Optimization of reference input of maximum power point tracking control for wind turbine

In the maximum power point tracking (MPPT) control of the wind power system, the tracking of the power curve corresponding to the maximum power point (MPP) is a means for obtaining the maximum wind energy efficiency. However, the traditional optimal power curve is not involved with the dynamic response characteristic of wind turbine (WT). Hence, it is not the actually optimal power curve tracked by the rotor speed to capture the maximum wind energy. This phenomenon is especially serious in low-speed wind power generation systems with large moments of inertia of turbines. As a matter of fact, in most servo-track control problems involved with benefit optimization, the tracking to the reference input is not the final aim of the control, instead, it is just a means for the control to obtain a best benefit. Therefore, when the reference input changes too fast relative to the slow dynamics of the controlled object, it will be very difficult to be tracked and then the benefit will be degenerated. To solve this problem, this paper presented a optimization method to design a virtual reference input for replace the original one in the servo-track problem involved with benefit optimization, which is easier to be tracked and hence a higher benefit will be achieved. An application of this method in MPPT control of the wind turbines is implemented on a single mass model. The results show that the suboptimal rotor speed curve can replace the optimal rotor speed curve to obtain higher wind energy capture.