Physical system simulation based on deep representation learning for 3D geometric features

doi:10.11772/j.issn.1001-9081.2023101505

Journal of Computer Applications ›› 2024, Vol. 44 ›› Issue (11): 3548-3555.DOI: 10.11772/j.issn.1001-9081.2023101505

• Multimedia computing and computer simulation • Previous Articles Next Articles

Physical system simulation based on deep representation learning for 3D geometric features

Fu LIN¹, Jiasheng SHI¹, Ze GAO², Zunkang CHU², Qiongmin MA³, Haiyan YU², Weixiong RAO¹()

^1.School of Software Engineering，Tongji University，Shanghai 201804，China
^2.School of Automotive Studies，Tongji University，Shanghai 201804，China
^3.National Key Laboratory for Complex Systems Simulation，Beijing 100101，China

Received:2023-11-03 Revised:2023-12-08 Accepted:2023-12-15 Online:2024-11-13 Published:2024-11-10
Contact: Weixiong RAO
About author:LIN Fu， born in 2000， M. S. candidate. His research interests include graph neural network， AI for FEM.
SHI Jiasheng， born in 2000， Ph. D. candidate. His research interests include graph neural network， AI for FEM.
GAO Ze， born in 1998， Ph. D. candidate. His research interests include application of graph neural network in automotive structural simulation.
CHU Zunkang， born in 1999， Ph. D. candidate. His research interests include application of machine learning in automotive structural design.
MA Qiongmin， born in 1988， Ph. D.， research assistant. Her research interests include system design and evaluation， computer simulation.
YU Haiyan， born in 1976， Ph. D.， professor. Her research interests include automotive structural optimization， automotive manufacturing processes.
Supported by:
National Key Research and Development Program of China(2022YFE0208000)

基于3D几何特征深度表达学习的物理系统仿真

林滏¹, 石稼晟¹, 高泽², 楚遵康², 马琼敏³, 余海燕², 饶卫雄¹()

^1.同济大学软件学院，上海 201804
^2.同济大学汽车学院，上海 201804
^3.复杂系统仿真总体重点实验室，北京 100101

通讯作者: 饶卫雄
作者简介:林滏（2000—），男，福建宁德人，硕士研究生，CCF会员，主要研究方向：图神经网络、面向有限元的人工智能
石稼晟（2000—），男，四川成都人，博士研究生，CCF会员，主要研究方向：图神经网络、面向有限元的人工智能
高泽（1998—），男，江苏常州人，博士研究生，主要研究方向：图神经网络在汽车结构仿真中的应用
楚遵康（1999—），男，山东济宁人，博士研究生，主要研究方向：机器学习在汽车结构设计中的应用
马琼敏（1988—），女，广东普宁人，助理研究员，博士，主要研究方向：体系设计与评估、计算机仿真
余海燕（1976—），女，安徽安庆人，教授，博士，主要研究方向：汽车结构优化、汽车制造工艺
基金资助:
国家重点研发计划项目(2022YFE0208000)

Abstract

Abstract:

To address the limitations of the existing deep learning methods in handling scenarios where both geometric boundaries and initial conditions vary in physical simulation problems， a technical approach was proposed to decouple the representation of geometric boundary constraints from the physical system simulation， and a two-step technical route of geometric feature representation learning and physical system simulation was designed. After constructing an independent geometric feature extraction module which was unaffected by external physical conditions， the extracted geometric features were fused with physical features， and finally a neural network-based physical system simulation model was designed. In stress field prediction experiments， the proposed method achieves a prediction time of only 2.63 ms， which is much lower than 0.6 s of Finite Element Method （FEM）， and has a Mean Absolute Error （MAE） only 0.389 times of that of MeshNet. Experimental results demonstrate that the proposed method maintains high simulation accuracy while effectively adapting to different geometric boundaries and initial conditions.

