Cyberattack resilient load frequency control in renewable integrated power system considering controlled energy storage

The rapid proliferation of renewable energy sources (RESs), driven by stringent decarbonization policies and global sustainability initiatives, has introduced critical stability challenges to modern power grids. The displacement of conventional synchronous generators by RESs leads to a significant decline in system inertia, thereby impairing the frequency stability. This instability is further exacerbated by the stochastic nature of renewable energy-based generation and unpredictable load variations in interconnected multi-area power systems. To counteract these effects, this study employs a controlled energy storage strategy utilizing electric vehicles (EVs) through vehicle-to-grid (V2G) technology and redox flow batteries (RFBs) to reinforce frequency regulation. Additionally, system robustness is enhanced through the deployment of flexible alternating current transmission systems (FACTS) devices like the unified power flow controller (UPFC) for dynamic real and reactive power management and high voltage direct current (HVDC) transmission link operating in parallel with the AC network to boost controllability and transmission reliability under high RES penetration. Furthermore, the increasing reliance on communication-based infrastructure in smart grids enhances control capabilities but also exposes the load frequency control (LFC) loops to cyber threats, including denial of service (DOS) and false data injection (FDI) attacks. To address this, the present study proposes a novel cascaded control scheme combining a fractional-order proportional integral derivative with filter plus fractional derivative controller (PI^λ₁D^μ₁ND^μ₂) and a tilt fractional-order integral (TI^λ₂) controller and walrus optimizer (WO) is employed to precisely tune the controller parameters using an integral of the time square error (ITSE) performance index. This nature-inspired metaheuristic offers a robust global search capability while maintaining computational efficiency. WO outperforms the well-established optimization techniques, including sine cosine algorithm (SCA), zebra optimization algorithm (ZOA), sand cat swarm optimization (SCSO), and coot algorithm (CA). The findings indicate that the WO tuned cascaded PI^λ₁D^μ₁ND^μ₂-TI^λ₂ controller demonstrates significant improvements of 96.60%, 94.8%, 75.5%, and 37.5% compared to the ZOA optimized CC PI^λ₁D^μ₁ND^μ₂-TI^λ₂, SCA tuned CC PI^λ₁D^μ₁ND^μ₂-TI^λ₂, SCSO based CC PI^λ₁D^μ₁ND^μ₂-TI^λ₂, and COOT optimized CC PI^λ₁D^μ₁ND^μ₂-TI^λ₂ controllers, based on ITSE performance index. The performance of the proposed cascaded PI^λ₁D^μ₁ND^μ₂-TI^λ₂ controller has been rigorously evaluated across multiple challenging scenarios, including various load profiles, high penetration of variable renewable energy resources, cyber-attacks (DoS and FDI) and communication time delay while considering system inherent system nonlinearities such as generation rate constraints (GRC) and governor dead band (GDB). Simulation results demonstrate the superiority of proposed cascaded PI^λ₁D^μ₁ND^μ₂-TI^λ₂ controller over TI^λ-PD^μN, cascaded FOPID-FOPD and cascaded (1+PI)-(T+(1+I^λD^μ)) controllers in achieving up to 88.4 % reduction in frequency overshoots, 85.8 % in tie-line power overshoot, 75.3 % in frequency undershoots, and 70.4 % in tie-line power undershoot, demonstrating its strong capability in suppressing transient deviations and enhancing system stability across all scenarios. The robustness of the proposed controller is validated under variations in system parameters, renewable energy permeability, and energy storage capacity. Furthermore, stability is confirmed through Bode-based frequency stability analysis.