Enhanced nonlinear sliding mode control technique for wind power generation systems application: Theoretical design and comparative study

Abstract

Wind power generation systems (WPGSs) have emerged as a vital source of clean energy, appreciated for their renewable nature and zero fuel costs. However, controlling these systems is challenging due to their nonlinear dynamics, unpredictable disturbances, parameter uncertainties, and rapid wind speed fluctuations. To address these challenges, robust control strategies are essential. This study presents a novel robust nonlinear controller for regulating the electromagnetic torque of horizontal-axis, variable-speed WPGSs with three blades connected to the grid. The proposed controller is designed to enhance system efficiency and profitability by maximizing electricity production, reducing torque ripples, eliminating peak overshoots, and achieving precise setpoint tracking with optimal transient performance. The controller combines sliding mode (SM) control with fractional calculus (FC) to exploit the benefits of both methods. SM control, known for its effectiveness in controlling nonlinear systems, stabilizes the system and ensures finite-time convergence to the desired state. The incorporation of fractional-order operators into the sliding surface introduces greater flexibility, leveraging the long-term memory properties of FC to enhance system stability and robustness while mitigating chattering effects. The stability of the proposed controller is rigorously validated through Lyapunov theory. The simulation phase is carried out in MATLAB under various wind conditions and operating scenarios demonstrate the controller's superior performance. The results show its ability to reduce chattering phenomena, improve power quality, and minimize high peaks in the input controller by achieving reductions of 38.0856 KN.m in the first test and 280.2039 KN.m in the second test. Additionally, this controller ensures system stability, with a Lyapunov function standard deviation of 0.0105 and 0.1755 for the first and second tests, respectively. The controller also achieves a high efficiency of 47.45%, robustly driving the system to its desired state in finite time with tracking error standard deviations of 0.3171 in the first test and 0.1652 in the second test.