Data-driven Design Optimization of Heterogeneous Cellular Structures for Realizing the Specified Deformation

doi:10.3901/JME.2024.01.085

Abstract

Abstract: Regarding the challenge of realizing the design optimization of heterogeneous cellular structures, a design optimization method is proposed for designing the heterogeneous cellular structures with specified deformation. In the macro design part, the gradient-based optimization method is used to optimize the distribution of equivalent properties, and the feasibility constraint is introduced to ensure the realizability of microstructures. The data driven method is introduced into the microstructure design, where the deep learning is used to realize the inverse design of the microstructure. Then, the microstructures are mapped to the macro structure, which finally results into non-uniform design. At the same time, the offset and compatibility algorithm of microstructures are used to tackle the problem of incompatibility between adjacent microstructures. Finally, the effectiveness of the method is verified by several numerical examples of the specified deformation problem, the effects of size of elements, the number of microstructural offsets, and the microstructure mapping density on the structural performance are discussed. The proposed method successfully realizes the requirement of specifying the deformation mode of the heterogeneous cellular structure by introducing the data driven method and the compatibility of adjacent microstructures.

Key words: heterogeneous cellular structure, specified deformation, deep learning, microstructure offset

CLC Number:

TP391

XU Siyuan, ZHANG Weidong, MAO Yuming, NIU Bin. Data-driven Design Optimization of Heterogeneous Cellular Structures for Realizing the Specified Deformation[J]. Journal of Mechanical Engineering, 2024, 60(1): 85-95.

