Control of mechanical loads in wind turbines using an integrated aeroelastic model

This paper presents the integration of an artificial neural network (ANN) based control algorithm for minimizing the stress of a wind turbine (WT) operating above-rated wind speeds, and a coupled aeroelastic model derived from structural blade element momentum (BEM) and thin-walled beam (TWB) theory. The coupled BEM/TWB model is used for real-Time determination of stress, strain and displacement in arbitrary positions of rotating blades. The controller selected is a proportionalintegral-derivate (PID) whose tuning parameters are calculated online through an Adaline ANN. The results show significant improvements in the stress reduction on the blades through the verification of a normalized stress factor. The stress reduction with the proposed control algorithm is up to 27%.