Fire Hawk Optimizer Adjusted Tri-Stage (1+PI)-PI-TID Cascade Controller for Automatic Generation Control of PSI

Abstract

This study aims to enhance the performance of automatic generation control (AGC) in power systems interconnection (PSI) by employing the fire hawk optimizer (FHO), a green algorithm, to optimize a novel tri-stage cascade controllers (CCs) configuration: (1+PI)-PI-PID and (1+PI)-PI-TID CCs. The FHO draws inspiration from the foraging behaviors of whistling kites, black kites, and brown falcons, known for their unique prey-catching method involving fire propagation. Evaluation of the proposed FHO: (1+PI)-PI-T/PID scheme is conducted using the integral time multiplied absolute error (ITAE) criteria. The versatility of this technique is demonstrated across four PSI test models, including a two-area non-reheat thermal PSI (NRTPSI) and a multi-unit multi-area (MUMA) system incorporating reheat, hydro, and gas units. Challenges are addressed by initially testing NRTPSI with and without governor dead band (GDB) nonlinearity, showing the superior performance of the proposed method. Next, the HVDC link in MUMA is employed under different scenarios. To assess the robustness of the FHO: (1+PI)-PI-T/PID CCs, real world AGC issues, such as sudden load perturbations and nonlinearities like generation rate constraint, GDB, boiler dynamics, and communication time delay in MUMA, are employed. Also, sensitivity analysis involving a range variation of }40% in PSI model parameters is conducted for validation. The proposed scheme also undergoes evaluation under random step-load perturbations. The results demonstrate significant improvements in the performance of ITAE by 84.3%, 10.7%, 42.1%, and 16.7% for scenario 1 across all four tested models. The FHO: (1+PI)-PI-TID CC significantly improves system performance compared to existing controllers like the HHO: PD-PI CC, FHO: TID, FHO: PID, and the proposed FHO: (1+PI)-PI-PID CCs. This shows that this approach is effective at improving AGC capabilities and can be used in real life. Update each group’s position using their respective equations