuses on the modeling and design of a PID controller for managing voltage and frequency in hybrid microgrid systems. To enhance the performance of these systems, the Q-learning algorithm has been employed as an adaptive method. The primary goal of this research is to increase the stability and efficiency of microgrids. Initially, the mathematical model of hybrid microgrid systems is developed, and then a PID controller is designed for voltage and frequency control. To optimize system performance in response to variations and changing conditions, the Q-learning algorithm has been integrated into this controller. The results demonstrate that using the Q-learning algorithm can improve the performance of frequency and voltage control, adapting to system changes and determining the optimal PID parameters. The Q-learning PI controller is introduced as a more efficient method compared to conventional PI, fuzzy PI, and PSO controllers. This controller is capable of restoring frequency and voltage to their reference states more quickly and stably. Specifically, the Q-learning PI controller has shown superior performance in frequency control, whereas the conventional PI controller returns to the desired state with slower speed and less accuracy, and the fuzzy PI controller failed to achieve satisfactory performance. Consequently, the Q-learning PI controller is recognized as a more effective and superior method for frequency control in microgrids.

کلیدواژه ها:

PID Controller Design ، Hybrid Microgrids ، Frequency Control ، Voltage Control ، Q-Learning Algorithm