Fast terminal integral sliding mode pitch angle control based on hierarchical power coefficient estimator for wind turbine

Wind turbines show substantial nonlinearity because of the uncertainties and highly nonlinear aerodynamic features. This nonlinearity becomes particularly pronounced while operating in the full load region, where linear controllers exhibit suboptimal performance. Therefore, it is advised to employ a nonlinear controller, yet it necessitates implementing additional sensors. To address these issues, this study introduces three novel sensorless fast terminal wind turbine pitch angle controllers. Namely, Sliding Mode (ISM), Terminal Integral sliding mode (TISM), and Fast Integral Sliding Mode (FISM). These controllers are based on a synthesized mathematical model that expresses the pitch angle as a function of wind speed. A hierarchical intelligent estimator is developed and verified to enhance the power coefficient observability. This estimator employed the Ordered Weighted Averaging (OWA) operator to integrate three intelligent estimators: multilayer perceptron neural network (MLP-NN), radial basic function neural network (RBF-NN), and adaptive neuro-fuzzy inference system (ANFIS). This is based on measurements of the pitch angle and tip speed ratio. The stability of the proposed sensorless three controllers (ISM, TISM and FISM) was assessed using Lyapunov stability criteria. The efficacy of the three control approaches is then tested on Controls Advanced Research Turbine (CART 2) wind turbines using FAST. The proposed controllers present a significant improvement of wind turbine speed fluctuations’, recording a reduction of standard deviation of the signal from 0.45 for baseline controller to 0.08 (rpm) for the proposed FTISMCE. Subsequenty a smoothy signals are obtained for other variables measurement. A sensitivity analysis is carried out, where it was observed that the controllers, operate under a maximum of 7° as additive disturbance on pitch.