Lessons learnt from back-analysis of offshore monopile lateral response-Paper

Paper

Pdf Summary

The study presents a monitoring-based workflow for improving predictions of the lateral response and natural frequencies of offshore wind turbine monopiles. Because fatigue is often the governing design limit state, accurately estimating natural frequency—and avoiding excessive conservatism—is essential.<br /><br />The researchers back-analyzed measurements from 73 turbines at two Belgian offshore wind farms. They compared measured first natural frequencies with integrated structural models using different pile–soil interaction methods. The commonly used API RP2 GEO method generally underestimated natural frequency because it is based largely on soil strength and does not adequately represent the higher, strain-dependent stiffness mobilized under typical operating loads. The PISA method, which incorporates small-strain soil stiffness, gave the closest match, with a mean relative error of −1.48%.<br /><br />The analysis also showed that scour protection materially affects monopile stiffness. It increases effective stress and stiffness in shallow soil layers and provides additional lateral restraint. Modelling all three effects produced the best results, reducing the mean frequency error to −0.51%, with a standard deviation of 1.30%. The restraint provided by the blocky protection layer remains difficult to quantify and warrants further research.<br /><br />Overall, reliable predictions require realistic modelling of soil stiffness at relevant strain levels, the scour protection system, structural properties, and operating conditions. Site-specific stiffness estimates are important, and uncertainty in CPT-based correlations should be assessed. The findings also have potential relevance to other structures, including onshore piles and retaining walls, where soil stiffness influences lateral response.

Keywords

offshore wind turbine

monopile

natural frequency

lateral response

soil stiffness

pile-soil interaction

PISA method

API RP2 GEO

scour protection

fatigue design