Three-dimensional online dynamic deformation monitoring of wind turbine blades

Online monitoring of the dynamic responses of wind turbine blades is an effective way to reduce maintenance costs and improve their operating life. This paper proposes a new method for three-dimensional (3D) online dynamic deformation monitoring of wind turbine blades by combining the mode superposition method and piezoelectric sensors. The approximate mode shape functions of flap-wise, edge-wise, and twist displacements for the wind turbine blade model are obtained based on the function fitting method and finite element modal analyses. The conversion method between the mode shape functions, the dynamic responses of blades, and the signals of piezoelectric sensors are derived according to the mode superposition theory. The least squares method and the physics-informed neural network (PINN) are used to obtain the correction parameters of the transformation parameters. Furthermore, 3D dynamic deformation monitoring experiments are conducted on the wind turbine blades. The dynamic displacements and twist angle of the blades collected by the electronic gyroscope are used as the calibration data. It is shown that the proposed method can accurately reconstruct the 3D dynamic deformations of wind turbine blades. The method developed in this study provides a new technical direction for the dynamic deformation monitoring of wind turbine blades and has application potential in many complex engineering structures in other fields.