Prediction of remaining fatigue life of welded joints in wind turbine support structures considering strain measurement and a joint distribution of oceanographic data

• Published in: Marine structures Vol. 66; pp. 307 - 322
• Main Authors: Mai, Quang A., Weijtjens, Wout, Devriendt, Christof, Morato, Pablo G., Rigo, Philippe, Sørensen, John D.
• Format: Journal Article Web Resource
• Language: English
• Published: Barking Elsevier Ltd 01.07.2019; Elsevier BV; Elsevier
• Subjects: Bayesian analysis; Civil engineering; Corrosion; Crack propagation; Distribution; Engineering, computing & technology; Fatigue; Fatigue failure; Fatigue life; Ingénierie civile; Ingénierie, informatique & technologie; Inspection; Joints (timber); Life extension; Life prediction; Measured strain; Measurement; Monitoring; Oceanographic data; Offshore; Offshore construction; Offshore engineering; Offshore structures; Probability theory; Product design; Reliability; Remaining life; SCADA; Service life; Strain measurement; Support structures; Turbine engines; Turbines; Uncertainty; Welded joints; Wind power; Wind speed; Wind turbines; Measurement; Welded joints; Measured strain; Offshore; Remaining life; Inspection; Fatigue; Wind turbines; Reliability; Support structures; Monitoring; SCADA
• ISSN: 0951-8339; 1873-4170
• DOI: 10.1016/j.marstruc.2019.05.002

Reassessing the remaining fatigue life of the wind turbine support structures becomes more and more crucial for operation, maintenance, and life extension when they are reaching the end of their design service life. By using measured oceanographic and strain data, each year, remaining fatigue life can be updated to adapt the operation to real loading conditions. Previous works have not put attention to address the complexity of offshore loading combinations and as-constructed state of the structure in estimating structural responses for fatigue behaviour to stochastically predict the remaining fatigue life. The present paper links the oceanographic data to fatigue damage by using measured strain, and uses the Bayesian approach to update the joint distribution of the oceanographic data. Consequently, the failure probability of the support structure can be updated and so the predicted fatigue life. The year-to-year variation of the 10-min mean wind speed, the unrepresentativeness of measured strain, the measurement uncertainty, and corrosion are considered together with uncertainties in Miner′s rule and S–N curves. The present research shows that the real oceanographic data can be used to adjust the predicted remaining fatigue life and eventually give decision support for the wind turbine operation. •Wind and wave are included in the stress-life (S-N) fatigue damage calculation of existing structures.•Wind speed distribution is updated using Bayesian approach and included in the limit state function.•Effect of year-to-year variation of annual mean wind speed is considered•The estimated fatigue life is very sensitive to the accuracy of the method used to derive stresses at the interested location from the measuring location.•The most probable hot-spot location is found by combining measured strain data of different wind directions.