Wind Turbine Control in Partial Load Operation by Using the Optimization Algorithms from the Deep Learning Techniques

The most common operation of wind turbines is in the case of under-rated wind speed, i.e., when the wind speed cannot reach the rated wind speed. In this situation, the control system has to bring the machine to an operating point that makes possible the maximum power extraction. To this end, many control strategies and control approaches are available. Two of them are the Optimal Torque Control (OTC) and the Hill-Climb Search Control (HCS). On the other hand, machine learning activities have signifi-cantly improved several optimization algorithms in recent years, which can also be used with HCS. In the present work, some of these algorithms are selected and investigated with the goal of being used with HCS combined with OTC for a joint goal of power extraction maximization and rotor inertia reduction in very large wind turbines. The interest is set on the evaluation of the algorithms for the real-time control implementation.