物理规律监督的RC柱地震破坏模式可解释机器学习方法

成浩, 喻泽成, 余波

建筑结构学报 ›› 2023, Vol. 44 ›› Issue (11) : 69-79.

PDF(6644 KB)
PDF(6644 KB)
建筑结构学报 ›› 2023, Vol. 44 ›› Issue (11) : 69-79. DOI: 10.14006/j.jzjgxb.2022.0370
结构抗灾及减灾专题

物理规律监督的RC柱地震破坏模式可解释机器学习方法

  • 成浩1, 喻泽成1, 余波1,2,3
作者信息 +

A physics-supervised interpretable machine learning approach for seismic failure modes prediction of RC columns

  • CHENG Hao1, YU Zecheng1, YU Bo1,2,3
Author information +
文章历史 +

摘要

针对传统机器学习模型难以有效兼顾预测准确率与可解释性,且存在模型预测结果与实际物理规律不相符的问题,提出物理规律监督的RC柱地震破坏模式可解释机器学习方法。首先结合RC柱的地震破坏模式演化机制、试验数据规律分析和工程经验,揭示RC柱关键特征参数与地震破坏模式之间的物理规律;然后采用多分类逻辑回归算法提出了RC柱地震破坏模式的可解释模型;进而融合物理规律和等比例K-折交叉验证算法确定了符合物理规律且预测误差最小的最优超参数;最后基于多分类逻辑回归可解释模型和最优超参数,建立物理规律监督的RC柱地震破坏模式多分类逻辑回归预测模型。分析结果表明:该模型可以有效兼顾准确率与可解释性,预测结果不仅满足关键特征参数与地震破坏模式之间的物理规律,而且可以合理反映RC柱地震破坏模式之间的竞争关系,避免了传统逻辑回归模型存在预测结果与实际物理规律不相符的缺陷;与传统的经验判别方法和机器学习模型相比,该模型的准确率可以分别提高12%～35%和4%～9%。

Abstract

Traditional machine learning models have difficulties in balancing the prediction accuracy and model interpretability, as well as have inconsistent results between model predictions and actual physical laws. In order to overcome these limitations, a physics-supervised interpretable machine learning approach was proposed to predict the seismic failure modes of reinforced concrete (RC) columns. The physical laws between key characteristic parameters and seismic failure modes of RC columns were revealed first based on the seismic failure modes evolution mechanism of RC columns, laws analysis of test data, and engineering experience. Then, an interpretable model for predicting the seismic failure modes of RC columns was proposed by multiclass logistic regression algorithm. Furthermore, the optimal hyperparameter satisfying the physical laws and with the minimum error were obtained by encoding the physical laws and equal proportion K-fold cross-validation algorithm. Finally, a physics-supervised multiclass logistic regression (PMLR) prediction model for failure modes of RC columns was established based on the multiclass logistic regression interpretable model and optimal hyperparameter. The results show that the proposed model has satisfactory accuracy and natural interpretability. The prediction results not only meet the actual physical laws between key characteristic parameters and seismic failure modes of RC columns, but also reflect the competitive relationship between seismic failure modes of RC columns.The proposed model overcomes the limitation that the prediction results are inconsistent with actual physical laws by traditional logistic regression (LR) model. The prediction accuracy of the proposed model is improved by 12%-35% and 4%-9% respectively when compared with traditional empirical classification methods and machine learning models.

关键词

钢筋混凝土柱 / 物理规律 / 可解释性 / 机器学习 / 破坏模式

Key words

reinforced concrete column / physical law / interpretability / machine learning / failure mode

引用本文

EndNote

Ris (Procite)

Bibtex

导出引用
成浩, 喻泽成, 余波. 物理规律监督的RC柱地震破坏模式可解释机器学习方法[J]. 建筑结构学报, 2023, 44(11): 69-79. https://doi.org/10.14006/j.jzjgxb.2022.0370
CHENG Hao, YU Zecheng, YU Bo. A physics-supervised interpretable machine learning approach for seismic failure modes prediction of RC columns[J]. Journal of Building Structures, 2023, 44(11): 69-79. https://doi.org/10.14006/j.jzjgxb.2022.0370

