Strut and tie model for predicting shear behavior of interior RC beam column joints under seismic loading Authors: Wentong Zhao, Yulin Yan, Yaxin Yue, Ru Xue, Xiang Yan Published: 2025 ABSTRACT Beam–column joint with complex mechanical behavior is an important part of reinforced concrete (RC) structures. The present study aims to assess the joint shear stress–strain relationship by adopting strut–and–tie approach of the interior beam-column subassemblages. Totally six interior RC beam–column joint specimens were tested under cyclic loading. The test results of beam–column subassemblages show that under high shear stress, the joint cores always perform obvious inelastic shear deformations. It is evident that when using finite element simulation to study the seismic behavior of RC beam–column joints, accurately modeling the shear stress–strain relationship of joint core is critical for successfully implementing the finite element analysis. Based on the experimental results, this study innovatively proposed a strut–and–tie model (STM) for RC interior beam–column joints to investigate shear behavior within joint core. This STM was established according to the deformation characteristics and the crack patterns of the joint region. The force equilibrium conditions and deformation compatibility requirements of this model were formulated. Comparisons were made between the predicted and published experimental results of the twenty–three RC interior beam-column joints with respect to joint shear stress and strain envelope curve at this state to investigate the adequacy of the proposed STM. The results demonstrate that the joint shear stress and strain envelope curves predicted by the proposed STM show closer agreement with experimental data. The mean absolute error (MAE) of peak shear stress calculated by the proposed STM is significantly lower than that of the Zhao-Yang model. The coefficient of determination (R2) is significantly higher than that of the Zhao-Yang model. These findings confirm that the proposed STM exhibits high prediction accuracy for the shear stress–strain relationship of beam-column joints.The calculation results of the proposed STM were used in finite element analysis of the beam–column joint tests based on the Beam Column Joint Element in the OpenSees software. The load–displacement hysteresis curves and joint shear deformations obtained from the finite element analysis were compared against experimental results of beam–column joints. Analytical results further validated the applicability and efficacy of the proposed STM. Introduction In the seismic resistant design of moment-resisting frames, the beam-column joint is regarded as a crucial point for consideration. Post–earthquake research indicates that, under high shear stress, the joint core always exhibits obvious inelastic deformations, which can lead to large lateral displacement or even collapse of the RC structures. Hence, comprehending the shear behavior of joints is essential to control the overall performance of RC beam-column frame structures. Since the mid-twentieth century, a range of experimental research and finite element analyses have been carried out to explore the seismic behavior of the beam–column joint. When using finite element method to analyze the mechanical behavior of beam–column joint, it is vital to calculate the joint shear stress versus strain relationship of the joint core correctly. Research indicates that the force transfer mechanisms within the core of beam–column joint can be characterized by diagonal compression strut, truss, and confined mechanism (Fig. 1). For calculating the joint shear stress and strain (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\tau { - }\gamma$$\end{document}) relationship correctly, it is imperative to consider the contributions of the three kinds of force transfer mechanisms appropriately. Lowes and Altoontash1 and Shi et al.2 used the modified compression field theory (MCFT) to calculate the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\tau { - }\gamma$$\end{document} relationship of the joint. The MCFT was proposed according to the shear plate test carried out by Vicchio and Collins3. The force transferred by longitudinal beam and column bars has been considered in MCFT, in other words, only the participation of truss mechanism has been taken into account. Xue et al.4 proposed to use the diagonal compression strut model to \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\tau { - }\gamma$$\end{document} relationship of steel fiber RC beam–column joint. However, this model presumes that the joint shear force is transmitted through the diagonal compression strut, and the confined mechanism is considered by confined concrete model, the participation of the truss mechanism is not taken into account.Fig. 1Force mechanisms within beam–column joint core. Force mechanisms within beam–column joint core. In the process of designing structural concrete members, it is a standard practice to conceptually divide these elements into two zones for better analysis. One effective method for representing how loads are transferred in the more complex zones, known as disturbed regions (D-regions), is through the strut-and-tie model. STM simplifies the intricate patterns of force flow within a structure, modeling them as a network of compression elements called