连续梁桥减隔震支座参数多目标优化设计

Multi-objective optimization design of seismic isolation bearings of continuous girder bridges

摘要

减隔震支座通过增加桥梁柔性，能够有效降低主梁传递至桥墩的内力，但过大的柔性可能导致主梁位移过大，进而引发碰撞或落梁问题。本文旨在优化减隔震支座的参数设计，实现内力与位移之间的合理平衡。本文提出了一种结合机器学习代理模型：极端梯度提升(eXtreme gradient boosting，XGBoost)、多目标遗传算法：非支配遗传算法Ⅱ(non-dominated sorting genetic algorithm-Ⅱ，NSGA-Ⅱ)与多目标决策方法的优化框架，用于减隔震支座参数的高效设计。首先，建立考虑非线性效应的桥梁三维有限元模型，利用XGBoost算法构建回归模型以提高NSGA-Ⅱ的计算效率；然后，通过多目标决策方法从帕累托前沿解集中选择最优设计方案。结果表明，使用XGBoost构建的代理模型替代有限元模型，在保证精度(拟合优度R²>0.95)的情况下，优化效率提高了83％以上。优化方案相比初始设计，墩底弯矩仅增加4.0％，支座变形显著减少35.5％，在不显著增加弯矩的情况下，有效降低了结构位移。本文提出的优化框架在减隔震支座设计等多目标优化问题中具有良好的应用潜力。

Abstract

Although seismic isolation devices can reduce the internal forces in bridges, they may increase the displacement of the main girder, potentially leading to collisions between girders or even girder dropping. Therefore, it is crucial to achieve a reasonable balance between internal forces and displacements within the seismic isolation bridge, making the parameter design of the seismic isolation bearings a critical aspect. Traditional design methods typically rely on extensive time history analysis and empirical calculations, resulting in low computational efficiency and difficulties in achieving a balance among multiple objectives. This paper proposes an optimization method combining machine learning surrogate models, multi-objective genetic algorithm (NSGA-Ⅱ), and multi-criteria decision theory for the design of seismic isolation parameters in continuous girder bridges. Firstly, a three-dimensional finite element model of the bridge considering nonlinear effects is established. The XGBoost machine learning algorithm is utilized to build regression models for target responses, enhancing the computational efficiency of the NSGA-Ⅱ genetic algorithm. Subsequently, NSGA-Ⅱ is applied to find the Pareto front solutions, and the entropy-weighted TOPSIS method is employed to select the final optimized design from the Pareto front. Research results demonstrate that using a machine learning method (XGBoost herein) to develop a surrogate model as a substitution for the finite element model improves optimization efficiency by at least 83%. Without compromising accuracy (coefficient of determination R ² > 0.95). Compared to the initial design, pier bottom moments only increased by 4.0%, while bearing deformations decreased by 35.5%. The optimization framework presented in this study demonstrates promising applicability to multi-objective optimization problems, particularly in the context of seismic isolation bearing design.

关键词

减隔震 / 多目标优化 / 机器学习 / 遗传算法 / 多目标决策

Key words

seismic isolation / multi-objective optimization / machine learning / genetic algorithm / multi-criteria decision-making

参考文献

[1] 袁一凡 . 四川汶川8.0级地震损失评估[J]. 地震工程与工程振动， 2008， 28(5): 10-19.

[2] YUAN Yifan . Loss assessment of Wenchuan earthquake[J]. Earthquake Engineering and Engineering Dynamics, 2008, 28(5): 10-19. (in Chinese)

[3] 石岩, 王东升, 韩建平, 等 . 桥梁减隔震技术的应用现状与发展趋势[J]. 地震工程与工程振动， 2017， 37(5): 118-128.

[4] SHI Yan, WANG Dongsheng, HAN Jianping, et al. Application status of seismic isolation for bridges and its development tendency[J]. Earthquake Engineering and Engineering Dynamics, 2017, 37(5): 118-128. (in Chinese)

[5] 张煜敏, 翁光远, 程凯凯, 等 . 不同减隔震支座组合对连续梁桥抗震性能影响分析[J]. 灾害学， 2023， 38(4): 34-38，55.

[6] ZHANG Yumin, WENG Guangyuan, CHENG Kaikai, et al. Study on the influence of different isolation bearing combinations on the seismic performance of continuous beam bridge[J]. Journal of Catastrophology, 2023, 38(4): 34-38, 55. (in Chinese)

[7] 许莉, 李煜民, 丁自豪, 等 . 采用改进多目标粒子群算法的斜拉桥阻尼器参数优化[J]. 振动工程学报， 2024， 37(6): 1006-1014.

[8] XU Li, LI Yumin, DING Zihao, et al. Parameter optimization of viscous damper for cable-stayed bridge using improved multi-objective particle swarm algorithm[J]. Journal of Vibration Engineering, 2024, 37(6): 1006-1014. (in Chinese)

[9] SONEJI B B, JANGID R S . Passive hybrid systems for earthquake protection of cable-stayed bridge[J]. Engineering Structures, 2007, 29(1): 57-70.