Key words: Finite Element Method (FEM), physical system simulation, 3D mesh, feature representation, Mean Absolute Error (MAE)

摘要：

针对现有深度学习方法在物理系统仿真中无法处理几何边界与初始条件同时变化的场景的问题，提出将几何边界约束的表达与物理系统仿真解耦的技术思路，设计了几何特征表达学习和物理系统仿真双步骤的技术路线。在构建与外部物理条件无关的几何特征提取模块之后，融合提取的几何特征与物理特征，最后设计基于神经网络的物理系统仿真方法。在应力场预测实验中，所提方法的预测时间为2.63 ms，远低于有限元法（FEM）的0.6 s，且平均绝对误差（MAE）仅为MeshNet的0.389倍。实验结果表明，所提方法能够保持较高仿真精度，同时能够较好地适应不同的几何边界与初始条件。

关键词: 有限元法, 物理系统仿真, 3D网格, 特征表达, 平均绝对误差

CLC Number:

TP183

Fu LIN, Jiasheng SHI, Ze GAO, Zunkang CHU, Qiongmin MA, Haiyan YU, Weixiong RAO. Physical system simulation based on deep representation learning for 3D geometric features[J]. Journal of Computer Applications, 2024, 44(11): 3548-3555.

林滏, 石稼晟, 高泽, 楚遵康, 马琼敏, 余海燕, 饶卫雄. 基于3D几何特征深度表达学习的物理系统仿真[J]. 《计算机应用》唯一官方网站, 2024, 44(11): 3548-3555.

Figures/Tables 24

References 22

1	周晔欣，戴如玥，黄争鸣.复合材料结构力学分析CAE软件现状［J］.应用力学学报，2020，37（1）：114-122.
	ZHOU Y X， DAI R Y， HUANG Z M. Current status of CAE software for composite structural analysis［J］. Chinese Journal of Applied Mechanics， 2020， 37（1）： 114-122.
2	唐成顺，孙丹，唐威，等.基于LSTM循环神经网络的汽轮机转子表面应力预测模型［J］.中国电机工程学报，2021，41（2）：451-460.
	TANG C S， SUN D， TANG W， et al. A turbine rotor surface stress prediction model based on LSTM recurrent neural network［J］. Proceedings of the CSEE， 2021， 41（2）： 451-460.
3	赵翔，茹东恒，王鹏，等.基于NARX神经网络方法的汽轮机转子关键部位应力预测［J］.应用数学和力学，2021，42（8）：771-784.
	ZHAO X， RU D H， WANG P， et al. On the stress prediction of key components in steam turbine rotors based on the NARX neural network［J］. Applied Mathematics and Mechanics， 2021， 42（8）： 771-784.
4	RAISSI M， PERDIKARIS P， KARNIADAKIS G E. Physics-informed neural networks： a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations［J］. Journal of Computational Physics， 2019， 378： 686-707.
5	ZHANG L， CHENG L， LI H， et al. Hierarchical deep-learning neural networks： finite elements and beyond［J］. Computational Mechanics， 2021， 67（1）： 207-230.
6	刘淼，孙成志，王潇嵩.基于深度学习的汽车结构件应力预测方法［C］// 2021中国汽车工程学会年会论文集.北京：机械工业出版社，2021：1534-1537.
	LIU M， SUN C Z， WANG X S. Stress prediction method of automobile structural components based on deep learning［C］// Proceedings of the 2021 China-SAE Congress. Beijing： China Machine Press， 2021： 1534-1537.
7	LIANG L， LIU M， MARTIN C， et al. A deep learning approach to estimate stress distribution： a fast and accurate surrogate of finite-element analysis［J］. Journal of the Royal Society Interface， 2018， 15（138）： No.20170844.
8	PFAFF T， FORTUNATO M， SANCHEZ-GONZALEZ A， et al. Learning mesh-based simulation with graph networks［EB/OL］. ［2023-09-22］..
9	SHI J， LIN F， RAO W. Learning to simulate complex physical systems： a case study［C］// Proceedings of the 32nd ACM International Conference on Information and Knowledge Management. New York： ACM， 2023： 4284-4288.
10	FENG Y， FENG Y， YOU H， et al. MeshNet： mesh neural network for 3D shape representation［C］// Proceedings of the 33rd AAAI Conference on Artificial Intelligence. Palo Alto， CA： AAAI Press， 2019： 8279-8286.
11	LIEN S L， KAJIYA J T. A symbolic method for calculating the integral properties of arbitrary nonconvex polyhedra［J］. IEEE Computer Graphics and Applications， 1984， 4（10）： 35-42.
12	ZHANG C， CHEN T. Efficient feature extraction for 2D/3D objects in mesh representation［C］// Proceedings of the 2001 International Conference on Image Processing — Volume 3. Piscataway： IEEE， 2001： 935-938.
13	HUBELI A， GROSS M. Multiresolution feature extraction for unstructured meshes［C］// Proceedings of the 2011 IEEE Conference on Visualization. Piscataway： IEEE， 2011： 287-294.
14	KAZHDAN M， FUNKHOUSER T， RUSINKIE-WICZ S. Rotation invariant spherical harmonic representation of 3D shape descriptors［C］// Proceedings of the 2003 Symposium on Geometry Processing. Goslar： Eurographics Association， 2003： 156-164.
15	QI C R， SU H， MO K， et al. PointNet： deep learning on point sets for 3D classification and segmentation［C］// Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2017： 77-85.
16	QI C R， YI L， SU H， et al. PointNet++： deep hierarchical feature learning on point sets in a metric space［C］// Proceedings of the 31st International Conference on Neural Information Processing Systems. Red Hook， NY： Curran Associates Inc.， 2017： 5105-5114.
17	SHEN Y， FENG C， YANG Y， et al. Mining point cloud local structures by kernel correlation and graph pooling［C］// Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2018： 4548-4557.
18	林瑶瑶，郁梅，何周燕，等.三维网格客观质量评价［J］.激光与光电子学进展，2021，58（2）：No.0200003.
	LIN Y Y， YU M， HE Z Y， et al. Objective quality assessment for three-dimensional meshes［J］. Laser and Optoelectronics Progress， 2021， 58（2）： No.0200003.
19	HUGHES J F， VAN DAM A， McGUIRE M， et al. Computer Graphics Principles and Practice［M］. 3rd ed. Boston： Addison-Wesley Professional， 2013： 635-668.
20	曾攀.有限元分析及应用［M］.北京：清华大学出版社，2004： 362-369.
	ZENG P. Finite Element Analysis and Applications［M］. Beijing： Tsinghua University Press， 2004： 362-369.
21	FANG Y， XIE J， DAI G， et al. 3D deep shape descriptor［C］// Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2015： 2319-2328.
22	LEYS C， LEY C， KLEIN O， et al. Detecting outliers： do not use standard deviation around the mean， use absolute deviation around the median［J］. Journal of Experimental Social Psychology， 2013， 49（4）： 764-766.