参考文献

[1] YU B, NING C L, LI B. Hysteretic model for shear-critical reinforced concrete columns [J]. Journal of Structural Engineering, 2016, 142(9): 04016056.
[2] ELWOOD K J, MOEHLE J P. Dynamic shear and axial-load failure of reinforced concrete columns [J]. Journal of Structural Engineering, 2008, 134(7): 1189-1198.
[3] MA Y, GONG J X. Probability identification of seismic failure modes of reinforced concrete columns based on experimental observations [J]. Journal of Earthquake Engineering, 2018, 22(10): 1881-1899.
[4] QI Y L, HAN X L, JI J. Failure mode classification of reinforced concrete column using Fisher method [J]. Journal of Central South University, 2013, 20(10): 2863-2869.
[5] GHEE A B, PRIESTLEY M J N, PAULAY T. Seismic shear strength of circular reinforced concrete columns [J]. ACI Structural Journal, 1989, 86(1): 45-59.
[6] 万海涛, 韩小雷,季静. 基于性能设计方法的钢筋混凝土柱构件分析[J]. 中南大学学报(自然科学版), 2010, 41(4): 1584-1589. (WAN Haitao, HAN Xiaolei, JI Jing. Analyses of reinforced concrete columns by performance-based design method [J]. Journal of Central South University (Science and Technology),2010,41(4):1584-1589.(in Chinese))
[7] American Society of Civil Engineers. Seismic evaluation and retrofit of existing buildings:ASCE/SEI 41-17[S]. Reston, Virginia: American Society of Civil Engineers, 2017.
[8] 刘力维, 易伟建. 弯剪破坏与轴力附加弯矩对钢筋混凝土柱极限变形能力的影响综述[J]. 建筑结构学报, 2019, 40(8): 1-11. (LIU Liwei, YI Weijian. A critical review of influence of flexural-shear failure mode and P-Delta effect on ultimate deformation capacity of RC columns [J]. Journal of Building Structures, 2019, 40(8): 1-11. (in Chinese))
[9] 张勤, 王娜,贡金鑫. 钢筋混凝土柱地震破坏模式及考虑剪切变形的抗震性能研究进展[J]. 建筑结构学报,2017, 38(8): 1-13. (ZHANG Qin, WANG Na, GONG Jinxin. State of the art of seismic performance including shear effects and failure modes of reinforced concrete columns [J]. Journal of Building Structures, 2017, 38(8): 1-13. (in Chinese))
[10] 马颖, 张勤,贡金鑫. 钢筋混凝土柱弯剪破坏恢复力模型骨架曲线[J]. 建筑结构学报, 2012, 33(10): 116-125. (MA Ying, ZHANG Qin, GONG Jinxin. Skeleton curves of restoring force model of reinforced concrete columns failed in flexure-shear [J]. Journal of Building Structures, 2012, 33(10): 116-125. (in Chinese))
[11] 高向玲, 李杰. 反复荷载作用下钢筋与纤维混凝土黏结性能试验研究[J]. 建筑结构学报, 2012, 33(12): 121-129. (GAO Xiangling, LI Jie. Study on bond performance between steel and fiber reinforced concrete under cyclic loading [J]. Journal of Building Structures, 2012, 33(12): 121-129. (in Chinese))
[12] 任晓丹, 刘凯,魏公涛,等. 不同加载速率下箍筋约束混凝土力学性能试验研究[J]. 建筑结构学报, 2017, 38(3): 141-150. (REN Xiaodan, LIU Kai, WEI Gongtao, et al. Experimental study on mechanical behavior of stirrups confined concrete under loadings at different rates [J]. Journal of Building Structures, 2017, 38(3): 141-150. (in Chinese))
[13] 于晓辉, 钱凯,吕大刚. 考虑悬链线效应的钢筋混凝土框架结构抗连续倒塌能力分析[J]. 建筑结构学报, 2017, 38(4): 28-34. (YÜ Xiaohui, QIAN Kai, LÜ Dagang. Progressive collapse capacity analysis of reinforced concrete frame structures considering catenary action [J]. Journal of Building Structures, 2017, 38(4): 28-34. (in Chinese))
[14] MANGALATHU S, JEON J S. Machine learning-based failure mode recognition of circular reinforced concrete bridge columns: comparative study [J]. Journal of Structural Engineering, 2019, 145(10): 04019104.