struts, tension elements referred to as ties, and their interconnecting points or nodes. This simplified representation, akin to a truss, provides a clear and discretized depiction of the paths along which forces are transferred within the actual structure. The origins of the STM can be traced back to the foundational proposals by Ritter5 and Morsch6. This method has been the subject of extensive research and refinement, especially after the detailed studies by Schlaich et al.7,8. Building on this foundational work, a significant number of researchers have since contributed a variety of STM types9–18. These methods have gained widespread acceptance and have been integrated into contemporary design codes19–24. Although the STM is extensively recognized and valued as a logical technique for the design of D-regions, research on STMs specific to beam-column joints is sparse, with only a handful of studies undertaken by researchers such as Hwang and Lee25, You et al.26, Anwar and Ismail27, Zhang et al.28, Saad and Sakr29, Xia et al.30, Mashhadi and Dastan31. Nevertheless, the main focus of these models is to estimate the joint shear strength, the calculation of the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\tau { - }\gamma$$\end{document} relationship has been ignored. Compared with the calculation of the shear capacity of the joint, the calculation of the shear stress–strain relationship of the joint involves the whole process analysis of the force and deformation of the joint, and the calculation process is more complex. The existing STM is not equipped to directly predict the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\tau { - }\gamma$$\end{document} relationship. Beam-column joints often experience significant shear deformation under seismic loading, whose impact on the overall structural performance cannot be overlooked. Thus, it is imperative to develop a new STM capable of simultaneously predicting both joint shear strength and joint shear deformation (i.e., the joint shear stress–strain relationship). Therefore, in this research a STM, named as Crack–based STM, was proposed according to the deformation characteristics and the crack patterns in joint region. The calculation method of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\tau { - }\gamma$$\end{document} envelope curve was established based on the Crack–based STM. The establishment, calibration, and validation of the Crack–based STM were carried out focusing on the interior beam–column joint with stirrups. The Crack–based STM was verified by twenty–three available interior joint experimental results. The numerical analysis was conducted using the OpenSees software. The calculation results of the crack-based STM were applied to improve the parameter determination method of critical parameters of the Beam Column Joint Element in OpenSees software. The load–displacement hysteresis curves and joint shear deformations obtained from the finite element analysis were compared against experimental results of beam–column joints to validate the applicability of the Crack–based STM. Description of the joint specimens and main experimental results Six full-scale beam-column joint specimens were designed to simulate typical interior joints in RC frame structures32. The design of the specimens conformed to the strong column-weak beam principle and the comprehensive guidelines specified in the Chinese Code33. The geometry, dimensions, and reinforcement detailing are depicted in Fig. 2. The mechanical properties of concrete and reinforcements are recorded in Table 1, and the main design parameters of beam-column joint specimens are recorded in Table 2.Fig. 2Geometry, dimensions, and reinforcement detailing of test specimens32.Table 1Material properties of concrete and steels32.SpecimenCompressive Strength fcu150, MPaBeam and column longitudinal reinforcementJoint transverse reinforcementDiameter, mmYield strength, MPaDiameter, mmYield strength, MPaCL-135.51652812407CL-238.22054012368CL-334.31644412464CL-430.21655312362CL-537.91644412457CL-638.71643212442Table 2Main design parameters of beam-column joint specimens.SpecimensJoint horizontal transverse reinforcement characteristic value (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\lambda_{{\text{v}}}$$\end{document})Joint shear index (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\nu$$\end{document})Column depth to longitudinal bar diameter ratio (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$h_{{\text{c}}} /d_{{\text{b}}}$$\end{document})CL-10.1710.27325CL-20.1610.3120CL-30.1540.23428.1CL-40.2010.3325CL-50.1370.21128.1CL-60.1240.19434.4Joint horizontal transverse reinforcement characteristic value \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\lambda_{{\text{v}}} = \rho_{{\text{j}}} \frac{{f_{{\text{y}}} }}{{f_{{\text{c}}} }} = \frac{{A_{{{\text{st}}}} f_{{\text{y}}} }}{{s_{{\text{j}}} b_{{\text{j}}} f_{{\text{c}}} }}$$\end{document}, joint shear index \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\nu = \left( {\frac{{A_{{{\text{sb}}}} f_{{\text{y}}} }}{{f_{{\text{c}}} b_{{\text{j}}} h_{{\text{c}}} }}} \right)\left( {1 - 0.85\frac{{h_{{\text{b}}} }}{{H - h_{{\text{b}}} }}} \right)$$\end{document}. The meanings of symbols in each formula are shown in Nomenclature. Geometry, dimensions, and reinforcement detailing of test specimens32. Table 1Material