基于可解释机器学习的锈蚀RC构件抗剪承载力预测模型

doi:10.13229/j.cnki.jdxbgxb.20230024

Abstract

Abstract:

The prediction model for shear capacity of corroded reinforced concrete （RC） beams based on the mechanism method usually introduces a series of assumptions and correction coefficients， resulting in low accuracy of calculation results and limited applicability. In the present study， based on the data drive， considering the reliability of the black box model and the rationality of the input features， the key basic characteristics of the corroded RC beams， such as geometric dimensions， longitudinal reinforcement ratio， stirrup yield strength， stirrup corrosion loss， and concrete strength， were selected. A practical model of shear capacity based on an interpretable machine learning algorithm was established. The results show that the corrosion loss， effective height， shear-to-span ratio and beam width are sensitive to the structural shear capacity. The practical model of shear capacity of corroded RC beams based on the data-driven can reveal the underlying mapping relationship between the basic features and the shear capacity. The proposed model have good applicability and prediction accuracy compared with the empirical model and black box model.

Key words: civil engineering, RC beams, corrosion, shear capacity, interpretability machine learning, feature selection

CLC Number:

TU375.1

Li-zhao DAI,Chong WANG,Ping YUAN,Lei WANG. Prediction model for shear capacity of corroded RC beams based on interpretable machine learning[J].Journal of Jilin University(Engineering and Technology Edition), 2024, 54(11): 3231-3243.

Figures/Tables 12

Table 1

Shear test data of corroded reinforced concrete beams"

学者	$f c$ /MPa	$b$ /mm	$h 0$ /mm	$ρ l$ /mm	$ρ v$ /%	$f y v$ /kN	$A s v$ /mm²	$s$ /mm	$λ$	$η l$ /%	$η w$ /%	$V u$ /kN	加载方式
薛昕等^［3］	15.8~18.8	120	220	2.17	0.39	300	57	120	1.5~3.2	0~17.3	0~16.7	70.5~124.3	单点
柳世涛^［27］	19.5~23.9	200	265	2.15	0.14~0.25	339~524	57~101	150~250	2~3.5	0~12.1	0~60.1	93.8~181.5	双点
Xia等^［28］	12.97	120	200	2.62	0.48~0.56	321~463	57~101	100~150	1.5	0	0~54.2	85.2~138.2	双点
霍飞格^［29］	16.8~17.6	250	475	1.63	0.08~0.13	331~337	43~66	200	1.74	0	0~36.3	205.8~275.6	双点
赵冰等^［30］	9.5	100	175	1.94	0.44	324	66	150	1.5~2.5	0	4~23.3	41.9~53.0	双点
杨晓明等^［31］	15.8~18.7	120	220	2.17	0.39	300	57	120	1.5~3.2	0~17.3	0~16.8	70.5~123.3	双点
李冰等^［32］	18.3~26.7	150	150	2.68	0.19~0.25	444.15	57	150~200	2.0~3.0	0~26.2	0~32.7	52.1~78.2	双点
赵羽习等^［9］	11.9	150	155	2.26~2.79	0.19~0.45	331.52	66~101	100~200	2.2~3.1	0~26.0	0~9.2	40.0~72.0	单点
李学田等^［33］	18.1	150	175	2.3	0.25	275	57	150	1.5~2.2	0~3.4	0~2.8	75.9~131.0	单点
霍艳华^［4］	9.2~12.8	100	175	1.94	0.44	324	66	150	1.5~2.5	0~20.0	3.0~30.8	41.9~53.0	双点
祝建军^［34］	17.9	100	175	2.06	0	0	0	0	1.5~2.5	0~12.5	0	19.3~44.2	双点
Higgins等^［35］	17.6~20.0	254	521	1.90	0.16~0.20	585	101	203~305	2.04	0	0~42.5	324.0~475.2	双点
王小惠等^［36］	17.0	150	170	1.58	0	0	0	0	3.97	1.0~4.7	0	24.4~36.3	双点
徐善华等^［8］	15.1~16.3	120	200	1.92	0.32	275	57	150	1~2	0	0~38.7	57.7~146.8	双点
Rodriguez等^［2］	14.3	150	170	1.77	0.22	626	57	85~170	4.7	19.0~31.9	0~93.8	27.7~38.6	双点

