System Identification and model comparison of a Tension Leg Platform for Floating Offshore Wind Turbines

Abstract

This paper will focus on using system identification on experimental data for building a mathematical model for the platform of a floating offshore wind turbine and analyzing the behavior of the structure. The floating offshore wind turbine examined in this paper uses a scaled tension leg platform as its foundation and the wind turbine is a 1:35 scaled model of the 5 MW NREL offshore wind turbine. The mathematical model of the platform will describe the displacement of the TLP in surge when affected by an irregular wave series generated from a scaled Pierson-Moskowitz wave spectrum. To obtain such a mathematical model, an examination of the displacement of the platform due to the hydrodynamic loads will be conducted on the foundation of the floating offshore wind turbine. The height of the waves and the displacement of the floating offshore wind turbines will be measured by resistive wave gauges and OptiTrack cameras, respectively, at the offshore laboratory at Aalborg University Esbjerg. System identification is used on the data obtained from the experiments, to build multiple mathematical models with different model structures, in order to find the most appropriate model structure. It is concluded from the analysis of the different mathematical models, that the Autoregressive Moving Average and Extra input model structure is the most accurate model at describing the dynamics of the platform of a floating offshore wind turbine. The model is valid for a specific operating range of Pierson-Moskowitz waves generated with a wind speed corresponding to 2 meters per seconds.

General Methodology for the Identification of Reduced Dynamic Models of Barge-Type Floating Wind Turbines

“…In [15], the authors present a model-based blade pitch controller for a spar-type FOWT scale prototype. They obtain the model from experimental input-output measurements using systems identification methods.…”

“…The identification methodology presented here has been successfully used in the investigation of control devices that have been applied to reduce structural loads in a barge FOWT turbine [12][13][14][15].…”