Abstract

Fatigue assessment of bolted ring-flanges in offshore wind turbine structures is a critical step in the overall structural design process. Analytical and numerical methods are employed to predict bolt stress, offering the convenience of speed (analytical) or accuracy (numerical). In this paper, we investigate the application of a Gaussian Process (GP) surrogate model, leveraging the accuracy of numerical techniques with significant gains in computation times. We present the notion of predictive uncertainty available from the GP surrogate model. Finally, we demonstrate the capacity of the GP surrogate model to propagate input space uncertainty in a computationally efficient manner.

Introduction

There is a global need for reliable non-fossil fuel alternatives that is driving development of offshore wind installed capacity. This capacity is projected to grow from 29.1 GW by end-2019 to 98.9 GW by end-2025 (Lee and Zhao 2020). The life cycle costs of managing an offshore wind farm are significantly affected by the costs of offshore inspection and maintenance (Zhong et al. 2019). Appropriate timing of offshore inspection activities for offshore wind turbines (OWT) is crucial to manage costs and limit the safety exposure of maintenance personnel.

Our interest is in fatigue life prediction for the bolted ring-flanges within an OWT structure, which can be subjected to a highly dynamic load regime of > 109 load cycles (Schaumann, Eichstädt, and Stang 2018). Analytical and numerical methods are currently used in industry to predict bolt load as a function of tower load. However, both of these methods suffer from drawbacks. The analytical methods have been shown to deviate from detailed numerical model predictions due to uncertainties inherent in a) operating conditions and hence tower load estimation, b) in installation practices, most notably bolt pre-load, and c) mechanical behaviour such as flange gapping. Numerical solutions, such as structural Finite Element Method (FEM) simulations, permit additional model complexity and hence are better positioned to approximate the physical process of interest. However, residual uncertainty persists (Girolami et al. 2021), and FEM simulations can be subject to significant computational costs (Schaumann and Eichstädt 2015).

This content is only available via PDF.
You can access this article if you purchase or spend a download.