基于一维卷积GRU网络的导管架平台动力响应实时预测与分析

张振, 崔春义, 张鹏, 刘海龙, 王坤鹏, 李雪

doi:10.16411/j.cnki.issn1006-7736.2023.02.003

大连海事大学学报 >

2023 , Vol. 49 >Issue 2: 23 - 32

DOI: https://doi.org/10.16411/j.cnki.issn1006-7736.2023.02.003

船舶与海洋工程

基于一维卷积GRU网络的导管架平台动力响应实时预测与分析

展开

(大连海事大学交通运输工程学院，辽宁大连 116026)

张振（1998—），男，硕士生，E-mail：alanmonster@163.com

网络出版日期: 2023-02-21

基金资助

辽宁省航运联合基金项目（2020-HYLH-48）；大连市科技创新基金重点学科重大项目（2020JJ25CY016）

收起

Real-time prediction and analysis of jacket platform dynamic response based on one-dimensional convolution and GRU network

Expand

(College of Transportation Engineering, Dalian Maritime University, Dalian 116026, China)

Online published: 2023-02-21

Fold

摘要

为有效地进行波浪荷载作用下海洋导管架平台动力响应实时预测，提出一种综合一维卷积与门控循环复合神经网络（1DCNN+GRU）的动力响应预测方法。基于SACS程序平台建立单层甲板的四桩腿导管架数值计算模型，通过非线性数值模型分析其在波浪作用下动力响应状态，从而得到结构动力响应时程样本数据，并经标准化处理后，输入到由Pytorch框架搭建的复合神经网络中进行训练与测试。计算分析结果表明：该1DCNN+GRU复合神经网络预测方法具有良好的求解精度和稳定性，各求解步时长远小于预测提前量，能够实现海洋导管架平台动力响应的实时预测，可为相关工程实践提供参考。

关键词： 导管架平台; GRU网络; 动力响应预测; 一维卷积

本文引用格式

张振, 崔春义, 张鹏, 刘海龙, 王坤鹏, 李雪 . 基于一维卷积GRU网络的导管架平台动力响应实时预测与分析[J]. 大连海事大学学报, 2023 , 49(2) : 23 -32 . DOI: 10.16411/j.cnki.issn1006-7736.2023.02.003

Abstract

In order to effectively achieve a real-time prediction of jacket platforms dynamic response under wave load,the dynamic response prediction method combining one-dimensional convolutional neural network and gated recurrent composite neural network(1DCNN+GRU) was proposed. Based on the SACS program platform,the numerical calculation model of single deck four-legged jacket was established.The dynamic dynamic response state of the jecket under wave action was analyzed by using a nonlinear numerical model to obtain the structural dynamic response time course sample data,and then input into a composite neural network built by the Pytorch framework for training and testing after standardization processing.The calculation and analysis results show that the 1DCNN+GRU composite neural network prediction method has good solution accurate and stability, and each solution step length was less than the amount of forecast advance,which can achieve the dynamic dynamic response real-time prediction of offshore jacket platforms, and provie reference for relevant engineering practices.

Key words： jacket platform; GRU network; dynamic response prediction; one dimensional convolution

参考文献

[1]吴志浩,崔春义,张新程,等. 直立腿海洋平台冰激振动响应参数敏感性分析[J]. 大连海事大学学报,2021,47(4):93-99.

WU Z H, CUI C Y, ZHANG X C, et al. Sensitivity analysis of ice-induced vibration response parameters of the vertical leg offshore platform[J]. Journal of Dalian Maritime University,2021,47(4):93-99. (in Chinese)

[2]黄礼敏. 海浪中非平稳非线性舰船运动在线预报研究[D]. 哈尔滨:哈尔滨工程大学, 2016.

HUANG L M. On-line prediction of non-stationary and nonlinear ship motions at sea[D]. Harbin:Harbin Engineering University, 2016. (in Chinese)

[3]SØRENSEN P H, NØRGAARD M, RAVN O, et al. Implementation of neural network based non-linear predictive control[J]. Neurocomputing (Amsterdam), 1999,28(1): 37-51.

[4]THIRUMALAISELVI A, MOHIT V, ANANDAVALLI N, et al. Response prediction of laced steel-concrete composite beams using machine learning algorithms[J].Structural Engineering and Mechanics, 2018, 66(3):399-409.

[5]SAKARIDIS E, KARATHANASOPOULOS N, MOHRD. Machine-learning based prediction of crash response of tubular structures[J]. International Journal of Impact Engineering, 2022, 166: 104240.

[6]SAHOO D M, CHAKRAVERTY S. Functional link neural network learning for response prediction of tall shear buildings with respect to earthquake data [ J]. IEEE Transactions on Systems, Man, and Cybernetics: Systems,2018, 48(1): 1-10.

[7]OH B K, GLISIC B, PARK S W, et al. Neural networkbased seismic response prediction model for building structures using artificial earthquakes [ J]. Journal of Sound and Vibration, 2020, 468: 115109.

[8]AHMED B, MANGALATHU S, JEON J. Seismic damage state predictions of reinforced concrete structures using stacked long short-term memory neural networks[ J ]. Journal of Building Engineering, 2022, 46:103737.

[9]ZHAO Y, DONG S, JIANG F, et al. Mooring tension prediction based on BP neural network for semi-submersible platform [ J ]. Ocean Engineering, 2021, 223:108714.

[10]WANG Z, QIAO D, YAN J, et al. A new approach to predict dynamic mooring tension using LSTM neural network based on responses of floating structure[J]. Ocean Engineering, 2022, 249: 110905.

[11]YAO J, WU W, LI S. Anomaly detection model of mooring system based on LSTM PCA method [J].Ocean Engineering, 2022, 254: 111350.

[12]GUO X, ZHANG X, TIAN X, et al. Probabilistic prediction of the heave motions of a semi-submersible by a deep learning model[J]. Ocean Engineering, 2022, 247: 110578.

[13]李昊波,肖龙飞,魏汉迪,等. 基于LSTM 网络的浮式海洋平台运动在线预报研究[J]. 船舶力学, 2021,25(5): 576-585.

LI H B, XIAO L F, WEI H D, et al. Research on online motion prediction of floating offshore platform based on LSTM network [ J]. Journal of Ship Mechanics, 2021, 25(5): 576-585. (in Chinese)

[14]张德庆,王超,杜君峰. 基于人工神经网络算法的深海浮式系统动力响应预报方法[J]. 中国造船,2021, 62(1): 123-132.

ZHANG D Q, WANG C, DU J F. Prediction method of dynamic response of deep-sea floating system based on artificial neural network algorithm[J]. Shipbuilding of China, 2021, 62(1): 123-132. (in Chinese)

[15]CHO K, MERRIENBOER B V, GULCEHRE C, et al.Learning phrase representations using RNN encoder-decoder for statistical machine translation[C] / / Proceedings of the 2014 Conference on Empirical Methods inNatural Language Processing (EMNLP). [S. l. :s. n. ],2014: 1724-1734.

Options

文章导航

模态框（Modal）标题

摘要

本文引用格式

Abstract

参考文献