A power boosting controller is a controller which increases the generator torque in order to increase the wind turbine’s nominal output power by up to 5%. This increase in power production can lead to an increase in the wind turbine’s AEP (annual energy production) by up to 2%.
The current master thesis, starts with the development of a 5 MW NREL wind turbine nonlinear mathematical model which is linearized twice, once for being used in the design of the 퐻∞ standard robust controller, and once for being used in the design of the 퐻∞ power booster robust controller. Next, the parametric uncertainties are added to the LTI (linear time invariant)/ss (state space) model and then reconstructed for the control design purposes. Two 퐻∞ optimal controllers are obtained for operating the wind turbine in two modes: standard and the power boosting mode. The logic for the boosting mode was built to switch between the controllers when certain wind speeds are reached. A pre-analysis for determining the stability, performance and robustness was executed on the open loops and gain loops bode diagrams and pole-zero maps.
The response, robustness and performance of the 퐻∞ robust controllers with and without the power booster are compared with the gain scheduled PI controller (base line). It is discovered that the pitch angle of the baseline controller variates more than the other controllers and there is not a significant difference in pitch angle between power booster and standard H-infinity controllers. On the other hand, the generator speed and torque of the base line controller oscillates less than the other controllers. The insignificant difference between boost and standard H-infinite controllers is kept just for the generator speed, while for the generator torque the difference becomes more clearly. This fact can be explained by the significant difference between linearization and operation points associated with the generator torque. From the analysis, it can be also seen that the generator speeds of the two H-infinity robust controllers is decreasing in higher speeds whereas the generator torque increases dramatically.
By analyzing the output power response of the different controllers, the H-infinity robust controller with power booster, boosts without any difficulties the power from 5 MW to 5.25 MW when the defined conditions (the wind speed lays between 15 and 22 m/s) are meet.
The visual analysis of the tower and blade moments is also confirmed by the fatigue analysis with the DELs (damage equivalent loads) calculated based on the standard deviation of the bending moments. Finally, it can be concluded, that by adding a power booster to h-infinity controller increases the tower fatique loads just 1.2% and the blades fatique loads just 1.07%.