Multi-objective Optimization Design Method Based on a Hybrid Metric Adaptive Sampling Surrogate Model

doi:10.3901/JME.2024.13.081

Abstract

Abstract: The high nonlinearity, computational complexity, and dimensionality of existing engineering problems, along with the demand for low-cost and high-fidelity simulation models, have increased the difficulty of solving multi-objective optimization designs of engineering structures based on multidisciplinary coupling effects. This has led to extensive research in this field due to the large computational volume involved. To address this challenge, this paper proposes a method to realize the multi-objective optimization design of engineering structures based on a new hybrid metric adaptive sampling surrogate model. Firstly, to reduce the cost of optimization design, a new hybrid metric adaptive sampling method based on the Voronoi tessellation is proposed by comprehensively considering the characteristics of global exploration and local exploitation of the entire design space. It is used for constructing a global surrogate model. After comparative testing with different cases and methods, the proposed hybrid metric adaptive sampling method can significantly reduce the required samples while ensuring the same accuracy. And then, to realize the multi-objective optimization design of engineering structure, a new NSGA-II-RD(Improved non-dominated sorting genetic algorithm II based on a rotation and density operator, NSGA-II-RD) multi-objective optimization design algorithm based on the new congestion operator of dominant surface rotational projection and Voronoi tessellation is proposed, which is tested in comparison with different algorithms and shows faster convergence speed and accurate calculation results. Finally, the proposed hybrid metric adaptive sampling surrogate modeling method is combined with the NSGA-II-RD algorithm and applied to the design of busbar structures for insulated-gate bipolar transistors to perform multi-objective optimization design considering the quality, circuit voltage drops, and fatigue damage of the busbars. The results demonstrate that the optimization design method ensures excellent lightweight and conductivity performance and enhances fatigue resistance during ultrasonic welding. Meanwhile, it has been verified that this method can effectively solve multi-objective optimization problems in practical engineering while ensuring low-cost and high-accuracy simulation models.

Key words: hybrid metric, adaptive sampling, surrogate model, NSGA-II-RD, multi-objective optimization design

CLC Number:

TP202

ZHAO Feng, HU Weifei, LI Guang, DENG Xiaoyu, LIU Zhenyu, GUO Yunfei, TAN Jianrong. Multi-objective Optimization Design Method Based on a Hybrid Metric Adaptive Sampling Surrogate Model[J]. Journal of Mechanical Engineering, 2024, 60(13): 81-91.

