Fractional order PI λD μ controller with optimal parameters using Modified Grey Wolf Optimizer for AVR system

Abstract

In this paper, an automatic voltage regulator (AVR) embedded with fractional order PID (FOPID) is employed for the alternator terminal voltage control. A novel meta-heuristic technique, a modified version of grey wolf optimizer (mGWO) is proposed to design and optimize the FOPID AVR system. The parameters of FOPID, namely, proportional gain ( 𝐾 𝑃 ) , the integral gain ( 𝐾 𝐼 ) , the derivative gain ( 𝐾 𝐷 ) , 𝜆 and 𝜇 have been optimally tuned with the proposed mGWO technique using a novel fitness function. The initial values of the 𝐾 𝑃 , 𝐾 𝐼 , and 𝐾 𝐷 of the FOPID controller are obtained using Ziegler-Nichols (ZN) method, whereas the initial values of 𝜆 and 𝜇 have been chosen as arbitrary values.The proposed algorithm offers more benefits such as easy implementation, fast convergence characteristics, and excellent computational ability for the optimization of functions with more than three variables. Additionally, the hasty tuning of FOPID controller parameters gives a high-quality result, and the proposed controller also improves the robustness of the system during uncertainties in the parameters. The quality of the simulated result of the proposed controller has been validatedby other state-of-the-art techniques in the literature.