TSM-TLHS Prediction Method for Assembly Deformation of Large Curved Thin Plates in Shipbuilding

doi:10.16183/j.cnki.jsjtu.2023.576

Abstract

Abstract:

During the block assembly, large curved thin plates (such as outer plates) undergo deformation due to the force of gravity when they are placed on the jigs, which affects the accuracy and quality of the block assembly in shipbuilding. In order to predict the deformation of these large curved thin plates within a given jig layout, this paper introduces a Transformer-based surrogate model with two-stage Latin hypercube sampling (TSM-TLHS). Primarily, compared to traditional approaches, the two-stage Latin hypercube sampling (TLHS) method enables direct sampling of irregularly shaped thin plates. Simultaneously, this paper uses a Transformer-based surrogate model (TSM) incorporating multi-head attention modules and positional encoding to comprehensively consider the impact of jig positions and corresponding node displacements on thin plate deformation. Real case results demonstrate that the prediction error of this TSM-TLHS method is only 61 μm, meeting the on-site assembly precision requirements for predicting plate deformation. This facilitates timely anti-deformation compensation by block in shipyards, ensuring assembly quality.

Key words: block assembly, curved thin plates, deformation prediction, surrogate model, Latin hypercube sampling

CLC Number:

TH183

JIN Xuancheng, HONG Ge, GAO Shuo, XIA Tangbin, HU Xiaofeng, XI Lifeng. TSM-TLHS Prediction Method for Assembly Deformation of Large Curved Thin Plates in Shipbuilding[J]. Journal of Shanghai Jiao Tong University, 2025, 59(8): 1092-1102.

Figures/Tables 16

Fig.1

Fig.2

Fig.3

Fig.4

Fig.5

Fig.6

Fig.7

Fig.8

Fig.9

Tab.1

Tab.2

Tab.3

Fig.10

Fig.11

Fig.12

Tab.4

References 18

[1]	应长春. 船舶工艺技术[M]. 上海: 上海交通大学出版社, 2013: 359.
	YING Changchun. Shipbuilding technology[M]. Shanghai: Shanghai Jiao Tong University Press, 2013: 359.
[2]	CAI W, HU S, YUAN J. Deformable sheet metal fixturing: Principles, algorithms, and simulations[J]. Journal of Manufacturing Science and Engineering, 1996, 118(3): 318-324.
[3]	王威, 王珉, 胡俊聪, 等. 面向汽车覆盖件的有限元仿真虚拟匹配方法[J]. 上海交通大学学报, 2020, 54(5): 532-543. doi: 10.16183/j.cnki.jsjtu.2020.05.011
	WANG Wei, WANG Min, HU Juncong, et al. Virtual matching method based on finite element simulation in automotive panel[J]. Journal of Shanghai Jiao Tong University, 2020, 54(5): 532-543.
[4]	VINOSH M, RAJ T, PRASATH M. Optimization of sheet metal resistance spot welding process fixture design[J]. Materials Today: Proceedings, 2021, 45: 1696-1700.
[5]	JU K, DUAN C, KONG J, et al. Prediction of clamping deformation in vacuum fixture-workpiece system for low-rigidity thin-walled precision parts using finite element method[J]. The International Journal of Advanced Manufacturing Technology, 2020, 109(7): 1895-1916.
[6]	SLON C, PANDEY V. An optimization framework for fixture layout design for nonrigid parts[J]. SAE International Journal of Materials and Manufacturing, 2020, 13(1): 5-18.
[7]	LOOSE J, CHEN N, ZHOU S. Surrogate modeling of dimensional variation propagation in multistage assembly processes[J]. IIE Transactions, 2009, 41(10): 893-904.
[8]	杨国林, 孙学先, 锁旭宏, 等. 桥梁形变监测中LSTM预测方法研究[J]. 兰州交通大学学报, 2022, 41(5): 1-5.
	YANG Guolin, SUN Xuexian, SUO Xuhong, et al. The application of LSTM method to bridge deformation prediction[J]. Journal of Lanzhou Jiaotong University, 2022, 41(5): 1-5.
[9]	ZHANG Y, MA N, GU X, et al. Surrogate model based on an MLP neural network and Bayesian hyperparameter tuning for ship hull form optimization[J]. International Journal of Offshore and Polar Engineering, 2023, 33(2): 184-195.
[10]	胡新明, 王德禹. 基于迭代均值组合近似模型和序贯优化与可靠性评估法的船舶结构优化设计[J]. 上海交通大学学报, 2017, 51(2): 150-156.
	HU Xinming, WANG Deyu. Optimization of ship structures using ensemble of surrogates with recursive arithmetic average and sequential optimization and reliability assessment[J]. Journal of Shanghai Jiao Tong University, 2017, 51(2): 150-156.
[11]	QIU W, LIU K, ZONG S, et al. An optimization method for anti-blast performance of corrugated sandwich plate structure based on neural network and sparrow search algorithm[J]. Ships and Offshore Structures, 2024, 19(8): 1028-1043.
[12]	何维, 孙宏磊, 陶袁钦, 等. 开挖引起的隧道位移动态多目标优化反演预测[J]. 上海交通大学学报, 2022, 56(12): 1688-1699. doi: 10.16183/j.cnki.jsjtu.2021.282
	HE Wei, SUN Honglei, TAO Yuanqin, et al. Dynamic multi-objective optimization inverse prediction of excavation-induced tunnel displacement[J]. Journal of Shanghai Jiao Tong University, 2022, 56(12): 1688-1699.
[13]	DU J, LIU C, LIU J, et al. Optimal design of fixture layout for compliant part with application in ship curved panel assembly[J]. Journal of Manufacturing Science and Engineering, 2021, 143(6): 061007.
[14]	TANG B. Orthogonal array-based Latin hypercubes[J]. Journal of the American Statistical Association, 1993, 88(424): 1392-1397.
[15]	JOSEPH V, HUNG Y. Orthogonal-maximin Latin hypercube designs[J]. Statistica Sinica, 2008, 18: 171-186.
[16]	夏立, 邹早建, 袁帅, 等. 基于非侵入式混沌多项式法的随机阻曳流CFD模拟不确定度量化[J]. 上海交通大学学报, 2020, 54(6): 584-591.
	XIA Li, ZOU Zaojian, YUAN Shuai, et al. Uncertainty quantification for CFD simulation of stochastic drag flow based on non-intrusive polynomial chaos method[J]. Journal of Shanghai Jiao Tong University, 2020, 54(6): 584-591.
[17]	VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]// Advances in Neural Information Processing Systems 30. La Jolla, USA: NIPS, 2017: 6000-6010.
[18]	YUE X, SHI J. Surrogate model-based optimal feed-forward control for dimensional-variation reduction in composite parts’ assembly processes[J]. Journal of Quality Technology, 2018, 50(3): 279-289.

参数名称	数值
编码器	4
heads	16
channels	80
学习率	0.001(40 epoch之前);0.000 01(40 epoch之后)
epoch	100
训练批量	20

运算名称	T/s
模型训练	3 600
模型预测	0.03
FEA方法计算	480

层数	类型	输出通道数
1-1	全连接层	180
1-2	ReLU	180
2-1	全连接层	160
2-2	ReLU	160
3	全连接层	1

方法	平均形变/mm	M/mm	平均输出误差/%
TSM-TLHS	10.61	0.08	0.7
实际案例结果	10.69	0.08	0.7