References

[1] BAUER J,HENGSBACH S,TESARI I,et al. High-Strength cellular ceramic composites with 3D microarchitecture[C]//Proceedings of the National Academy of Sciences,2014,111(7):2453-2458.
[2] ELNASRI I,PATTOFATTO S,ZHAO H,et al. Shock enhancement of cellular structures under impact loading:Part i experiments[J]. Journal of the Mechanics and Physics of Solids,2007,55(12):2652-2671.
[3] GOLOVIN I S,SINNING H R. Damping in some cellular metallic materials[J]. Journal of Alloys and Compounds,2003,355(1-2):2-9.
[4] 王向明,苏亚东,吴斌,等. 微桁架点阵结构在飞机结构/功能一体化中的应用[J]. 航空制造技术,2018,61(10):16-25. WANG Xiangming,SU Yadong,WU Bin,et al. Application for additive manufacturing of lattice materials on integrated aircraft structures and functions[J]. Aeronautical Manufacturing Technology,2018,61(10):16-25.
[5] DUMAS M,TERRIAULT P,BRAILOVSKI V. Modelling and characterization of a porosity graded lattice structure for additively manufactured biomaterials[J]. Materials & Design,2017,121:383-392.
[6] RODRIGUES H,GUEDES J M,BENDSOE M P. Hierarchical optimization of material and structure[J]. Structural and Multidisciplinary Optimization,2002,24:1-10.
[7] LIU Ling,YAN Jun,CHENG Gengdong. Optimum structure with homogeneous optimum truss-like material[J]. Computers & Structures,2008,86(13-14):1417-1425.
[8] NIU Bin,YAN Jun,CHENG Gengdong. Optimum structure with homogeneous optimum cellular material for maximum fundamental frequency[J]. Structural and Multidisciplinary Optimization,2009,39:115-132.
[9] WANG Chuang,GU Xiaojun,ZHU Jihong,et al. Concurrent design of hierarchical structures with three-dimensional parameterized lattice microstructures for additive manufacturing[J]. Structural and Multidisciplinary Optimization,2020,61:869-894.
[10] HAN Yafeng,LU Wenfeng. A novel design method for nonuniform lattice structures based on topology optimization[J]. Journal of Mechanical Design,2018,140(9):091403.
[11] HU Jingqiao,LI Ming,YANG Xingtong,et al. Cellular structure design based on free material optimization under connectivity control[J]. Computer-Aided Design,2020,127:102854.
[12] 阎军,许琦,张起,等. 人工智能在结构拓扑优化领域的现状与未来趋势[J]. 计算力学学报,2021,38(4):412-422. YAN Jun,XU Qi,ZHANG Qi,et al. Current and future trends of artificial intelligence in the field of structure topology optimization[J]. Chinese Journal of Computational Mechanics,2021,38(4):412-422.
[13] ZHANG Yan,GAO Liang,XIAO Mi. Maximizing natural frequencies of inhomogeneous cellular structures by kriging-assisted multiscale topology optimization[J]. Computers & Structures,2020,230:106197.
[14] QIU Zheng,LI Quhao,LIU Shutian,et al. Clustering-based concurrent topology optimization with macrostructure,components,and materials[J]. Structural and Multidisciplinary Optimization,2021,63(3):1243-1263.
[15] WANG Liwei,CHAN Yuchin,AHMED F,et al. Deep generative modeling for mechanistic-based learning and design of metamaterial systems[J]. Computer Methods in Applied Mechanics and Engineering,2020,372:113377.
[16] ZHU B,SKOURAS M,CHEN D,et al. Two-scale topology optimization with microstructures[J]. ACM Transactions on Graphics,2017,36(5):1-16.
[17] RADFORD A,METZ L,CHINTALA S. Unsupervised representation learning with deep convolutional generative adversarial networks[EB/OL]. (2015-11-19) arXiv preprint arXiv:1511. 06434,2015. https://arxiv.org/abs/1511.06434
[18] GOODFELLOW I,BENGIO Y,COURVILLE A. Deep learning[M]. Cambridge:MIT Press,2017.
[19] SVANBERG K. The method of moving asymptotes-a new method for structural optimization[J]. International Journal for Numerical Methods in Engineering,1987,24(2):359-373.
[20] LI Xiaolin,YANG Zijiang,BRINSON L C,et al. A deep adversarial learning methodology for designing microstructural material systems[C]//Proceedings of the ASME 2018 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference 2018,Quebec,Canada,2018.
[21] TAN Renkai,ZHANG Nevin,YE Wenjing. A Deep learning-based method for the design of microstructural materials[J]. Structural and Multidisciplinary Optimization,2020,61(4):1417-1438.
[22] GOODFELLOW I J,POUGET-ABADIE J,MIRZA M,et al. Generative adversarial nets[R]. 2014,27.
[23] WANG Liwei,CHAN Yuchin,LIU Zhao,et al. Data-driven metamaterial design with laplace-beltrami spectrum as "Shape-DNA"[J]. Structural and Multidisciplinary Optimization,2020,61(6):2613-2628.
[24] HEUSEL M,RAMSAUER H,UNTERTHINER T,et al. GANs trained by a two time-scale update rule converge to a local nash equilibrium[EB/OL].(2017-06-26) arXiv preprint arXiv:1706. 08500,2017. https://arxiv.org/abs/1706.08500.
[25] MIYATO T,KATAOKA T,KOYAMA M,et al. Spectral normalization for generative adversarial networks[EB/OL]. (2018-02-16) arXiv preprint arXiv:1802. 05957,2018. https://arxiv.org/abs/1802.05957.
[26] BERGSTRA J,YAMINS D,COX D D. Making a science of model search:hyperparameter optimization in hundreds of dimensions for vision architectures[C]//Proceedings of the 30th International Conference on Machine Learning (ICML 2013),Atlanta,USA,2013,28:115-123.
[27] ANDREASSEN E,ANDREASEN C S. How to determine composite material properties using numerical homogenization[J]. Computational Materials Science,2014,83:488-495.
[28] SETHIAN J A. A fast marching level set method for monotonically advancing fronts[J]. Proceedings of the National Academy of Sciences of the United States of America,1996,93(4):1591-1595.
[29] MORITZ P,NISHIHARA R,WANG S,et al. Ray:A distributed framework for emerging AI applications[EB/OL]. (2017-12-16). arXiv preprint arXiv:1712. 05889,2017. https://arxiv.org/abs/1712.05889.