A Machine-Learning Approach for the Development of a FOWT Model Integrated with Four OWCs

Abstract

The wind-wave excitations cause structural vibrations on the Floating Offshore Wind Turbines (FOWT) pressing the power generation efficiency and reducing the life expectancy. In particular, tower-top displacement and barge-type platform pitch dynamics are extremely sensitive to wind speed and wave elevation to the point that may even lead to structural instability in extreme conditions. Having into account that computational techniques such as Artificial Neural Networks (ANNs) are widely used in artificial intelligence because of their strong predicting and forecasting capabilities, the aim of this article is to create a deep-layer ANN model that incorporates Oscillating Water Columns (OWCs) into the barge platform. This ANN model enables to address stability issues of the hybrid floating offshore wind platform. The proposed control-oriented model has been successfully validated to achieve adequate dynamic behavior and structural performance using FAST.