Design and real-time implementation of a fractional-order adaptive PID controller for load frequency control in hybrid renewable power systems

Abstract In recent years, fractional calculus operators have proven to be highly beneficial in electrical engineering applications, such as control theory, signal processing, image processing, electrical circuits theory, transmission lines, photovoltaic curves formulation, and automatic voltage regulator (AVR). Moreover, fractional-order controllers have proven their superior performance in load frequency control (LFC) applications, offering enhanced tuning flexibility, improved dynamic response, and robust disturbance rejection. These advantages make fractional-order controllers particularly suitable for modern power systems, where the growing penetration of renewable energy sources (RES) introduces significant challenges in maintaining frequency stability due to their intermittent and variable nature. Therefore, utilizing fractional-order controllers for LFC in such hybrid systems has become a critical research focus. In this study, a novel fractional-order adaptive proportional-integral-derivative (FOAPID) controller that combines Caputo fractional-order derivatives with model reference adaptive control to enhance system dynamic performance is presented for frequency regulation in a two-area hybrid power system. The parameters of the controller were determined by using trial and error approach and harmony search (HS) algorithm. The first area consists of photovoltaic (PV) and thermal generation, while the second area includes wind and thermal generation. The proposed HS FOAPID and FOAPID controllers are evaluated under multiple operating scenarios, including step load disturbances, renewable energy source uncertainties, and power system parameter variations, system nonlinearities such as generation rate constraint (GRC) and governor dead band (GDB), and are benchmarked against existing controllers reported in the literature. The simulation results demonstrate that the FOAPID controller achieves the shortest settling time of approximately 4 s compared to 12–18 s for the other controllers, while HS FOAPID significantly reduces frequency undershoot by more than 44%. Furthermore, a real-time simulation analysis was conducted using a Speedgoat performance real-time target machine, and the results closely matched the simulation outcomes, confirming the performance of the FOAPID controller. Additionally, FOAPID has shown remarkable improvements in the error indices in reducing the sum of squares error (SSE) and root mean square error (RMSE) by more than 90%. Accordingly, the gained outcomes have highlighted the effectiveness of the proposed HS FOAPID and FOAPID controllers in supporting frequency stability in multi-area hybrid power systems.