|
|
Publications
Xu, S., Wang, S. and Guedes Soares, C. (2024), A hybrid deep learning approach to predict dynamic mooring tension of a wave energy converter, 43rd International Conference on Ocean, Offshore and Arctic Engineering (OMAE2024), 9-14 June, Singapore, Singapore, Paper No: OMAE2024-129671, V007T09A078.
Enhancing the precision of short-term mooring tension prediction holds the key to bolstering safety measures within marine operations. In this study, a hybrid model which takes the advantages of convolutional neural networks (CNN) and gated recurrent units (GRU) is developed to predict dynamic mooring tension time series of a wave energy converter. The datasets used to train the hybrid model were collected from model tests conducted in a wave flume. The hybrid model is trained using motion responses of a wave energy converter. The study explores and discusses the impact of various factors on the performance of the hybrid GRU model, including the number of neurons, the choice of optimizer, and the number of dully connected layers. Optimal model parameters are determined through an assessment of error indexes. Once the optimum GRU model is determined, a hybrid model which consists of the CNN and the optimum model is proposed for predicting mooring tensions. To evaluate the effectiveness of the hybrid model, a comparison is made between its performance in dynamic mooring tension prediction and that of GRU. This research offers highlights the substantial potential of the proposed method in predicting mooring tensions for other offshore floating structures.
If you did not manage to obtain a copy of this paper: Request a copy of this article
For information about all CENTEC publications you can download: Download the Complete List of CENTEC Publications
|