Design of FOPID Controller for Optimizing AVR System using Neural Network Algorithm

Automatic Voltage Regulator (AVR) system is responsible for keeping the terminal voltage of generation unit within safe limits. But the response is slow and oscillatory. To improve this, a controller is used along with AVR system. Fractional-Order-Proportional-Integral-Derivative (FOPID) controllers are popular amongst researchers for providing more degrees of freedom and broad applicability. In this research paper, first, a novel tuning method based on Neural Network Algorithm (NNA) has been proposed to optimize parameters of FOPID controller: K
_P
, K
_I
, K
_D
, λ and μ. NNA is a parameter free algorithm and have a faster convergence rate. Results of proposed controller have been compared with state-of-the-art techniques like Zeigler-Nichols (Z-N), Artificial Bee Colony (ABC), Grass Hopper Optimization Algorithm (GOA) and Multi-Objective Extremal Optimization (MOEO) for set point tracking, noise suppression capability, load rejection capability, controller effort and model uncertainty in various components of AVR system.