摘要
在风、浪、流等复杂环境荷载的联合作用下,海洋浮式平台运动响应呈现强非线性及非平稳特征。为了解决海洋平台运动响应复杂多变、难以预测的问题,本文结合图像化特征提取方法与卷积神经网络(convolutional neural network,CNN),提出一种半潜式平台六自由度响应预测方法。首先根据海洋荷载计算方程,结合实测荷载数据,提出图像化海洋环境特征提取方法;其次,基于CNN建立平台响应预测模型,并对不同特征输入、模型参数等影响进行分析;最终,利用实测数据验证了所提预测方法的误差仅为3.84%,对比直接基于原始数据的CNN模型,精度提高了64.24%。
Semi-submersible platforms are subjected to complex load environments with coupled influences of wind,waves,and currents.The platforms' motions may present strong nonlinear characteristics.In order to solve the problem that the motion response is difficult to predict due to its complexity and changeability,a six-degree-of-freedom(6 DOFs) prediction method for semi-submersible platforms based on image feature selection method and convolutional neural network(CNN) was proposed in this paper.Firstly,according to the ocean environment load calculation equation and the measured data,a novel image feature selection method of environment loads was proposed.Secondly,a 6 DOFs motion prediction model was established based on CNN method,and the effects of different feature inputs and model parameters were analyzed.Finally,it was verified that the prediction error of the proposed method was only 3.84%.Compared with the CNN model based on the original data without feather selection,the prediction accuracy of the proposed method was improved by 64.24%.
作者
姚骥
武文华
徐海博
顾学康
张欣玉
YAO Ji;WU Wen-hua;XU Hai-bo;GU Xue-kang;ZHANG Xin-yu(China Ship Scientific Research Center,Wuxi 214082,China;Taihu Laboratory of Deepsea Technological Science,Wuxi 214082,China;State Key Laboratory of Analysis of Industrial Equipment,Faculty of Vehicle Engineering and Mechanics,Dalian University of Technology,Dalian 116024,China;Ningbo Research Institute,Dalian University of Technology,Ningbo 315000,China;School of Naval Architecture,Ocean and Civil Engineering,Shanghai Jiao Tong University,Shanghai 200240,China)
出处
《船舶力学》
EI
CSCD
北大核心
2023年第5期617-626,共10页
Journal of Ship Mechanics
基金
国家重点研发计划资助项目(2017YFC0307203)
国家自然科学基金资助项目(U1906233)
山东省重点研发计划资助项目(2019JZZY010801)
基本科研业务费重点类项目(DUT20ZD213,DUT20LAB308)。
作者简介
姚骥(1994-),男,博士研究生;通讯作者:武文华(1973-),男,博士,教授,E-mail:lxyuhua@dlut.edu.cn。
