ABSTRACT

We discuss the use of the environmental contour method to derive design loads for an active stall-regulated offshore wind turbine. Two different Danish offshore environments, Rødsand and Horns Rev, are considered for the locations of the turbine. The accuracy of the derived design loads is assessed by comparing them with exact solutions derived using full integration over an accurate description of the failure domain. The error in estimating design loads is introduced because two key assumptions of the method are violated:

  • the limit state surface especially in the operating range of the turbine is not well approximated by a tangent hyperplane at the design point; and

  • failure in any of the possible turbine states (e.g., operating or parked states discussed here) needs to be considered in computing accurate failure probabilities.

It is recommended that environmental contours and iso-response curves should be plotted and interpreted before establishing design load levels.

INTRODUCTION

Inverse reliability techniques are commonly used when there is interest in establishing design levels associated with a specified reliability of probability of failure. The heuristically appealing Inverse First-Order Reliability Method (Inverse FORM) also called the "environmental contour" method is an example of an inverse reliability technique that has been applied to estimate design loads in many applications including offshore platforms (Winterstein et al., 1993) and onshore wind turbines (Fitzwater et al., 2003; Saranyasoontorn and Manuel, 2003). Even though application of the environmental contour approach to deriving design loads for wind turbine structures has been well established (see, for example, Saranyasoontorn and Manuel, 2003), it has not been clearly explained how the treatment of multiple turbine states (or accounting for the possibility of failure under different conditions such as "operating" or "parked") typical for wind turbine applications can affect the accuracy of the derived design loads.

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