Assessment and control of wind turbine by support vector machines

Abstract

In this paper, a control strategy by support vector machines (SVM) for a wind turbine is proposed. It aims at operating the wind conversion system perfectly so that it can extract the available maximum power in wind flow under normal conditions, also, the system should be stabilised by this control when the wind is stronger. The SVM’s role is identifying the behaviours of optimal electromagnetic torque and blade pitch angle with respect to wind speed changing. Therefore, data should be collected at first. Among of these data:wind speed, the matching optimal torque and pitch angle. SVM model may be designed through using these data show. Yet, SVM model ought to be used correctly through providing us optimal electro magnetic torque and optimal pitch angle for becoming wind system controller. To show the extent of reliability compared to other controllers in the literature, the method what is called sliding mode control is proposed. The latter is going to be designed and applied for comparing it with SVM. Finally, the two methods (SVM, SMC) are testified on real wind profile.