References

[1] 韩忠华. Kriging模型及代理优化算法研究进展[J]. 航空学报, 2016, 37(11)：3197-3225. HAN Zhonghua. Kriging surrogate model and its application to design optimization：A review of recent progress[J]. Acta Aeronautica et Astronautica Sinica, 2016, 37(11)：3197-3225.
[2] 于梦阁, 潘振宽, 蒋荣超, 等. 基于近似模型的高速列车头型多目标优化设计[J]. 机械工程学报, 2019, 55(24)：178-186. YU Mengge, PAN Zhenkuan, JIANG Rongchao, et al. Multi-objective optimization design of the high-speed train head based on the approximate model[J]. Journal of Mechanical Engineering, 2019, 55(24)：178-186.
[3] BEZERRA M, SANTELLI R, OLIVEIRA E, et al. Response surface methodology (RSM) as a tool for optimization in analytical chemistry[J]. Talanta, 2008, 76(5)：965-977.
[4] ZHOU Q, JIANG P, SHAO X, et al. A variable fidelity information fusion method based on radial basis function[J]. Advanced Engineering Informatics, 2017, 32：26-39.
[5] KLEIJNEN J. Kriging metamodeling in simulation：A review[J]. European Journal of Operational Research, 2009, 192(3)：707-716.
[6] HU W, ZHAO F, DENG X, et al. A new sequential sampling method for surrogate modeling based on a hybrid metric[J]. Journal of Mechanical Design, 2024, 146(6)：061705.
[7] FUHG J, FAU A, NACKENHORST U. State-of-the-art and comparative review of adaptive sampling methods for kriging[J]. Archives of Computational Methods in Engineering, 2021, 28(4)：2689-2747.
[8] XU S, LIU H, WANG X, et al. A robust error-pursuing sequential sampling approach for global metamodeling based on Voronoi diagram and cross validation[J]. Journal of Mechanical Design, 2014, 136(7)：071009.
[9] YAO W, CHEN X, LUO W. A gradient-based sequential radial basis function neural network modeling method[J]. Neural Computing and Applications, 2009, 18(5)：477-484.
[10] LIU H, CAI J, ONG Y. An adaptive sampling approach for kriging metamodeling by maximizing expected prediction error[J]. Computers & Chemical Engineering, 2017, 106：171-182.
[11] 王平, 郑松林, 吴光强. 基于协同优化和多目标遗传算法的车身结构多学科优化设计[J]. 机械工程学报, 2011, 47(2)：102-108. WANG Ping, ZHENG SongPin, WU Guangqiang. Multidisciplinary design optimization of vehicle body structure based on collaborative optimization and multi-objective genetic algorithm[J]. Journal of Mechanical Engineering, 2011, 47(2)：102-108.
[12] 孙光永, 李光耀, 陈涛, 等. 多目标粒子群优化算法在薄板冲压成形中的应用[J]. 机械工程学报, 2009, 45(5)：153-159. SUN Guangyong, LI Guangyao, CHEN Tao. Application of multi-objective particle swarm optimization in sheet metal forming[J]. Journal of Mechanical Engineering, 2009, 45(5)：153-159.
[13] 邢宇飞, 王成恩, 柳强. 基于Pareto解集蚁群算法的拆卸序列规划[J]. 机械工程学报, 2012, 48(9)：186-192. XING Yufei, WANG Chengen, LIU Qiang. Disassembly sequence planning based on Pareto ant colony algorithm[J]. Journal of Mechanical Engineering, 2012, 48(9)：186-192.
[14] ZHANG Q, LI H. MOEA/D: A multiobjective evolutionary algorithm based on decomposition[J]. IEEE Transactions on Evolutionary Computation, 2007, 11(6)：712-731.
[15] QI Y, MA X, LIU F, et al. MOEA/D with adaptive weight adjustment[J]. Evolutionary Computation, 2014, 22(2)：231-264.
[16] AURENHAMMER F. Voronoi diagrams-a survey of a fundamental geometric data structure[J]. ACM Computing Surveys, 1991, 23(3)：345-405.
[17] 方伟, 宋鑫宏. 基于Voronoi图盲区的无线传感器网络覆盖控制部署策略[J]. 物理学报, 2014, 63(22)：2207011-22070110. FANG Wei, SONG Xinhong. A deployment strategy for coverage control in wireless sensor networks based on the blind-zone of Voronoi diagram[J]. Acta Physica Sinica, 2014, 63(22)：2207011-22070110.
[18] AUTE V, SALEH K, ABDELAZIZ O, et al. Cross-validation based single response adaptive design of experiments for kriging metamodeling of deterministic computer simulations[J]. Structural and Multidisciplinary Optimization, 2013, 48(3)：581-605.
[19] 程锦, 谭建荣, 余加红. 基于TOPSIS的注塑工艺参数多目标稳健优化设计[J]. 机械工程学报, 2011, 47(6)：27-32. CHENG Jin, TAN Jianrong, YU Jiahong. Multi-objective robust optimization of injection molding process parameters based on TOPSIS[J]. Journal of Mechanical Engineering, 2011, 47(6)：27-32.
[20] MO S, LU D, SHI X, et al. A Taylor expansion‐based adaptive design strategy for global surrogate modeling with applications in groundwater modeling[J]. Water Resources Research, 2017, 53(12)：10802-10823.
[21] DEB K, PRATAP A, AGARWAL S, et al. A fast and elitist multiobjective genetic algorithm: NSGA-II[J]. IEEE Transactions on Evolutionary Computation, 2002, 6(2)：182-197.
[22] 徐弓岳, 丁华锋, 孙玉玉. 基于改进非支配排序遗传算法的正铲挖掘机工作装置优化设计[J]. 机械工程学报, 2016, 52(21)：35-43. XU Gongyue, DING Huafeng, SUN Yuyu. Optimization of face-shovel excavator's attachment based on improved NSGA-II[J]. Journal of Mechanical Engineering, 2016, 52(21)：35-43.
[23] ZHANG Q, ZHOU A, ZHAO S, et al. Multiobjective optimization test instances for the CEC 2009 special session and competition[R]. University of Essex and Nanyang Technological University, Essex, UK and Singapore, technical report, 2008.
[24] BADER J, ZITZLER E. HypE: An algorithm for fast hypervolume-based many-objective optimization[J]. Evolutionary Computation, 2011, 19(1)：45-76.
[25] ZITZLER E, DEB K, THIELE L. Comparison of multiobjective evolutionary algorithms: Empirical results[J]. Evolutionary Computation, 2000, 8(2)：173-195.
[26] DEB K, THIELE L, LAUMANNS M, et al. Scalable test problems for evolutionary multiobjective optimization[M]. London：Springer London, 2005.
[27] TIAN Y, CHENG R, ZHANG X, et al. PlatEMO：A MATLAB platform for evolutionary multi-objective optimization[J]. IEEE Computational Intelligence Magazine, 2017, 12(4)：73-87.
[28] LI D, PACKWOOD M, QI F, et al. 3D multiphysics modelling of high voltage IGBT module packaging[C]//Proceedings of the 201516th International Conference on Electronic Packaging Technology (ICEPT). IEEE, 2015.
[29] HU W, CHOI K, CHO H. Reliability-based design optimization of wind turbine blades for fatigue life under dynamic wind load uncertainty[J]. Structural and Multidisciplinary Optimization, 2016, 54(4)：953-970.