properties of concrete and steels32.SpecimenCompressive Strength fcu150, MPaBeam and column longitudinal reinforcementJoint transverse reinforcementDiameter, mmYield strength, MPaDiameter, mmYield strength, MPaCL-135.51652812407CL-238.22054012368CL-334.31644412464CL-430.21655312362CL-537.91644412457CL-638.71643212442 Material properties of concrete and steels32. Table 2Main design parameters of beam-column joint specimens.SpecimensJoint horizontal transverse reinforcement characteristic value (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\lambda_{{\text{v}}}$$\end{document})Joint shear index (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\nu$$\end{document})Column depth to longitudinal bar diameter ratio (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$h_{{\text{c}}} /d_{{\text{b}}}$$\end{document})CL-10.1710.27325CL-20.1610.3120CL-30.1540.23428.1CL-40.2010.3325CL-50.1370.21128.1CL-60.1240.19434.4Joint horizontal transverse reinforcement characteristic value \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\lambda_{{\text{v}}} = \rho_{{\text{j}}} \frac{{f_{{\text{y}}} }}{{f_{{\text{c}}} }} = \frac{{A_{{{\text{st}}}} f_{{\text{y}}} }}{{s_{{\text{j}}} b_{{\text{j}}} f_{{\text{c}}} }}$$\end{document}, joint shear index \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\nu = \left( {\frac{{A_{{{\text{sb}}}} f_{{\text{y}}} }}{{f_{{\text{c}}} b_{{\text{j}}} h_{{\text{c}}} }}} \right)\left( {1 - 0.85\frac{{h_{{\text{b}}} }}{{H - h_{{\text{b}}} }}} \right)$$\end{document}. The meanings of symbols in each formula are shown in Nomenclature. Main design parameters of beam-column joint specimens. Joint horizontal transverse reinforcement characteristic value \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\lambda_{{\text{v}}} = \rho_{{\text{j}}} \frac{{f_{{\text{y}}} }}{{f_{{\text{c}}} }} = \frac{{A_{{{\text{st}}}} f_{{\text{y}}} }}{{s_{{\text{j}}} b_{{\text{j}}} f_{{\text{c}}} }}$$\end{document}, joint shear index \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\nu = \left( {\frac{{A_{{{\text{sb}}}} f_{{\text{y}}} }}{{f_{{\text{c}}} b_{{\text{j}}} h_{{\text{c}}} }}} \right)\left( {1 - 0.85\frac{{h_{{\text{b}}} }}{{H - h_{{\text{b}}} }}} \right)$$\end{document}. The meanings of symbols in each formula are shown in Nomenclature. For each specimen, the bottom of column was supported by a hinge, while the left and right ends of beams were supported by rollers. The load imposed on the column top was recorded with a load cell connected in series with an actuator that applied reversed cyclic lateral load, as illustrated in Fig. 3a and b. The experiment proceeded using a displacement-control strategy, incrementally increasing the story drift ratio. The displacement history of specimens is depicted in Fig. 4. Additionally, the columns were subjected to a constant axial load, with the axial load ratios for specimens CL-1, 2, 3, 4, 5, and 6 being 0.25, 0.3, 0.25, 0.25, 0.25, and 0.25, respectively.Fig. 3Test setup and instrumentation32 (Image in (a) and (c) were created using AutoCAD version 2018; https://www.autodesk.com/. Image in (b) was acquired with a Canon EOS 600D camera and cropped using Microsoft PowerPoint version 2508; https://www.microsoft.com/.)Fig. 4Loading history32. Test setup and instrumentation32 (Image in (a) and (c) were created using AutoCAD version 2018; https://www.autodesk.com/. Image in (b) was acquired with a Canon EOS 600D camera and cropped using Microsoft PowerPoint version 2508; https://www.microsoft.com/.) Loading history32. Strain measurements of the reinforcements were conducted using strain gauges with a length of 3 mm, which were positioned on the longitudinal beam and column reinforcements, as well as on the joint stirrups. To gauge the joint shear deformations and the slippage of the beam longitudinal bars, dial indicators were set up on both the joint core and on the longitudinal reinforcements within the plastic-hinge zone of the beam. The arrangement for these external measuring devices was presented in Fig. 3c. Figure 5 illustrates the load–displacement hysteresis responses for each specimen32. Table 3 details the peak lateral load (Fmax) and the corresponding drift ratio (∆Fmax) for each sample, as well as the strength degradation at a drift ratio of 3.5% (∆F3.5%). Additionally, Table 3 includes the drift ratio (DR) at which diagonal cracking first appeared in the joint core. The patterns of cracking in the beam-column joint specimens are depicted in Fig. 632. The distribution of damage was similar in these six specimens. Initial diagonal cracking on the surface of the joint core was detected at a DR of 0.75%, which became more pronounced and expansive with repeated iterations of the loading cycles. The bottom beam reinforcement yielded at a DR approaching 1.35%. The specimens demonstrated an increase in lateral load resistance in tandem with increasing lateral drift, achieving a maximum load when the average DR reached 2%. Thereafter, the joint area started to exhibit crushing and spalling. At the same time, the specimens showed a reduction in strength and stiffness due to the bond deterioration of longitudinal beam bars, the expansion of joint diagonal cracks, and the consequent degradation of joint shear strength. In the phase of larger displacement cycles, the concrete core steadily spalled from the joint area, revealing the