结构	种类数	结构	种类数	结构	种类数
长方体	5	三棱柱	3	T形柱	3
工形柱	4	梯形四棱柱	2	带孔长方体	2

结构	种类数	结构	种类数	结构	种类数
长方体	5	三棱柱	3	T形柱	3
工形柱	4	梯形四棱柱	2	带孔长方体	2

参数	值	参数	值
密度/（kg·m^-3）	7 850	泊松比	0.29
杨氏模量/GPa	205	屈服强度/GPa	355

参数	值	参数	值
密度/（kg·m^-3）	7 850	泊松比	0.29
杨氏模量/GPa	205	屈服强度/GPa	355

类型	型号	其他参数
CPU	Intel Xeon Platinum 8225C	内存： 47 GB；内核： 12核
GPU	NVIDIA GeForce RTX 3090	显存： 24 GB

Physical system simulation based on deep representation learning for 3D geometric features

基于3D几何特征深度表达学习的物理系统仿真

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 24

References 22

Related Articles 15

Recommended Articles

Metrics

模块	MLP参数
空间描述符	［3， 64， 64］
载荷描述符	［3， 64， 64］
面形状描述符	［6， 32， 32］，［32， 64， 64］
结构描述符	［3+64+64， 64， 64］
特征融合模块	［in×2， in］，［in， out］
多级特征融合	［128×L， 128］
输出层	［128+128， 128， 64， 1］

