Structural health monitoring of the monopile foundation structure of an offshore wind turbine

The structural components of a monopile foundation consist of the monopile (MP), the transition piece (TP), the grouted TP/MP connection and the seabed subsoil. The stiffness of the foundation structure has a high impact on the fundamental structural resonance and on the inclination (tilt) of the complete structure. A potential degradation of any such stiffness might move the fundamental resonance frequencies closer to the first harmonic of the rotor rotation frequency resulting in increased dynamic loads. Continuous identification of the foundation stiffness and the inclination together with the fundamental resonance frequencies were therefore considered the primary monitoring goals. The foundation stiffness is defined by the ratio of the bending moments and shear forces with respect to the inclination at TP-airtight platform level. The bending moments and shear forces were calculated from the strains measured in three directions at five locations along the TP-circumference. The inclinations were measured by a 2-axial inclination sensor located at the same TP level. The bending moments and shear forces were correlated with the inclinations and also with the environmental and operational (EO) data (wind velocity, rotor frequency, nacelle orientation and generated power). The state indicators are based on two different data sets: (a) the long-term distribution of the 10-minute averages and (b) the time histories of the strain and inclination data which are measured with a sampling rate of 50 Hz. The time histories were used for extracting the fundamental bending frequencies continuously over the monitoring period via time-frequency analysis based on 5-minute time intervals. Operational modal analysis (OMA) techniques were applied to the data of each interval. The measurement system (MPC-system) designed by the University of Kassel was installed on the TP airtight platform and connected via internet to the university lab-PC. The raw test data are stored locally by the MPC-system and can be downloaded to the lab-PC on demand for data evaluation. Results from continuous monitoring over a period of 20 months show the evolution (trend) of the derived state indicators. No significant variations of the indicators were observed. The indicators are considered to represent a valuable decision tool for condition based maintenance.