NEURAL NETWORK SLIDING MODE CONTROLLER FOR A VARIABLE SPEED WIND TURBINE

In this paper, a neural network sliding mode controller is proposed for a variable speed wind turbine. Below the rated wind speed, the main objective of the controller is to maximize the energy captured from the wind and minimize the stress on the drive train shafts. Sliding mode control approach (SMC) can be used for a variable speed wind turbine. However, in the presence of large uncertainties the SMC produces chattering phenomenon due to the higher switching gain. To reduce this gain, neural network with one hidden layer is used to uncertain parts of the system plant. This neural network is trained online using the backpropagation algorithm. A second robust control term is added to compensate the neural network errors. The performance of the proposed approach is investigated in simulations by comparison with traditional sliding mode control.