Study of A Hybrid Deep Learning Method for Forecasting the Short-Term Motion Responses of A Semi-Submersible

作     者：XU Sheng JI Chun-yan XU Sheng;JI Chun-yan

作者机构：School of Naval Architecture and Ocean EngineeringJiangsu University of Science and TechnologyZhenjiang 212100China 

出 版 物：《China Ocean Engineering》 (中国海洋工程(英文版))

年 卷 期：2024年第38卷第6期

页      面：917-931页

核心收录：

学科分类：08[工学] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：The first author gratefully acknowledges the National Natural Science Foundation of China(Grant No.52301322) the Jiangsu Provincial Natural Science Foundation(Grant No.BK20220653) The second author gratefully acknowledges the National Science Fund for Distinguished Young Scholars(Grant No.52025112) the Key Projects of the National Natural Science Foundation of China(Grant No.52331011) 

主　　题：short-term motion responses convolutional neural network bidirectional long short-term memory neural network attention mechanism hybrid model multi-step prediction semi-submersible 

摘      要：Accurately predicting motion responses is a crucial component of the design process for floating offshore *** study introduces a hybrid model that integrates a convolutional neural network(CNN),a bidirectional long short-term memory(BiLSTM)neural network,and an attention mechanism for forecasting the short-term motion responses of a ***,the motions are processed through the CNN for feature *** extracted features are subsequently utilized by the BiLSTM network to forecast future *** enhance the predictive capability of the neural networks,an attention mechanism is *** addition to the hybrid model,the BiLSTM is independently employed to forecast the motion responses of the semi-submersible,serving as benchmark results for ***,both the 1D and 2D convolutions are conducted to check the influence of the convolutional dimensionality on the predicted *** results demonstrate that the hybrid 1D CNN-BiLSTM network with an attention mechanism outperforms all other models in accurately predicting motion responses.