reinforcing bars. By the end of the testing procedures, as Fig. 6 indicates, the joint panels experienced substantial deformation and were extensively damaged.Fig. 5Hysteretic responses of the applied load–displacement diagrams of specimens32.Table 3Summary of test results.SpecimenPositive directionNegative directionDrift at first joint crack (%)Fmax(kN)∆Fmax(%)∆F3.5%(%)Fmax(kN)∆Fmax(%)∆F3.5%(%)CL-1175.32.2018.4− 191.2− 2.6023.00.75CL-2182.41.7520.8− 159.2− 1.7519.30.75CL-3147.21.3515.1− 158.6− 1.7521.80.75CL-4185.82.2020.7− 184.9− 2.2018.00.75CL-5126.92.2018.2− 133.9− 2.2015.50.75CL-6121.82.2025.8− 119.8− 1.7527.70.75The assumption is that the failure occurs when the lateral load drops to 85% of Fmax. For specimen CL-6, the failure occurs at a drift ratio of 3.0%, thus the ∆F3.5% for CL-6 is related to the strength degradation at a drift ratio of 3.0%Fig. 6The crack pattern of joint specimens at the end of the test (The photographic images were acquired with a Canon EOS 600D camera. The figures were cropped using Microsoft PowerPoint version 2508; https://www.microsoft.com/). Hysteretic responses of the applied load–displacement diagrams of specimens32. Table 3Summary of test results.SpecimenPositive directionNegative directionDrift at first joint crack (%)Fmax(kN)∆Fmax(%)∆F3.5%(%)Fmax(kN)∆Fmax(%)∆F3.5%(%)CL-1175.32.2018.4− 191.2− 2.6023.00.75CL-2182.41.7520.8− 159.2− 1.7519.30.75CL-3147.21.3515.1− 158.6− 1.7521.80.75CL-4185.82.2020.7− 184.9− 2.2018.00.75CL-5126.92.2018.2− 133.9− 2.2015.50.75CL-6121.82.2025.8− 119.8− 1.7527.70.75The assumption is that the failure occurs when the lateral load drops to 85% of Fmax. For specimen CL-6, the failure occurs at a drift ratio of 3.0%, thus the ∆F3.5% for CL-6 is related to the strength degradation at a drift ratio of 3.0% Summary of test results. The assumption is that the failure occurs when the lateral load drops to 85% of Fmax. For specimen CL-6, the failure occurs at a drift ratio of 3.0%, thus the ∆F3.5% for CL-6 is related to the strength degradation at a drift ratio of 3.0% The crack pattern of joint specimens at the end of the test (The photographic images were acquired with a Canon EOS 600D camera. The figures were cropped using Microsoft PowerPoint version 2508; https://www.microsoft.com/). The failure modes of specimens at the end of the test are illustrated in Fig. 7. The six beam-column subassemblage specimens in this study exhibited two failure modes: CL-6 failed in beam-end flexural failure (see in Fig. 7f), and the other five specimens failed in joint shear failure after longitudinal reinforcement yielding (see in Fig. 7a–e).Fig. 7Failure modes of specimens at the end of the test. Failure modes of specimens at the end of the test. Under cyclic loading, the diagonal compression force in the joint core was resisted by both the truss mechanism and diagonal compression strut mechanism, whereas‌ the diagonal tension force was solely carried by‌ the truss mechanism. As a result, the total diagonal compression in the joint core consistently exceeded the total diagonal tension throughout the loading process. With repeated loading cycles, the longitudinal beam reinforcement yielded. After yielding, the joint experienced severe bond degradation and yield penetration. As‌ longitudinal beam bar slip increased, the shear force transferred to the truss mechanism via ‌bonded reinforcement decreased significantly. Consequently, the ‌diagonal compression strut‌ carried a greater proportion of shear resistance, ultimately leading to ‌shear failure in the joint region‌. The beam-end flexural failure in CL-6 primarily resulted from the relatively weak beam flexural capacity of this specimen. After yielding of the longitudinal reinforcement, with repeated loading cycles, concrete crushing occurred at the beam-end compression zone with a reduction in effective compression zone height of beam. This led to a significant decrease in shear force transferred to the diagonal compression strut. At this stage, the two mechanisms effectively resisted the joint shear force. Ultimately, due to severe beam flexural failure, the bearing capacity of specimen CL-2 decreased significantly. The failure mode of specimen CL-6 at the end of the test is shown in Fig. 7f. The joint region of CL-6 exhibited relatively minor damage. As depicted in Fig. 3c, dial indicators I and II were employed for measuring the diagonal displacements within the joint. The overall joint shear strain is traditionally calculated from these measurements as described in reference34,1\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\gamma = \gamma_{{\text{b}}} + \gamma_{{\text{c}}} = \frac{{\sqrt {h_{{\text{b}}}^{2} + h_{{\text{c}}}^{2} } }}{{2h_{{\text{b}}} h_{{\text{c}}} }}(\Delta_{\rm I} - \Delta_{{{\rm I}{\rm I}}} )$$\end{document}where \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\gamma_{{\text{b}}}$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\gamma_{c}$$\end{document} are the beam and column side joint shear strain, respectively, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$h_{{\text{b}}}$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$h_{{\text{c}}}$$\end{document} are height and width of the joint region, respectively, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\Delta_{\rm I}$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\Delta_{{{\rm I}{\rm