[10] ZAKIAN P, KAVEH A . Multi-objective seismic design optimization of structures: A review[J]. Archives of Computational Methods in Engineering, 2024, 31(2): 579-594.

[11] XIE Y Z, ZHANG J . Optimal design of seismic protective devices for highway bridges using performance-based methodology and multiobjective genetic optimization[J]. Journal of Bridge Engineering, 2017, 22(3): 04016129.

[12] LI S, WANG J Q, SHAHRIA ALAM M . Multi-criteria optimal design and seismic assessment of SMA RC piers and SMA cable restrainers for mitigating seismic damage of simply-supported highway bridges[J]. Engineering Structures, 2022, 252: 113547.

[13] STEIN M . Large sample properties of simulations using Latin hypercube sampling[J]. Technometrics, 1987, 29(2): 143-151.

[14] ZHONG X, GAO H, LI J, et al. Multi-objective seismic design optimization of self-centering bridges with novel mechanical hinges[J]. Engineering Structures, 2025, 343: 121061.

[15] 张维科, 刘正楠, 陈兴冲, 等 . 基于深度学习的摇摆自复位桥墩设计参数优化方法[J]. 工程科学与技术， 2024， 56(6): 185-196.

[16] ZHANG Weike, LIU Zhengnan, CHEN Xingchong, et al. Design parameter optimization of self-centering pier based on deep learning[J]. Advanced Engineering Sciences, 2024, 56(6): 185-196. (in Chinese)

[17] CORADDAU A, ONETO L, LI S, et al. Surrogate models to unlock the optimal design of stiffened panels accounting for ultimate strength reduction due to welding residual stress[J]. Engineering Structures, 2023, 293: 116645.

[18] WANG J C, YEA J, WANG X W . Quantifying easy-to-repair displacement ductility and lateral strength of scoured bridge pile group foundations in cohesionless soils: A classification-regression combination surrogate model[J]. Journal of Bridge Engineering, 2023, 28(11): 04023080.

[19] WANG X W, MAZUMDER R K, SALARIEH B, et al. Machine learning for risk and resilience assessment in structural engineering: Progress and future trends[J]. Journal of Structural Engineering, 2022, 148(8): 03122003.

[20] FENG D C, WANG W J, MANGALATHU S, et al. Interpretable XGBoost-SHAP machine-learning model for shear strength prediction of squat RC walls[J]. Journal of Structural Engineering, 2021, 147(11): 04021173.

[21] WANG X W, LI Z Q, SHAFIEEZADEH A . Seismic response prediction and variable importance analysis of extended pile-shaft-supported bridges against lateral spreading: Exploring optimized machine learning models[J]. Engineering Structures, 2021, 236: 112142.

[22] WAKJIRA T G, RAHMZADEH A, ALAM M S, et al. Explainable machine learning based efficient prediction tool for lateral cyclic response of post-tensioned base rocking steel bridge piers[J]. Structures, 2022, 44: 947-964.

[23] 徐亚峰, 付晓, 关波, 等 . 基于多目标优化的独柱墩直线梁桥抗倾覆加固[J]. 公路交通科技， 2024， 41(6): 109-118.

[24] XU Yafeng, FU Xiao, GUAN Bo, et al. Anti-overturning reinforcement for single-column pier straight beam bridge based on multi-objective optimization[J]. Journal of Highway and Transportation Research and Development, 2024, 41(6): 109-118. (in Chinese)

[25] 戴理朝, 康哲, 陈瑞, 等 . 基于NSGA-Ⅲ的桥梁网络多目标维修决策优化研究[J]. 土木工程学报， 2024， 57(5): 41-52.

[26] DAI Lizhao, KANG Zhe, CHEN Rui, et al. Study on multi-objective maintenance decision optimization for bridge networks based on NSGA-Ⅲ[J]. China Civil Engineering Journal, 2024, 57(5): 41-52. (in Chinese)

[27] 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.

[28] 彭涛, 田仲初, 张建仁, 等 . 基于多目标优化的混凝土斜拉桥静动力有限元模型修正[J]. 振动与冲击， 2018， 37(21): 108-116.

[29] PENG Tao, TIAN Zhongchu, ZHANG Jianren, et al. Static and dynamic finite element model updating for a concrete cable-stayed bridge based on multi-objective optimization[J]. Journal of Vibration and Shock, 2018, 37(21): 108-116. (in Chinese)

[30] 彭雪梅, 刘夫云, 孙永厚, 等 . 基于熵权-改进TOPSIS法的驾驶室多目标优化[J]. 机械设计， 2023， 40(11): 122-127.

[31] PENG Xuemei, LIU Fuyun, SUN Yonghou, et al. Cab multi-objective optimization based on entropy weight-improved TOPSIS method[J]. Journal of Machine Design, 2023, 40(11): 122-127. (in Chinese)

[32] 张先起, 梁川, 刘慧卿 . 基于熵权的改进TOPSIS法在水质评价中的应用[J]. 哈尔滨工业大学学报， 2007， 39(10): 1670-1672.

