Optimize AVR system performance by using enhanced genetic algorithm

Abstract

Automatic voltage regulator (AVR) systems play an important role in adjusting the terminal voltage of synchronous generators. Proportional-Integral-Derivative (PID) and Proportional-Integral-Derivative-Acceleration (PIDA) are two types of controllers that are widely used on this issue. To achieve better performance, optimization techniques are used to obtain the best controller gains values that achieve minimum feedback error. However, most of optimization techniques utilized so far are computationally exhausting or suffer slow convergence. Thus, in this research, the Enhanced Genetic Algorithm (EGA) method is proposed to optimize AVR system with PID and PIDA controllers. EGA enforces local search ability using directed operators without loss of genetic divesity in order to achieve better performance. Simulation results and comparative step response analysis show that EGA outperforms other optimization techniques.