[15] 于晓辉, 王猛,宁超列. 基于机器学习的钢筋混凝土柱失效模式两阶段判别方法[J]. 建筑结构学报, 2022, 43(8): 220-231. (YU Xiaohui, WANG Meng, NING Chaolie. A machine-learning-based two-step method for failure mode classification of reinforced concrete columns [J]. Journal of Building Structures, 2022, 43(8): 220-231. (in Chinese))
[16] NADERPOUR H, MIRRASHID M, PARSA P. Failure mode prediction of reinforced concrete columns using machine learning methods [J]. Engineering Structures, 2021, 248: 113263.
[17] FENG D C, LIU Z T, WANG X D, et al. Failure mode classification and bearing capacity prediction for reinforced concrete columns based on ensemble machine learning algorithm [J]. Advanced Engineering Informatics, 2020, 45: 101126.
[18] 冯德成, 吴刚. 混凝土结构基本性能的可解释机器学习建模[J]. 建筑结构学报, 2022, 43(4): 228-238. (FENG Decheng, WU Gang. Interpretable machine learning-based modeling approach for fundamental properties of concrete structures [J]. Journal of Building Structures, 2022, 43(4): 228-238. (in Chinese))
[19] LIU Z L, GUO A X. Empirical-based support vector machine method for seismic assessment and simulation of reinforced concrete columns using historical cyclic tests [J]. Engineering Structures, 2021, 237: 112141.
[20] 喻泽成, 李启明,谢龙隆,等. 基于穷举搜索策略与逻辑回归算法的RC柱地震破坏模式判别模型[J]. 工程力学, 2022, 39(10): 99-110. (YU Zecheng, LI Qiming, XIE Longlong, et al. Discrimination model of seismic failure mode of RC columns based on exhaustive search strategy and logistic regression algorithm [J]. Engineering Mechanics, 2022, 39(10): 99-110. (in Chinese))
[21] 温增平, 徐超,陆鸣,等. 汶川地震重灾区典型钢筋混凝土框架结构震害现象[J]. 北京工业大学学报, 2009, 35(6): 753-760. (WEN Zengping, XU Chao, LU Ming, et al. Damage features of RC frame structures in Wenchuan earthquake [J]. Journal of Beijing University of Technology, 2009, 35(6): 753-760. (in Chinese))
[22] 霍林生, 李宏男,肖诗云,等. 汶川地震钢筋混凝土框架结构震害调查与启示[J]. 大连理工大学学报, 2009, 49(5): 718-723. (HUO Linsheng, LI Hongnan, XIAO Shiyun, et al. Earthquake damage investigation and analysis of reinforced concrete frame structures in Wenchuan earthquake [J]. Journal of Dalian University of Technology, 2009, 49(5): 718-723. (in Chinese))
[23] BISHOP C M. Pattern recognition and machine learning [M]. New York: Springer, 2006: 209-210.
[24] PEDREGOSA F, VAROQUAUX G, GRAMFORT A, et al. Scikit-learn: machine learning in python [J]. Journal of Machine Learning Research, 2011, 12: 2825-2830.
[25] BOYD S, VANDENBERGHE L. Convex optimization [M]. Cambridge: Cambridge University Press, 2004.
[26] 李启明, 喻泽成,余波,等. 钢筋混凝土柱地震破坏模式判别的两阶段支持向量机方法[J]. 工程力学, 2022, 39(2): 148-158. (LI Qiming, YU Zecheng, YU Bo, et al. Two-stage support vector machine method for failure mode classification of reinforced concrete columns [J]. Engineering Mechanics, 2022, 39(2): 148-158. (in Chinese))

基金

国家自然科学基金项目(52278162, 62266005),广西杰出青年科学基金项目(2019GXNSFFA245004),广西重点研发计划项目(桂科AB23026026)。
PDF(6644 KB)

140

Accesses

0

Citation

Detail
段落导航
相关文章
总访问量： 　 今日访问： 　 在线人数：
版权所有 © 《建筑结构学报》 杂志社有限公司
地址：北京三里河路13号      邮编：100835      电话：010-88029882
推荐采用谷歌、火狐、Edge等浏览器登录网站。
京ICP备20000062号-1　 京公网安备 11010802041942号　网络出版服务许可证（署）网出证（京）字第373号　出版物经营许可证：新出发京批字第版240007号　技术支持：北京玛格泰克科技发展有限公司　51La

/

〈 〉