Identifying Structural Parameters of an Idling Offshore Wind Turbine Using Operational Modal Analysis

Abstract

The design of modern day offshore wind turbines (OWTs) relies on numerical models, which are used for simulating the dynamic behavior in different operational and environmental conditions. From these results one can estimate ultimate and fatigue loads, which are needed for determining the design life of the turbines. The dynamic behavior, and thus the lifetime, of the turbines are influenced for a large part by its structural properties, such as the natural frequencies and damping ratios. Hence, it is important to obtain accurate estimates of these modal properties. For this purpose Operational Modal Analysis (OMA) techniques are used to estimate the modal domain of the OWT. As, for instance, the loads in the structure and the damping ratios are inversely related, higher damping values will results in lower loads, and hence in more optimized and less costly support structures. In this paper the data-driven Stochastic Subspace Identification (SSI) method is used to evaluate the modal domain of the OWT by using output-only measurements obtained from an installed 3.6 MW offshore wind turbine. In order to better satisfy the OMA assumptions of having a Linear Time-Invariant (LTI) system and white noise uncorrelated input, the analyses are performed in case of idling turbines, thereby avoiding the effect of rotational harmonic components, changing system properties due to yawing and pitching actions, as well as strong aerodynamic nonlinearities. In this paper we focuss on the first four global eigenfrequencies that were found and the associated damping ratios. Even though a sensor mix of several strain gauges and a single bi-directional accelerometer are available, the best results were obtained by only using the accelerometer on the nacelle.