Adaptive Neuro-fuzzy Algorithm for Pitch Control of Variable-speed Wind Turbine

With increasing size of wind turbines (WTs), the power regulation and fatigue loads on WT structures emerge as major problems to wind power industry. Pitch angle is scheduled above the rated wind speed to keep the power captured by variable-speed wind turbine (VSWT) around its rated value and release the fatigue load on WT structure. In this paper, a hybrid intelligent learning based adaptive neuro-fuzzy algorithm is proposed to schedule the pitch angle of 2 Megawatt (MW) VSWT. The artificial neural network (ANN) trains the parameters of fuzzy membership functions (MFs) using least squares estimator method in forward pass and back propagation gradient descent method in backward pass. The simulation is done in MATLAB and results are compared with multilayer perceptron feed-forward neural network (MLPFFNN) and fuzzy logic-based pitch controllers. The results indicate the effectiveness of proposed neuro-fuzzy algorithm which outperforms the other two methods.