Session: 06-17-03 AI Technology for Ocean Engineering III

Paper Number: 124429

124429 - A Response Frequency Informed Lstm Model for Ultra-Short-Term Mooring Line Forces Prediction of Floating Wind Turbines 

Accurate dynamic response forecasting is pivotal for operational monitoring, maintenance, and dynamic control of floating wind turbines (FWT), ensuring both efficiency and safety in real-world applications. This paper conducts the measured validation and application of the ultra-short-term forecast of a full-scale FWT’s mooring line tension by recurrent neural networks with the frequency decomposition method (RNN-FD). The RNN-FD method is one of the modules of our in-house program DARwind-AI to realize real-time or ultra-short-term forecasting of dynamic response of FWT. Initially, the principles of the RNN-FD method are introduced, including the Long Short-Term Memory (LSTM) encoder-decoder network, wavelet transform applicaiton and frequency decomposition method was demonstrated. Then, the validation of RNN-FD method is carried out using measured data from Hywind Scotland and the results of RNN-FD are promising, with a good agreement observed datasets from 2018 January to July. For instance, the Mean Squared Error (MSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), and One-Nth Largest Error (ONLE) for mooring line forces forecasting at lead times of 60 seconds are 0.0065, 0.063, 0.034 and 0.073, respectively. Conclusively, this innovative methodology, rooted in both traditional engineering principles and artificial intelligence technology, offers promising prospects for real-time FWT monitoring and dynamic control.

Presenting Author: Peng Chen SJTU

Presenting Author Biography: Dr. Chen Peng is a postdoctoral researcher from Shanghai Jiao Tong University. His research focuses on the floating wind turbines and the application of artificial intelligence in terms of numerical simulations in both time and frequency domains, design optimisation, and basin experiment.

Authors:

Peng Chen SJTU
Yirou Kang SJTU
Ruihan Zhang Tsinghua University
Zhengshun Cheng SJTU
Shi Deng Offshore Oil Engineering Co. Ltd.
Gareth Erfort Stellenbosch University
Zhiqiang Hu Newcastle University