Wind estimation with a non-standard extended Kalman filter and its application on maximum power extraction for variable speed wind turbines

To maximize power extraction at below-rated wind speeds, variable-speed wind turbines must be controlled by tracking the optimal TSR (tip speed ratio) and pitch angle, which depend on the wind speed measured by nacelle anemometers or provided by an EWS (effective wind speed) estimator. However, the measured values are imprecise and existing estimators cannot provide qualified estimates. This paper addresses this problem by presenting a novel solution with a non-standard extended Kalman filter. To avoid using imprecise wind speed measurements or other costly measurement devices, the proposed solution employs a virtual measurement that is calculated from related estimated states. In addition, the solution presents an internal EWS model by considering the tower shadow effect, so the obtained model is more general than the statistical model that is difficult to obtain in practice. Compared with existing estimators, the proposed estimator provides more precise estimated results and is suitable for control application. Its application is investigated on the MPE (maximum power extraction) of a variable speed wind turbine, for which an industrial baseline controller is optimized by enhancing the optimal TSR tracking and pitch adjustment. The proposed solutions are validated using both simulation and field testing results. Comparing the proposed estimation solution to two existing methods demonstrates that the former gives the best estimate results. Moreover, its application for the MPE increases annual energy production by approximately 0.8% in comparison with the baseline controller, which is a considerable energy production increment.