Table 1

Fig.1

Fig.2

Table 2

Fig.3

Fig.4

Fig.5

Table 3

Feature importance ranking of each training model"

训练模型	特征重要性排序
训练模型	1	2	3	4	5	6	7	8	9	10	11
RF	$h 0$	$b$	$λ$	$f c$	$f y v$	$ρ v$	$η w$	$A s v$	$ρ l$	$s$	$η l$
RF	46.3%	14.7%	12.7%	6.9%	6.6%	4.4%	2.9%	2.0%	1.9%	0.9%	0.7%
XGBoost	$h 0$	$b$	$λ$	$η w$	$f c$	$η l$	$ρ l$	$f y v$	$ρ v$	$s$	$A s v$
XGBoost	32.6%	24.7%	21.4%	4.6%	3.8%	3.2%	3.2%	3.1%	1.7%	1.1%	0.6%
SVR	$h 0$	$λ$	$b$	$ρ l$	$η w$	$A s v$	$f y v$	$η l$	$s$	$ρ v$	$f c$
SVR	24.8%	20.3%	19.1%	7.0%	5.8%	5.1%	4.5%	3.7%	3.6%	3.2%	2.9%

Table 3

Fig.6

Table 4

Calculation model for shear capacity of RC beams"

作者	计算模型
李士斌等^［1］	集中荷载下的梁： $V c s = 1.75 f t b c h 0 c 1 + λ + f y v c A s v c h 0 c s$ ，矩形截面梁： $V c s = 0.7 f t b c h 0 c + 1.25 f y v A s v h 0 c s$
Ahmed K^［47］	$V c s = 0.17 λ f c' b e f f h 0 + A v, e f f f y v h 0 s$
赵羽习等^［9］	$V c s = P v V c s 0, P v = 1.0, ?????????????? η w ≤ 10 % 1.17 - 1.7 η w, η w > 10 %, V C S 0 = b h 0 λ C s h 0 1 - 0.5 C s h 0 f c' + 0.5 ρ v f y v 1 - C s h 0 2 λ 2$
霍艳华^［4］	$V c s = φ 1.75 1 + λ f t b h 0 + γ A s v f y v h 0 s, φ = 1.0, ????????????????? η l ≤ 5 % 1.098 - 1.96 η l, η l > 5 %, γ = 1 - 1.059 η w$

Table 4

Table 5

Evaluation index of each calculation model for shear capacity of RC beams"

评价指标	李士斌等^［1］	Ahmed K^［47］	赵羽习等^［9］	霍艳华^［4］	本文模型	SVR	MLP	XGBoost	RF
$μ$	1.58	0.94	2.13	1.99	0.89	0.96	0.96	0.96	0.92
RMSE	61.46	49.37	92.07	68.5	25.6	13.3	14.6	14.9	20.1
MAE	41.88	37.7	65.13	47.53	17.8	8.8	9.9	10.1	13.0
MAPE/%	36.50	64.61	57.91	40.52	20.68	10.57	12.56	10.84	14.77
R2	0.844	0.898	0.653	0.807	0.948	0.985	0.982	0.982	0.966