I}}}$$\end{document} are deformations measured by dial indicator I and II , respectively (positive corresponds to extension and negative to compression). The joint shear stress can be determined by Eq. (2).2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\tau = \frac{{V_{{{\text{jh}}}} }}{{b_{{\text{j}}} h_{{\text{c}}} }}{ = }\frac{{(M_{{\text{L}}} + M_{{\text{R}}} )/jh_{{\text{b}}} - V_{{\text{c}}} }}{{b_{{\text{j}}} h_{{\text{c}}} }}$$\end{document}where \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$jh_{{\text{b}}}$$\end{document} is the distance between the tension and compression resultants in the beam section at the beam-joint interface, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$b_{{\text{j}}}$$\end{document} is effective joint width19, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{L}$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{{\text{R}}}$$\end{document} are moments at the interface on joint left and right side, respectively. Figure 8 presents the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\tau - \gamma$$\end{document}relationship for the specimens. The joint shear stress showed a direct proportionality to the lateral force. After the peak load was achieved at a DR of 2.2%, there was an observable increase in joint shear deformation, which led to a considerable degradation on both strength and stiffness. At the final cycle of loading, the average joint shear strain reached 0.03 rad for all specimens except CL-6, where beam-end flexural failure occurred. Due to its beam-end flexural failure mode, specimen CL-6 exhibited significantly lower joint shear deformation compared to the other five specimens with joint shear failure mode. For specimens that experienced shear failure, the joint distortion was affected by the joint horizontal transverse reinforcement characteristic value. CL-5 with the lowest joint horizontal transverse reinforcement characteristic value exhibited the highest joint shear strain at joint failure condition, whereas CL-4 with the highest joint horizontal transverse reinforcement characteristic value exhibited the lowest joint shear strain at joint failure condition.Fig. 8Joint shear stress versus strain hysteretic response. Joint shear stress versus strain hysteretic response. Comprehensive details on the analysis of the experimental results of specimens CL-1, CL-2, CL-3, and CL-4 are available in32. Crack–based strut–and–tie model Model introduction and basic assumptions Figure 6 depicts the typical crack pattern observed in beam–column joint tests. Based on the deformation characteristics of the joint and the force transfer mechanisms in the joint, a Crack–based STM has been established. Figure 9 shows the proposed Crack–based STM for interior beam–column joint. The tension ties (Fig. 9), which represent the horizontal joint stirrups and longitudinal column bars, are developed to incorporate the influence of the tie rod of truss mechanism. At the same time, the compression struts are developed to take the impact of diagonal struts into account. The main diagonal compression strut AE is developed to account for the influence of diagonal strut mechanism and a portion of struts of truss mechanism. The subdiagonal compression struts (AD, BD, HE, and HF) are established to account for the influence of the remaining part of struts of truss mechanism. In addition, in the proposed Crack–based STM, the confined mechanism is considered by the Mander confined concrete model35.Fig. 9Crack–based STM. Crack–based STM. The Crack–based STM shown in Fig. 9 is an idealized model which employs the following assumptions.As shown in Fig. 7, the concrete cover experienced extensive spalling during the experimental study, thus the deformation of the concrete cover was neglected.The compression struts and tension ties are assumed to remain straight when the joint shear deformation occurs. According to the measured strain of the steel bars in the experimental study, it can be concluded that in the joint area, the overall strength of the longitudinal bars in the column is markedly greater than the strength of the horizontal joint stirrups32. Therefore, it is assumed that the axial deformation can only occur in the tension tie HD which represents the horizontal joint stirrups.The nodes in this model are ideal hinges.According to the geometric characteristics of the Crack–based STM, it can be inferred that the total height of the cross-sectional height of each compression strut is equal to the projection length of the diagonal length of the joint core on the plane perpendicular to another diagonal.The horizontal joint shear force is an aggregate of the forces from each compression strut in horizontal direction, then the joint shear stress can be performed as follow29:3\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\tau_{{\text{j}}} = \frac{{\sum {F_{{{\text{n}}i}} \cdot \cos \alpha_{i} } }}{{b_{{\text{j}}} \cdot h_{{\text{c}}} }}$$\end{document}where \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\tau_{{\text{j}}}$$\end{document} is the joint shear stress computed based on the Crack–based STM, Fni is the axial force of the compression strut with serial number i, αi is the included angle between the compression strut with serial number i and the horizontal direction. As shown in Fig. 7, the concrete cover experienced extensive spalling