Identification of wind turbine main-shaft torsional loads from high-frequency SCADA (supervisory control and data acquisition) measurements using an inverse-problem approach

Abstract. To assess the structural health and remaining useful life of wind turbines within farms, site-specific response modal parameters primary structures are required. In this regard, a novel inverse-problem-based methodology is proposed here to identify dynamic quantities drivetrain main shaft, i.e. torsional displacement coupled stiffness. As model-based approach, an inverse problem mathematical model concerning coupled-shaft dynamics with high-frequency SCADA (supervisory control data acquisition) measurements as input solved. It involves Tikhonov regularisation minimise measurement noise irregularities on shaft obtained from measured rotor generator speed. Subsequently, regularised along necessary used model, system identification method called collage employed estimate also demonstrated that estimated stiffness can be main-shaft loads. shown loads by in good agreement aeroelastic simulations Vestas V52 turbine. Upon successful verification, applied turbine damage-equivalent load. Since does not require design basis or additional sensors, it directly farm possess measurements.