Three‐Level NF‐TDN Controller for AGC in Thermal‐Hydro‐Gas Power Systems Integrated With TCPS Under Nonlinear Conditions

Automatic generation control (AGC) is vital for preserving frequency and power balance in modern interconnected power systems (IPSs), which are increasingly challenged by nonlinear dynamics, parameter uncertainties, and unpredictable load variations. This study introduces a three‐level neuro‐fuzzy‐tilt‐derivative controller with a low‐pass filter ( 3 TLNF‐TDN), designed to achieve intelligent, robust, and adaptive control performance in AGC applications. The controller parameters are optimally tuned using the secretary bird optimization algorithm (SBOA), a recently developed metaheuristic inspired by the predatory behavior of secretary birds. To validate the effectiveness of the proposed approach, a three‐area hydro‐thermal‐gas power system model incorporating realistic nonlinearities is developed. The performance of the 3 TLNF‐TDN controller is comprehensively compared with several benchmark AGC controllers, including the tilt‐derivative with filter coefficient (TDN) controller, the single‐level neuro‐fuzzy TDN ( 1 SLNF‐TDN) controller, and the two‐level neuro‐fuzzy TDN ( 2 TLNF‐TDN) controller. Simulation results demonstrate that the proposed controller exhibits markedly superior dynamic performance, yielding faster settling time (ST), reduced overshoot (OS)/undershoot (US), and lower frequency oscillations, alongside improved performance indices under varying load disturbances and parameter uncertainties. Moreover, the integration of a thyristor‐controlled series capacitor (TCSC) further enhances tie‐line power flow and frequency stability. Overall, the results confirm that the SBOA‐optimized 3 TLNF‐TDN controller provides a high degree of robustness, adaptability, and practical feasibility for real‐world AGC implementations.