Adaptive simulation and data-driven hybrid modeling for predicting shear strength and failure modes of interior reinforced concrete beam-column joints

被引:0
作者
Mehta, Vikas [1 ]
Jang, Sung Hyun [1 ]
Chey, Min Ho [1 ]
机构
[1] Keimyung Univ, Dept Civil Engn, Smart Struct Lab, Daegu 42601, South Korea
关键词
Beam-column joints; Machine learning; Reinforced concrete; Shear strength; SEISMIC BEHAVIOR; PERFORMANCE; CONNECTIONS; DESIGN; LOAD; CLASSIFICATION; ECCENTRICITY;
D O I
10.1016/j.istruc.2025.108835
中图分类号
TU [建筑科学];
学科分类号
0813 ;
摘要
This study investigates the key factors influencing failure modes and shear strength in interior reinforced concrete beam-column joints (RCBCJs) using an integrated machine learning framework. A dataset of 200 experimental records was analyzed, with Principal Component Analysis (PCA) refining essential parameters and the Synthetic Minority Over-sampling Technique (SMOTE) addressing class imbalance through the generation of 136 synthetic instances, equilibrating the minority class to 140 samples. Five classification and regression algorithms were evaluated, with the Random Forest (RF) model demonstrating superior predictive performance. For shear strength prediction, the RF model achieved a training relative absolute error (RAE) of 0.11 and a coefficient of determination (R2 = 0.99), outperforming conventional design codes (ACI 318-14, EN1998-I:2004). Testing yielded an RAE of 0.27 and R2 = 0.94, demonstrating robust generalizability. In failure mode classification, the model attained 98 % training accuracy and 84 % testing accuracy, surpassing the performance of empirical codebased methods. SHapley Additive exPlanations (SHAP) analysis revealed beam width (bb) and column height (hc) as the most influential factors for failure modes (mean absolute SHAP = 0.09 and 0.05). For shear strength, column height (hc) had the highest impact (mean absolute SHAP = 76.52), followed by top (Asb,top; 64.02) and bottom (Asb,bot; 54.74) beam reinforcement areas. The RF model consistently surpassed existing design standards, validating its capacity to capture complex parameter interactions. To bridge research and practice, a user-friendly graphical user interface (GUI) was developed, enabling streamlined RCBCJ design optimization by integrating data-driven insights with structural engineering principles.
引用
收藏
页数:19
相关论文
共 105 条
[1]  
ACI Committee 318, 2014, Building code requirements for structural concrete (ACI 318-14) and commentary (ACI 318R-14), P28
[2]   Neural network prediction of joint shear strength of exterior beam-column joint [J].
Alagundi, Shreyas ;
Palanisamy, T. .
STRUCTURES, 2022, 37 :1002-1018
[3]  
Alire DA, 2002, Seismic evaluation of existing unconfined RC beam-column joints
[4]   Load-carrying capacity and mode failure simulation of beam-column joint connection: Application of self-tuning machine learning model [J].
Alwanas, Afrah Abdulelah Hamzah ;
Al-Musawi, Abeer A. ;
Salih, Sinan Q. ;
Tao, Hai ;
Ali, Mumtaz ;
Yaseen, Zaher Mundher .
ENGINEERING STRUCTURES, 2019, 194 :220-229
[5]   Shape quantization and recognition with randomized trees [J].
Amit, Y ;
Geman, D .
NEURAL COMPUTATION, 1997, 9 (07) :1545-1588
[6]  
[Anonymous], 2004, Eurocode 8, UNI EN 1998-1:2005, P30
[7]  
Asadi S., 2016, RipMC: RIPPER for multiclass classification
[8]  
Attaalla SA, 2004, ACI STRUCT J, V101, P65
[9]  
Au FTK, 2005, P I CIVIL ENG-STR B, V158, P21, DOI 10.1680/stbu.2005.158.1.21
[10]   Seismic Behavior of RC wide beam-column connections under dynamic loading [J].
Benavent-Climent, A. .
JOURNAL OF EARTHQUAKE ENGINEERING, 2007, 11 (04) :493-511