Table 5

Fig.7

References 47

1	李士彬, 张鑫, 贾留东, 等. 箍筋锈蚀钢筋混凝土梁的抗剪承载力分析[J]. 工程力学, 2011, 28(S1): 60-63.
	Li Shi-bin, Zhang Xin, Jia Liu-dong, et al. Analysis for shear capacity of reinforced concrete beams with corrosion stirrups[J]. Engineering Mechanics, 2011, 28(S1): 60-63.
2	Rodriguez J, Ortega L M, Casal J. Load carrying capacity of concrete structures with corroded reinforcement[J]. Construction and Building Materials, 1997, 11(4): 239-248.
3	薛昕, 杨成. 纵向受拉钢筋锈蚀钢筋混凝土梁的抗剪性能[J]. 上海交通大学学报, 2016, 50(11): 1767-1772.
	Xue Xin, Yang Cheng. Shear performance of reinforced concrete beams containing corroded longitudinal bars[J]. Journal of Shanghai Jiaotong University, 2016, 50(11): 1767-1772.
4	霍艳华. 锈蚀钢筋混凝土简支梁受剪承载力研究[J]. 工业建筑, 2006, 36(z1): 910-912.
	Huo Yan-hua. Research on shear capacity of simply supported concrete beam with corroded reinforcement[J]. Industrial Construction, 2006, 36(z1): 910-912.
5	梁兴文, 汪萍, 徐明雪, 等. 配筋超高性能混凝土梁受弯性能及承载力研究[J]. 工程力学, 2019, 36(5): 110-119.
	Liang Xing-wen, Wang Ping, Xu Ming-xue, et al. Investigation on flexural capacity of reinforced ultra high performance concrete beams[J]. Engineering Mechanics, 2019, 36(5): 110-119.
6	何世钦, 贡金鑫. 钢筋混凝土梁中锈蚀钢筋粘结性能的试验研究[J]. 哈尔滨工业大学学报, 2006, 38(12): 2167-2170.
	He Shi-qin, Gong Jin-xin. Experimental studies on bond characteristics of corroded steel bar in reinforced concrete beams[J]. Journal of Harbin Institute of Technology, 2006, 38(12): 2167-2170.
7	戴理朝, 周亮, 杨晓文, 等. 基于Connector单元的锈蚀RC梁界面粘结性能细观数值模拟[J]. 吉林大学学报: 工学版, 2023, 53(10): 2886-2896.
	Dai Li-chao, Zhou Liang, Yang Xiao-wen, et al. Microscopic numerical simulation of interface bond performance of corroded RC beam based on connector element[J]. Journal of Jilin University (Engineering and Technology Edition), 2023, 53(10): 2886-2896.
8	徐善华, 牛荻涛. 锈蚀钢筋混凝土简支梁斜截面抗剪性能研究[J]. 建筑结构学报, 2004, 25(5): 98-104.
	Xu Shan-hua, Niu Di-tao. The shear behavior of corroded simply supported reinforced concrete beam[J]. Journal of Building Structures, 2004, 25(5): 98-104.
9	赵羽习, 金伟良. 锈蚀箍筋混凝土梁的抗剪承载力分析[J]. 浙江大学学报: 工学版, 2008, 42(1): 19-24.
	Zhao Yu-xi, Jin Wei-liang. Analysis on shearing capacity of concrete beams with corroded stirrups[J]. Journal of Zhejiang University (Engineering Science), 2008, 42(1): 19-24.
10	卢朝辉, 李海, 赵衍刚, 等. 锈蚀钢筋混凝土梁抗剪承载力预测经验模型[J]. 工程力学, 2015, 32(z1): 261-270.
	Lu Zhao-hui, Li Hai, Zhao Yan-gang, et al. An empirical model for shear strength prediction of corroded RC beams[J]. Engineering Mechanics, 2015, 32(z1): 261-270.
11	金浏, 王涛, 杜修力. 无腹筋RC悬臂梁抗剪强度及尺寸效应理论研究[J]. 计算力学学报, 2020, 37(4): 396-404.
	Jin Liu, Wang Tao, Du Xiu-li. Size effect theory on shear strength of RC cantilever beams without stirrups[J]. Chinese Journal of Computational Mechanics, 2020, 37(4): 396-404.
12	Huang H, Burton H V. Classification of in-plane failure modes for reinforced concrete frames with infills using machine learning[J]. Journal of Building Engineering, 2019, 25: No.100767.
13	Mangalathu S, Jeon J S. Classification of failure mode and prediction of shear strength for reinforced concrete beam-column joints using machine learning techniques[J]. Engineering Structures, 2018, 160: 85-94.