during the experimental study, thus the deformation of the concrete cover was neglected. As shown in Fig. 7, the concrete cover experienced extensive spalling during the experimental study, thus the deformation of the concrete cover was neglected. The compression struts and tension ties are assumed to remain straight when the joint shear deformation occurs. According to the measured strain of the steel bars in the experimental study, it can be concluded that in the joint area, the overall strength of the longitudinal bars in the column is markedly greater than the strength of the horizontal joint stirrups32. Therefore, it is assumed that the axial deformation can only occur in the tension tie HD which represents the horizontal joint stirrups. The compression struts and tension ties are assumed to remain straight when the joint shear deformation occurs. According to the measured strain of the steel bars in the experimental study, it can be concluded that in the joint area, the overall strength of the longitudinal bars in the column is markedly greater than the strength of the horizontal joint stirrups32. Therefore, it is assumed that the axial deformation can only occur in the tension tie HD which represents the horizontal joint stirrups. The nodes in this model are ideal hinges. The nodes in this model are ideal hinges. According to the geometric characteristics of the Crack–based STM, it can be inferred that the total height of the cross-sectional height of each compression strut is equal to the projection length of the diagonal length of the joint core on the plane perpendicular to another diagonal. According to the geometric characteristics of the Crack–based STM, it can be inferred that the total height of the cross-sectional height of each compression strut is equal to the projection length of the diagonal length of the joint core on the plane perpendicular to another diagonal. The horizontal joint shear force is an aggregate of the forces from each compression strut in horizontal direction, then the joint shear stress can be performed as follow29:3\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\tau_{{\text{j}}} = \frac{{\sum {F_{{{\text{n}}i}} \cdot \cos \alpha_{i} } }}{{b_{{\text{j}}} \cdot h_{{\text{c}}} }}$$\end{document} The horizontal joint shear force is an aggregate of the forces from each compression strut in horizontal direction, then the joint shear stress can be performed as follow29:3\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\tau_{{\text{j}}} = \frac{{\sum {F_{{{\text{n}}i}} \cdot \cos \alpha_{i} } }}{{b_{{\text{j}}} \cdot h_{{\text{c}}} }}$$\end{document} (6)Based on the measured joint deformation data from this study (as shown in Fig. 6) and the existing research findings1, it can be concluded that after shear deformation occurs, the shape of the joint core region can be approximated as a parallelogram. Thus, the average joint shear strain is assumed equal to the rotation angle of the middle horizontal tension tie HD. The average joint shear strain can be calculated as:4\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\gamma_{{\text{j}}} = \theta_{{{\text{VA}}}}$$\end{document}where \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\gamma_{{\text{j}}}$$\end{document} is the joint shear strain computed by Crack–based STM, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\theta_{{{\text{VA}}}}$$\end{document} is the rotation angle of the tension tie HD. (6)Based on the measured joint deformation data from this study (as shown in Fig. 6) and the existing research findings1, it can be concluded that after shear deformation occurs, the shape of the joint core region can be approximated as a parallelogram. Thus, the average joint shear strain is assumed equal to the rotation angle of the middle horizontal tension tie HD. The average joint shear strain can be calculated as: Based on the measured joint deformation data from this study (as shown in Fig. 6) and the existing research findings1, it can be concluded that after shear deformation occurs, the shape of the joint core region can be approximated as a parallelogram. Thus, the average joint shear strain is assumed equal to the rotation angle of the middle horizontal tension tie HD. The average joint shear strain can be calculated as: (7)As shown in Fig. 7, during the experimental process, the crack propagation and damage development in the joint region exhibited approximately diagonal symmetry. Therefore, it is assumed that the cross–sectional areas of subdiagonal compression struts AD and HE are the same, and similarly, the cross–sectional areas of subdiagonal compression struts BD and HF are equal.