Motion simulation of moorings using optimized LSTM neural network

作     者：Zhiyuan ZHUANG Fangjie YU Ge CHEN Zhiyuan ZHUANG;Fangjie YU;Ge CHEN

作者机构：Department of Marine TechnologyCollege of Information Science and EngineeringOcean University of ChinaQingdao 266100China Laoshan LaboratoryQingdao 266237China 

出 版 物：《Journal of Oceanology and Limnology》 (海洋湖沼学报（英文）)

年 卷 期：2023年第41卷第5期

页      面：1678-1693页

核心收录：

学科分类：08[工学] 0816[工学-测绘科学与技术] 

基　　金：Supported by the Laoshan Laboratory (Nos.LSKJ202201302-5 LSKJ202201405-1 LSKJ202204304) 

主　　题：mooring motion simulation long short-term memory(LSTM) optimization strategy hybrid deep learning 

摘      要：Mooring arrays have been widely deployed in sustained ocean observation in high resolution to measure finer dynamic features of marine ***,the irregular posture changes and nonlinear response of moorings under the effect of ocean currents face huge challenges for the deployment of mooring arrays,which may cause the deviations of measurements and yield a vacuum of observation in the upper *** developed a data-driven mooring simulation model based on LSTM(long short-term memory)neural network,coupling the ocean current with position data from moorings to predict the motion of moorings,including single-step output prediction and multi-step *** on the predictive information,the formation of the mooring array can be adjusted to improve the accuracy and integrity of ***,we proposed the cuckoo search(CS)optimization algorithm to tune the parameters of LSTM,which improves the robustness and generalization of the *** utilize the datasets observed from moorings anchored in the Kuroshio Extension region to train and validate the simulation *** experimental results demonstrate that the model can remarkably improve prediction accuracy and yield stable ***,compared with other optimization algorithms,CS is more efficient and performs better in simulating the motion of moorings.