|
|
Publications
Xu, S., Wang, S. and Guedes Soares, C. (2024), Prediction of mooring tensions of a wave energy converter considering the effects of nonlinear axial stiffness by a deep learning neural network, Ocean Engineering, Vol. 305, 117810
A validated numerical tool is applied to conduct a fully coupled dynamic analysis for a wave energy converter moored by three taut nylon ropes with nonlinear axial stiffness considered. After that, a bi-directional long short-term memory neural network (Bi-LSTM) is developed to predict mooring tension time series, and the simulated data are used to train and check the accuracy of the Bi-LSTM. The influence of the optimizer, the length of memory and the training duration on the performance of the Bi-LSTM is discussed, and the best parameters are determined by evaluating error indexes, including the average absolute error, the root mean square error, the standard deviation error, the mean value error, the maximum value error and the R-squared value. After the architecture of the Bi-LSTM is determined, the model is used to predict mooring tension time series under different application scenarios. The mooring tensions are predicted precisely by the Bi-LSTM, as the relative errors are less than 7% and the R-squared error is greater than 0.96 in all study cases. It is seen that the proposed model is an efficient and accurate tool in designing mooring systems which use synthetic fibre ropes.
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
|