(8)According to the experimental results, during the loading process, the joint transverse reinforcements primarily serve to transfer and resist joint shear forces. Therefore, the bond effect between stirrup and concrete has been ignored. (7)As shown in Fig. 7, during the experimental process, the crack propagation and damage development in the joint region exhibited approximately diagonal symmetry. Therefore, it is assumed that the cross–sectional areas of subdiagonal compression struts AD and HE are the same, and similarly, the cross–sectional areas of subdiagonal compression struts BD and HF are equal. As shown in Fig. 7, during the experimental process, the crack propagation and damage development in the joint region exhibited approximately diagonal symmetry. Therefore, it is assumed that the cross–sectional areas of subdiagonal compression struts AD and HE are the same, and similarly, the cross–sectional areas of subdiagonal compression struts BD and HF are equal. (8)According to the experimental results, during the loading process, the joint transverse reinforcements primarily serve to transfer and resist joint shear forces. Therefore, the bond effect between stirrup and concrete has been ignored. According to the experimental results, during the loading process, the joint transverse reinforcements primarily serve to transfer and resist joint shear forces. Therefore, the bond effect between stirrup and concrete has been ignored. Compression strut Main diagonal compression strut AE In the Crack–based STM, the main strut AE is defined as a prismatic bar. The cross–sectional area of the main strut AE is able to be calculated using the following method. Cross–sectional height of main strut AE, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$h_{{{\text{strut}}}}$$\end{document}. Cross–sectional height of main strut AE, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$h_{{{\text{strut}}}}$$\end{document}. Cross–sectional height of main strut AE, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$h_{{{\text{strut}}}}$$\end{document}. Based on the suggestions of Hwang and Lee25 and Wang36, the height of the main strut AE can be determined using the empirical equations presented as Eq. (5). Figure 10 depicts a conceptual representation of the main diagonal compression strut in Crack–based STM.5\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$h_{{{\text{strut}}}} = h_{{{\text{strut1}}}} \cdot \sin \beta_{1} { = }\left( {\sqrt {c_{{\text{c}}}^{{2}} + c_{{\text{b}}}^{{2}} } } \right) \cdot \sin \beta_{1}$$\end{document}6\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$c_{{\text{c}}} = \left( {0.25 + \frac{{0.85N_{{\text{c}}} }}{{A_{{\text{c}}} f_{{\text{c}}} }}} \right)h_{{\text{c}}}$$\end{document}7\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$c_{{\text{b}}} = \frac{{A_{{{\text{sb}}}} f_{{\text{y}}} }}{{0.85b_{{\text{b}}} f_{{\text{c}}} }}$$\end{document}where \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$c_{{\text{c}}}$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$c_{{\text{b}}}$$\end{document} are depth of column and beam compression zones, respectively, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$f_{{\text{c}}}$$\end{document} is the concrete compressive strength, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$A_{{{\text{sb}}}}$$\end{document} is the cross section area of beam tensile steel reinforcement, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$b_{{\text{b}}} {\kern 1pt} {\kern 1pt} {\kern 1pt}$$\end{document} is beam width, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$f_{{\text{y}}}$$\end{document} is steel yield strength, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$N_{{\text{c}}} {\kern 1pt} {\kern 1pt} {\kern 1pt}$$\end{document} is joint axial load, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$A_{{\text{c}}}$$\end{document} is the cross section area of column, and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\beta_{{1}} {\kern 1pt} {\kern 1pt} {\kern 1pt}$$\end{document} is the included angle between \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$h_{{{\text{strut1}}}} {\kern 1pt} {\kern 1pt} {\kern 1pt}$$\end{document} and diagonal of the joint area (Fig. 11), \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$h_{{{\text{strut1}}}}$$\end{document} is able to be determined as \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\sqrt {c_{{\text{c}}}^{2} + c_{{\text{b}}}^{2} }$$\end{document}.Fig. 10Diagonal strut model.Fig. 11Inclined angle of cross–section of strut. Diagonal strut model. Inclined angle of cross–section of strut. (2)Cross–sectional width of main strut AE, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$w_{{{\text{strut}}}}$$\end{document}. (2)Cross–sectional width of main strut AE, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$w_{{{\text{strut}}}}$$\end{document}. Cross–sectional width of main strut AE, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$w_{{{\text{strut}}}}$$\end{document}. The cross–sectional width of the main strut AE could be assessed using the following approach36:8\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$w_{{{\text{strut}}}} = b_{{\text{j}}} - 2 \times (c + d_{{{\text{jh}}}} )$$\end{document}where c is the thickness of concrete cover, djh is the diameter of joint horizontal transverse reinforcement. The stress–strain relationship of the compression strut is characterized using the Mander confined concrete model35. As suggested by ACI 318–1919, the compression force of the main strut AE (FnAE) can be computed as follow:9\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$F_{{{\text{nAE}}}} = 0.85 \cdot \beta_{{\text{s}}} \cdot \sigma_{{\text{c}}} \cdot w_{{{\text{strut}}}} \cdot h_{{{\text{strut}}}}$$\end{document}where σc is the compression stress of the strut, and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\beta_{{\text{s}}}$$\end{document} is the softening coefficient of the compression stress of the strut, and the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\beta_{{\text{s}}}$$\end{document} is