Model predictive controller for performance enhancement of automatic voltage regulator system

Automatic voltage regulator (AVR) is a control circuit dedicated to stabilizing the terminal voltage of the synchronous generator whenever deviated from a predetermined setpoint. However, utilizing such this control circuit only to adjust the terminal voltage of the synchronous generator shows some deficiencies from the robust control system point of view. Significant criteria should the control system take it into consideration, which are the transient and steady-state characteristics. In this paper, a model predictive control (MPC) is introduced to enhance the output response of the AVR. Moreover, a realistic AVR control circuit is identified in this paper, where the saturation characteristic of the excitation system is incorporated into the AVR modeling circuit. To ensure the MPC fulfills an effective and outstanding performance, the proposed technique is compared with the proportional-integral-derivative (PID) controller tuned with the genetic algorithm (GA) executing the same fitness function. The results obtained by MPC and GA are validated using MATLAB-Simulink and MATLAB toolbox. This paper proves that the MPC is able to improve the terminal voltage of the synchronous generator more than the PID controller tuned with GA does.