[33] ZHANG Xianqi, LIANG Chuan, LIU Huiqing . Application of improved TOPSIS method based on coefficient of entropy to comprehensive evaluating water quality[J]. Journal of Harbin Institute of Technology, 2007, 39(10): 1670-1672. (in Chinese)

[34] 王力东, 付岱林, 卜秀孟, 等 . 基于自适应代理模型的桥梁风屏障参数多目标优化方法[J/OL]. 工程力学， 2024: 1-12[2025-01-03]. https://link.cnki.net/urlid/11.2595.O3.20240710.2101.007.

[35] WANG Lidong, FU Dailin, BU Xiumeng, et al. Multi-objective optimization method for bridge wind barrier parameters based on adaptive surrogate models[J/OL]. Engineering Mechanics, 2024: 1-12[2025-01-03]. https://link.cnki.net/urlid/11.2595.O3.20240710.2101.007. (in Chinese)

[36] CHEN C H . A novel multi-criteria decision-making model for building material supplier selection based on entropy-AHP weighted TOPSIS[J]. Entropy, 2020, 22(2): 259.

[37] MCKENNA F, FENVES G L, SCOTT M H . OpenSees: open system for earthquake engineering simulation[CP]. Berkeley, CA: Pacific Earthquake Engineering Research Center, 2006[2025-08-01]. http://opensees.berkeley.edu/.

[38] 丁永刚, 李政, 王建强 . 高阻尼橡胶支座压缩-水平双向剪切作用下力学性能试验研究[J]. 工程抗震与加固改造， 2021， 43(4): 128-132，159.

[39] DING Yonggang, LI Zheng, WANG Jianqiang . Experimental study on mechanical properties of high damping rubber bearing under compression-bilateral shear loading[J]. Earthquake Resistant Engineering and Retrofitting, 2021, 43(4): 128-132, 159. (in Chinese)

[40] MANDER J B, PRIESTLEY M J N, PARK R . Theoretical stress-strain model for confined concrete[J]. Journal of Structural Engineering, 1988, 114(8): 1804-1826.

[41] SCOTT M H, FENVES G L . Plastic hinge integration methods for force-based beam-column elements[J]. Journal of Structural Engineering, 2006, 132(2): 244-252.

[42] SHEN Y, LI Y X, XU W J, et al. Evaluation of seismic-induced impact interaction between a cable-stayed bridge and its approach spans using a simplified analysis model[J]. Journal of Earthquake Engineering, 2022, 26(7): 3343-3363.

[43] GAO H, LI J, CHEN X, et al. Combined concrete and cable restrainers to prevent longitudinal unseating of highway bridges during earthquakes[J]. Engineering Structures, 2024, 314: 118388.

[44] SHAMSABADI A, KAPUSKAR M . Nonlinear soil-abutment-foundation-structure interaction analysis of skewed bridges subjected to near-field ground motions[J]. Transportation Research Record: Journal of the Transportation Research Board, 2010, 2202(1): 192-205.

[45] GAO H Y, LI J Z, CHEN X, et al. Combined concrete and cable restrainers to prevent longitudinal unseating of highway bridges during earthquakes[J]. Engineering Structures, 2024, 314: 118388.

[46] JTG/T 2231-01- 2020 公路桥梁抗震设计规范[S]. 北京: 人民交通出版社， 2020.

[47] JTG/T 2231-01- 2020 Specifications for seismic design of highway bridges[S]. Beijing: China Communications Press, 2020. (in Chinese)

[48] OZBULUT O E, HURLEBAUS S . Optimal design of superelastic-friction base isolators for seismic protection of highway bridges against near-field earthquakes[J]. Earthquake Engineering & Structural Dynamics, 2011, 40(3): 273-291.

[49] FEMA P695. Quantification of building seismic performance factors[R]. Washington: Federal Emergency Management Agency, 2009.

[50] ZHANG H Y, YANG Y . Optimization of mechanical performance of seismic isolation bearings for continuous beam bridges[J]. Journal of Vibration Engineering & Technologies, 2024, 12(3): 3653-3665.

[51] JT/T 842- 2012 公路桥梁高阻尼隔震橡胶支座[S]. 北京: 人民交通出版社， 2012.

[52] JT/T 842- 2012 High damping seismic isolation rubber bearings for highway bridges[S]. Beijing: China Communications Press, 2012. (in Chinese)

[53] ZHENG W J, DOERR B . Mathematical runtime analysis for the non-dominated sorting genetic algorithm II (NSGA-Ⅱ)[J]. Artificial Intelligence, 2023, 325: 104016.

[54] CAMACHO V T, HORTA N, LOPES M, et al. Optimizing earthquake design of reinforced concrete bridge infrastructures based on evolutionary computation techniques[J]. Structural and Multidisciplinary Optimization, 2020, 61(3): 1087-1105.

[55] FANG C, PING Y W, GAO Y Q, et al. Machine learning-aided multi-objective optimization of structures with hybrid braces-Framework and case study[J]. Engineering Structures, 2022, 269: 114808.

基金资助

中建八局总承包公司课题基金项目(2022-04)

福建省自然科学基金创青项目(2026J008106)