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): No. 04019104.
15	Mangalathu S, Jeon J S, DesRoches R. Critical uncertainty parameters influencing seismic performance of bridges using Lasso regression[J]. Earthquake Engineering & Structural Dynamics, 2018, 47(3): 784-801.
16	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: No.101126.
17	Fu B, Feng D C. A machine learning-based time-dependent shear strength model for corroded reinforced concrete beams[J]. Journal of Building Engineering, 2021, 36: No.102118.
18	Nguyen T H, Nguyen D T, Nguyen D H, et al. Evaluation of residual strength of corroded reinforced concrete beams using machine learning models[J]. Arabian Journal for Science and Engineering, 2022, 47(8): 9985-10002.
19	Hoang N D, Tran X L, Nguyen H. Predicting ultimate bond strength of corroded reinforcement and surrounding concrete using a metaheuristic optimized least squares support vector regression model[J]. Neural Computing and Applications, 2020, 32: 7289-7309.
20	Lundberg S M, Lee S I. A unified approach to interpreting model predictions[C]∥Advances in Neural Information Processing Systems, Long Beach, USA, 2017: 4765-4774.
21	Koza J R. Genetic programming as a means for programming computers by natural selection[J]. Statistics and Computing, 1994, 4: 87-112.
22	Mangalathu S, Hwang S H, Jeon J S. Failure mode and effects analysis of RC members based on machine-learning-based shapley additive explanations(SHAP) approach[J]. Engineering Structures, 2020, 219: No.110927.
23	冯德成, 吴刚. 混凝土结构基本性能的可解释机器学习建模方法[J]. 建筑结构学报, 2022, 43(4): 228-238.
	Feng De-cheng, Wu Gang. Interpretable machine learning-based modeling approach for fundamental properties of concrete structures[J]. Journal of Building Structures, 2022, 43(4): 228-238.
24	Chaabene W B, Nehdi M L. Genetic programming based symbolic regression for shear capacity prediction of SFRC beams[J]. Construction and Building Materials, 2021, 280: No.122523.
25	Gandomi A H, Alavi A H, Mohammadzadeh S D, et al. An empirical model for shear capacity of RC deep beams using genetic-simulated annealing[J]. Archives of Civil and Mechanical Engineering, 2013, 13: 354-369.
26	Jeong H, Han S J, Choi S H, et al. Genetic programming approach and data generation for transfer lengths in pretensioned concrete members[J]. Engineering Structures, 2021, 231: No.111747.
27	柳世涛. 受腐蚀钢筋混凝土梁抗剪性能研究[D]. 长沙: 中南大学土木工程学院, 2013.
	Liu Shi-tao. Research on shear behavior of corroded RC beams[D]. Changsha: School of Civil Engineering, Central South University, 2013.
28	Xia J, Jin W, Li L. Shear performance of reinforced concrete beams with corroded stirrups in chloride environment[J]. Corrosion Science, 2011, 53(5): 1794-1805.
29	翟飞格. 箍筋锈蚀配斜筋RC梁抗剪承载力研究[D]. 长沙: 长沙理工大学土木工程学院, 2010.
	Zhai Fei-ge. The research on shear capacity of corroded stirrups RC beams with diagonal reinforcement[D]. Changsha: School of Civil Engineering, Changsha University of Science and Technology, 2010.
30	赵冰, 曾凡振. 锈蚀箍筋混凝土简支梁抗剪承载力模型及有限元分析[J]. 中国农村水利水电, 2010, 2010(10): 88-91.
