Optimal FOPID and RPIDD2 Controller Design Using Meta-Heuristic Optimization for AVR System

In electric power systems, adjusting the generator voltage through automatic voltage regulator (AVR) systems constitutes a very important problem for ensuring a stable voltage level and reliable electricity distribution. This paper aims to present the optimal design of fractional-order proportional-integral-derivative (FOPID) and real PID plus second-order derivative (RPIDD2) controllers, which have been designed not only to preserve the voltage of the power system within the permitted limits but also enhance its stability. The particle swarm optimization (PSO) has been implemented for the optimal tuning of the controller parameters, thereby achieving enhanced AVR dynamic performance. The search process is performed based on one of the current objective functions, which is a composition of Zwe-Lee Gaing (ZLG) and integral-time absolute error (ITAE) objectives. For this study, the AVR system is modeled in a simulation environment with known standard parameters, and a comparative analysis is conducted. The results of the extensive study demonstrate that the five-parameter FOPID controller exhibits superior dynamic performance, while the six-parameter optimized RPIDD2 controller is capable of achieving even more excellent dynamic performance. It is also shown that more impressive results are achieved than some optimized structures before. Furthermore, a basic robustness analysis is conducted to evaluate the system's response to parameter changes.