Enhanced wind energy conversion system performance using fast smooth second-order sliding mode control with neuro-fuzzy estimation and variable-gain robust exact output differentiator

Wind energy conversion systems (WECS) often face challenges due to the stochastic and intermittent nature of wind, leading to a mismatch between output power generation and fluctuating electrical load requirements. To effectively address this issue, an advanced Maximum Power Point Tracking (MPPT) strategy is crucial for maximizing power extraction. This study introduces a novel MPPT approach based on Fast Smooth Second-Order Sliding Mode Control (FSSOSMC) to optimize power output from a 3 kW fixed-pitch variable speed WECS coupled with a permanent magnet synchronous generator (PMSG). To ensure robustness despite uncertain, nonlinear system parameters, an offline neuro-fuzzy algorithm utilizing the Takagi–Sugeno–Kang (TSK) fuzzy inference system is implemented. This method accurately estimates the nonlinear and uncertain components of the control input, enhancing the performance and robustness of the MPPT control techniques. Additionally, a Variable-Gain Robust Exact Output Differentiator (VG-REOD) is employed to accurately estimate the shaft speed, addressing issues related to speed estimation and missing derivative information. The proposed FSSOSMC-based MPPT strategy was benchmarked against super-twisting sliding mode control (STSMC) and feedback linearization control (FBLC) MPPT strategies under stochastic wind speed profiles, parameter variations, and wind speed fluctuations. The results show that the proposed method achieves a tracking accuracy of 98.2% and an overall efficiency of 98.9%, significantly outperforming STSMC (97.1% accuracy, 95.75% efficiency) and FBLC (95.4% accuracy, 93.82% efficiency). The FSSOSMC method also reduced the settling time to 7.879 s and the rise time to 1.062 s, with minimal overshoot of 10.022% and a steady-state error of 0.0015088. These results demonstrate superior tracking performance, high precision, rapid dynamic response, minimal chattering, and robust global performance. The efficacy of the proposed method is validated through extensive MATLAB/Simulink simulations. •Utilizes Fast Smooth Second-Order Sliding Mode Control.•Offline neuro-fuzzy algorithm enhances robustness.•Incorporates Takagi–Sugeno–Kang fuzzy inference system.•Implements Variable-Gain Robust Exact Output Differentiator.•MATLAB/Simulink simulations under various wind profiles.