数据集	方法	MAE/MPa	MedAE/MPa	预测时间/s
悬臂梁	全连接网络	0.657	0.553	2.07×10^-3
	LSTM^［2］	0.224	0.188	6.82×10^-4
	PCAFeatureNN^［7］	0.485	0.460	3.25×10^-3
	MeshGraphNet^［8］	0.198	0.137	4.38×10^-3
	MeshNet^［10］	0.290	0.251	3.63×10^-3
	本文方法	0.049	0.036	2.63×10^-3
方向盘	MeshGraphNet^［8］	8.09	7.50	3.72×10^-1
	MeshNet^［10］	13.20	12.30	3.08×10^-1
	本文方法	6.85	6.16	2.23×10^-1

方法	MAE	MedAE
全连接网络	1.207	0.953
LSTM^［2］	0.686	0.344
PCAFeatureNN^［7］	0.666	0.497
MeshGraphNet^［8］	0.349	0.249
MeshNet^［10］	0.342	0.271
本文方法	0.133	0.052

[1]	Jieru JIA, Jianchao YANG, Shuorui ZHANG, Tao YAN, Bin CHEN. Unsupervised person re-identification based on self-distilled vision Transformer [J]. Journal of Computer Applications, 2024, 44(9): 2893-2902.
[2]	Zongze JIA, Pengfei GAO, Yinglong MA, Xiaofeng LIU, Haixin XIA. Multi-feature fusion attention-based hierarchical classification method for dialogue act [J]. Journal of Computer Applications, 2024, 44(3): 715-721.
[3]	Yongfeng DONG, Yahan DENG, Yao DONG, Yacong WANG. Survey of clustering based on deep learning [J]. Journal of Computer Applications, 2022, 42(4): 1021-1028.
[4]	CHEN Li, WANG Hongyuan, ZHANG Yunpeng, CAO Liang, YIN Yuchang. Video-based person re-identification method by jointing evenly sampling-random erasing and global temporal feature pooling [J]. Journal of Computer Applications, 2021, 41(1): 164-169.
[5]	DING Jianli, LI Yang, WANG Jialiang. Short text automatic summarization method based on dual encoder [J]. Journal of Computer Applications, 2019, 39(12): 3476-3481.
[6]	CHANG Bingguo, ZANG Hongying. Time series classifier design based on piecewise dimensionality reduction and updated dynamic time warping [J]. Journal of Computer Applications, 2018, 38(7): 1910-1915.
[7]	YANG Xiaoling, LI Zhiqing, LIU Yutong. Automatic image annotation based on multi-label discriminative dictionary learning [J]. Journal of Computer Applications, 2018, 38(5): 1294-1298.
[8]	ZHANG Luosheng, TONG Jing. Real-time interaction based modeling method for 3D objects with relief-texture [J]. Journal of Computer Applications, 2017, 37(8): 2302-2306.
[9]	LI Yandong, HAO Zongbo, LEI Hang. Survey of convolutional neural network [J]. Journal of Computer Applications, 2016, 36(9): 2508-2515.
[10]	LOU Ziting, ZHANG Yaping. Mesh layout algorithm based on greedy optimization strategy [J]. Journal of Computer Applications, 2016, 36(7): 1954-1958.
[11]	YANG Xingyao YU Jiong TURGUN Ibrahim LIAO Bin. Collaborative filtering model combining users' and items' predictions [J]. Journal of Computer Applications, 2013, 33(12): 3354-3358.
[12]	LIU Tong-tong DAI Min LI Zhong-yi. ECG waveform similarity analysis based on window-slope representation [J]. Journal of Computer Applications, 2012, 32(10): 2969-2972.
[13]	XIANG Jun DA Bang-you LIANG Juan HOU Jian-hua. New feature description based on feature relationships for gait recognition [J]. Journal of Computer Applications, 2012, 32(03): 885-888.
[14]	. A fast topological reconstruction algorithm for 3D mesh model [J]. Journal of Computer Applications, 2010, 30(11): 3002-3004.
[15]	Hua SONG Jiang LIU. Algorithm to generate progressive meshes based on quadric error metric [J]. Journal of Computer Applications, 2008, 28(12): 3160-3162.