Software-In-the-Loop Method to Predict the Global Dynamic Responses of Full-scale Floating Wind Turbines by Artificial Neural Network

The design of floating offshore wind turbines (FOWTs) need accurate predictions of full-scale global dynamic responses. Conventional basin experimental method can hardly be used directly to predict the full-scale global dynamic responses of FOWTs, due to the dissimilarity of aerodynamic load and hydrodynamic load. Besides, numerical simulation methods are not reliable enough at this moment due to the lack of full-scale data validation. Therefore, it is necessary to find an accurate, economic and efficient approach for FOWT design engineering practice. A new method, named as SILANN, is proposed in this study, which utilizing Software-in-the-Loop method with artificial neural network (ANN) approach for engineering prediction. Firstly, the methodology of combing ANN with in-house programme DARwind is introduced. Then, datasets and validation cases were selected from the results of basin experiment and DARwind in terms of 3DOFs (surge, heave and pitch motions). The predictions’ results show a good performance. The difference between simulation results and those of experiment can be significantly reduced by this method. This proposed method takes advantage of the AI technology, which brings new solution for overcoming the handicap impeding direct using of traditional basin experimental technology in floating wind turbine design.

Subjects

Artificial neural network

Basin model test

DARwind

Floating offshore wind turbines

Numerical simulation

DDC Class

600: Technik

Lizenz

http://rightsstatements.org/vocab/InC/1.0/

Name

paper_0040_file_0367_Software_In_the_Loop_Method_to_Predict_the_Global_Dynamic_Respon.pdf

Size

924.11 KB

Format

Adobe PDF

Options

Software-In-the-Loop Method to Predict the Global Dynamic Responses of Full-scale Floating Wind Turbines by Artificial Neural Network