New Tilt Fractional-Order Integral Derivative with Fractional Filter (TFOIDFF) Controller with Artificial Hummingbird Optimizer for LFC in Renewable Energy Power Grids

Abstract

Recent advancements in renewable generation resources and their vast implementation in power sectors have posed serious challenges regarding their operation, protection, and control. Maintaining operating frequency at its nominal value and reducing tie-line power deviations represent crucial factors for these advancements due to continuous reduction of power system inertia. In this paper, a new modified load frequency controller (LFC) method is proposed based on fractional calculus combinations. The tilt fractional-order integral-derivative with fractional-filter (TFOIDFF) is proposed in this paper for LFC applications. The proposed TFOIDFF controller combines the benefits of tilt, FOPID, and fractional filter regulators. Furthermore, a new application is introduced based on the recently presented artificial hummingbird optimizer algorithm (AHA) for simultaneous optimization of the proposed TFOIDFF parameters in the studied two-area power grids. The contribution of electric vehicle (EVs) is considered in the centralized control strategy using the proposed TFOIDFF controller. The performance of the proposed TFOIDFF controller has been compared with the existing tilt with filter, PID with filter, FOPID with filter and hybrid fractional-order with filter LFCs from the literature. Moreover, the AHA optimizer results are compared with the featured LFC optimization algorithms in the literature. The proposed TFOIDFF and AHA optimizer are validated against renewable energy fluctuations, load stepping, generation/loading uncertainty, and power-grid parameter uncertainty. The AHA optimizer is compared with the widely-used optimizers in the literature, including the PSO, ABC, BOA, and AEO optimizers at the IAE, ISE, ITAE, and ITSE objectives. For instance, the proposed AHA method has a minimized IAE after 34 iterations of 0.03178 compared to 0.03896 with PSO, 0.04548 with AEO, 0.04812 with BOA, and 0.05483 with ABC optimizer. Therefore, fast and better minimization of objective functions are achieved using the proposed AHA method.