	Zhao Bing, Zeng Fan-zhen. The model and finite element analysis of shear capacity of simply supported concrete beam with corroded stirrup[J]. China Rural Water and Hydropower, 2010, 2010(10): 88-91.
31	杨晓明, 吴桐, 王耀耀. 小剪跨比锈蚀钢筋混凝土梁受剪性能试验研究[J]. 建筑结构学报, 2019, 40(12): 147-154.
	Yang Xiao-ming, Wu tong, Wang Yao-yao. Experimental study on shear behavior of corroded reinforced concrete beams with low shear span ratio[J]. Journal of Building Structures, 2019, 40(12): 147-154.
32	李冰, 高向华, 王小惠, 等. 局部区段锈蚀的钢筋混凝土梁抗剪承载力试验研究[J]. 混凝土与水泥制品, 2010, 2010(6): 60-65.
	Li Bing, Gao Xiang-hua, Wang Xiao-hui, et al. Experimental study on shear capacity of reinforced concrete beams with local section corrosion[J]. Concrete and Cement Products, 2010, 2010(6): 60-65.
33	李学田, 殷惠光. 锈蚀钢筋混凝土梁抗剪能力退化机理和预计模型[J]. 徐州工程学院学报: 自然科学版, 2010, 25(4): 58-63.
	Li Xue-tian, Yin Hui-guang. Degradation mechanism and predicting models of shearing capacity for corroded reinforced concrete beams[J]. Journal of Xuzhou Institute of Technology (Natural Sciences Edition), 2010, 25(4): 58-63.
34	祝建军. 纵筋锈蚀对无腹筋梁斜截面受剪承载力的影响[D]. 南昌: 南昌大学建筑工程学院, 2007.
	Zhu Jian-jun. Effect of corroded longitutal reinforcements on shear capacity of simple supported concrete beam without stirrups[D]. Nanchang: School of Civil Engineering and Architecture, Nanchang University, 2007.
35	Higgins C, Farrow W C. Tests of reinforced concrete beams with corrosion-damaged stirrups[J]. ACI Materials Journal, 2006, 103(1): 133-141.
36	王小惠, 刘西拉. 基于粘结强度变化的锈蚀无腹筋梁承载能力的研究[J]. 四川建筑科学研究, 2007, 33(3): 56-61.
	Wang Xiao-hui, Liu Xi-la. Research on load carrying capacity of corroded RC beam without web reinforcement on the basis of the change of bond strength[J]. Sichuan Building Science, 2007, 33(3): 56-61.
37	. 混凝土结构设计规范 [S].
38	Liaw A, Wiener M. Classification and regression by random forest[J]. R News, 2002, 2(3): 18-22.
39	Safavian S R, Landgrebe D. A survey of decision tree classifier methodology[J]. IEEE Transactions on Systems, Man, and Cybernetics, 1991, 21(3): 660-674.
40	Freund Y, Schapire R E. A decision-theoretic generalization of on-line learning and an application to boosting[J]. Journal of Computer and System Sciences, 1997, 55(1): 119-139.
41	Chen T, Guestrin C. Xgboost: a scalable tree boosting system[C]∥Proceedings of the 22nd ACM Sigkdd International Conference on Knowledge Discovery and Data Mining, San Francisco, USA, 2016: 785-794.
42	Rumelhart D E, Hinton G E, Williams R J. Learning representations by back-propagating errors[J]. Nature, 1986, 323(6088): 533-536.
43	Hearst M A, Dumais S T, Osuna E, et al. Support vector machines[J]. IEEE Intelligent Systems and Their Applications, 1998, 13(4): 18-28.
44	Friedman J H. On bias, variance, 0/1—loss, and the curse-of-dimensionality[J]. Data Mining and Knowledge Discovery, 1997, 1: 55-77.
45	Kuhn M, Johnson K. Feature Engineering and Selection: A Practical Approach for Predictive Models[M]. Boca Raton: Chapman & Hall/CRC Press, 2019.
46	Breiman L. Random forests[J]. Machine Learning, 2001, 45: 5-32.
47	Elsayed A K. Shear capacity assessment of reinforced concrete beams with corroded stirrups[J]. Construction and Building Materials, 2017, 134: 176-184.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

Comments

Recommended 0

No Suggested Reading articles found!

数据集	模型	指标
数据集	模型	R2	MAE	MAPE/%	RMSE
训练结果	RF	0.990	4.102	4.626	7.17
	XGBoost	0.998	1.367	1.572	2.90
	MLP	0.995	3.285	3.940	5.204
	SVR	0.993	3.003	3.438	6.252
测试结果	RF	0.966	12.995	14.774	20.133
	XGBoost	0.982	10.063	10.836	14.887
	MLP	0.982	9.850	12.557	14.634
	SVR	0.985	8.834	10.568	13.344

[1]	Fu-cheng WANG,Xin-rong ZHAO,Jia-bing TIAN,Guo-liang XIE,Li-ming ZHOU. Influence of rice straw ash on compressive properties and microstructure of concrete [J]. Journal of Jilin University(Engineering and Technology Edition), 2024, 54(9): 2620-2630.
[2]	Lin XIAO,Huan-bo WEI,Xing WEI,Zhi-rui KANG. Numerical analysis on cracking behavior of concrete slab due to corrosion expansion of stud connector in steel-concrete composite beam [J]. Journal of Jilin University(Engineering and Technology Edition), 2024, 54(7): 1958-1965.
[3]	Yi-fan LIU,Zhi-wei MIAO,Chen SHEN,Xiang-dong GENG. Evaluation of mechanical properties of non-uniform corroded rebars based on Monte Carlo method [J]. Journal of Jilin University(Engineering and Technology Edition), 2024, 54(4): 1007-1015.
[4]	Qiong FENG,Hao-zheng TIAN,Hong-xia QIAO,Teng-fei NIAN,Wen-wen HAN. Corrosion deterioration and equivalent relationship between natural exposure and salt spray accelerated environment of reinforced concrete [J]. Journal of Jilin University(Engineering and Technology Edition), 2024, 54(2): 494-505.
[5]	Bo-song FAN,Chun-fu SHAO. Urban rail transit emergency risk level identification method [J]. Journal of Jilin University(Engineering and Technology Edition), 2024, 54(2): 427-435.
[6]	Wen-jie CHEN,Zhen-xing SU,Xian-tao SUN,Yuan-yuan LIU,Xiang-tao HU,Ya-li ZHI. Feature extraction of speech signals of exoskeleton devices in noise environments [J]. Journal of Jilin University(Engineering and Technology Edition), 2024, 54(10): 3050-3057.
[7]	Feng-feng ZHOU,Tao YU,Yu-si FAN. Generative adversarial autoencoder integrated voting algorithm based on mass spectral data [J]. Journal of Jilin University(Engineering and Technology Edition), 2024, 54(10): 2969-2977.
[8]	Ran AN,You-zhi WANG. Shear properties of shear stud connectors under combined tension and shear loading [J]. Journal of Jilin University(Engineering and Technology Edition), 2023, 53(9): 2554-2562.
[9]	Qing-feng YAN,Ji-gang ZHANG,Tao WANG,De-gang CHEN,You-sheng YU,Ying-chun YANG. Seismic performance of connection joints between prefabricated prefinished volumetric construction [J]. Journal of Jilin University(Engineering and Technology Edition), 2023, 53(2): 505-514.
[10]	Feng-feng ZHOU,Zhen-wei YAN. A model for identifying neuropeptides by feature selection based on hybrid features [J]. Journal of Jilin University(Engineering and Technology Edition), 2023, 53(11): 3238-3245.
[11]	Li-zhao DAI,Liang ZHOU,Xiao-wen YANG,Lei WANG. Meso-scale numerical simulation of interfacial bond behavior of corroded RC beams based on connector element [J]. Journal of Jilin University(Engineering and Technology Edition), 2023, 53(10): 2886-2896.
[12]	Hui GUO,Jie-di FU,Zhen-dong LI,Yan YAN,Xiao LI. SVM parameters and feature selection optimization based on improved whale algorithm [J]. Journal of Jilin University(Engineering and Technology Edition), 2023, 53(10): 2952-2963.
[13]	Pei-ze LI,Shi-shun ZHAO,Xiao-hui WENG,Xin-mei JIANG,Hong-bo CUI,Jian-lei QIAO,Zhi-yong CHANG. A new method for rapid detection of pesticide residues based on multi⁃sensor optimization [J]. Journal of Jilin University(Engineering and Technology Edition), 2022, 52(8): 1951-1956.
[14]	Feng-feng ZHOU,Hai-yang ZHU. SEE： sense EEG⁃based emotion algorithm via three⁃step feature selection strategy [J]. Journal of Jilin University(Engineering and Technology Edition), 2022, 52(8): 1834-1841.
[15]	Wei-hong CHEN,Yan CHEN,Qiu-rong HONG,Shuang-shuang CUI,Xue-yuan YAN. Seismic performance of earthquake⁃damaged precast concrete frame structures strengthened with BRBs [J]. Journal of Jilin University(Engineering and Technology Edition), 2022, 52(8): 1817-1825.

Prediction model for shear capacity of corroded RC beams based on interpretable machine learning

RICH HTML

PDF (PC)

Abstract

Cite this article

share this article

Figures/Tables 12

References 47

Related Articles 15

Metrics

Comments

Recommended 0