suggested to be defined as 0.75. Subdiagonal compression strut he cross–sectional width of each subdiagonal strut (AD, HE, BD and HF) is equal to that of the main strut (AE). As suggested by Wang36, the calculation procedure for the cross–sectional height of each substrut is established as follows: Calculate the diagonal length of the joint core, lstrut.According to assumption (5) mentioned above, the total height of diagonal compression struts (hz) can be determined as Eq. (10):10\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$h_{{\text{z}}} = l_{{{\text{strut}}}} \cdot \sin \beta_{2}$$\end{document}Calculate the total height of the four substruts, hd.11\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$h_{{\text{d}}} = h_{{\text{z}}} - h_{{{\text{strut}}}}$$\end{document}It is suggested that the cross-sectional area ratio between strut AD and strut BD be 5:4, then the height of strut AD and strut BD can be calculated according to the assumption (5) and (7). Calculate the diagonal length of the joint core, lstrut. Calculate the diagonal length of the joint core, lstrut. According to assumption (5) mentioned above, the total height of diagonal compression struts (hz) can be determined as Eq. (10):10\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$h_{{\text{z}}} = l_{{{\text{strut}}}} \cdot \sin \beta_{2}$$\end{document} According to assumption (5) mentioned above, the total height of diagonal compression struts (hz) can be determined as Eq. (10):10\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$h_{{\text{z}}} = l_{{{\text{strut}}}} \cdot \sin \beta_{2}$$\end{document} Calculate the total height of the four substruts, hd.11\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$h_{{\text{d}}} = h_{{\text{z}}} - h_{{{\text{strut}}}}$$\end{document} Calculate the total height of the four substruts, hd.11\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$h_{{\text{d}}} = h_{{\text{z}}} - h_{{{\text{strut}}}}$$\end{document} It is suggested that the cross-sectional area ratio between strut AD and strut BD be 5:4, then the height of strut AD and strut BD can be calculated according to the assumption (5) and (7). It is suggested that the cross-sectional area ratio between strut AD and strut BD be 5:4, then the height of strut AD and strut BD can be calculated according to the assumption (5) and (7). 12\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$h_{{{\text{AD}}}} = h_{{{\text{HE}}}} = \frac{1}{2} \cdot \frac{5}{9} \cdot h_{{\text{d}}} \cdot \sin \beta_{{3}}$$\end{document}13\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$h_{{{\text{BD}}}} = h_{{{\text{HF}}}} = \frac{1}{2} \cdot \frac{4}{9} \cdot h_{{\text{d}}} \cdot \sin \beta_{{4}}$$\end{document}where hAD, hHE, hBD, and hHF represent the cross–sectional heights of struts AD, HE, BD, and HF, respectively. As shown in Fig. 11, β2, β3 and β4 denote the inclined angles between the diagonal and struts AE, HE, and HF, respectively. Tension ties In Fig. 9, the vertical tension tie AG and CE are composed of the column longitudinal reinforcements arranged on the left and right sides of the column, along with a portion of the concrete concentric with these reinforcements in the axial direction. As for tension tie BF, when the column longitudinal reinforcements are arranged on the front and back of the joint, the vertical tension tie BF is composed of these reinforcements and the surrounding axially concentric concrete. When there is no reinforcement on the front and back of the joint, the top and bottom columns have binding force on the joint, which is equivalent to the vertical tension tie BF. The upper, middle, and lower horizontal tension ties (AC, HD, and GE) are composed of joint transverse reinforcements and a part of the surrounding concrete concentric with the axial direction of the transverse reinforcements. Test results of Beam–column subassemblages indicate that the stresses of joint transverse reinforcements are non–uniformity, and only a part of joint transverse reinforcements yield. Simultaneously, the strains of transverse reinforcements in the middle of the joint are generally higher than that of the transverse reinforcements at the joint upper and lower sides. According to the research on the effective transverse reinforcement layers within the joint area25, it is assumed that 65% of the joint transverse reinforcements yield37. Therefore, it is recommended to calculate the cross-section areas of horizontal tension ties HD, AC, and GE as follows:14\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$A_{{{\text{HD}}}} = 0.65 \cdot m \cdot n \cdot A_{{{\text{jh1}}}}$$\end{document}15\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$A_{{{\text{AC}}}} = A_{{{\text{GE}}}} = \frac{1}{2}(A_{{{\text{jh2}}}} - A_{{{\text{HD}}}} )$$\end{document}where m is the number of joint stirrup legs along the longitudinal beam reinforcement, n is the count of joint transverse reinforcement layers, Ajh1 is the cross–sectional area of single hoop, Ajh2 is the total cross–sectional area of joint transverse reinforcements. The axial force of the horizontal tension tie HD can be obtained from the equilibrium condition of node H. According to ACI31819, the correlation between axial force and stress in the tension tie HD can be determined as:16\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$T_{{{\text{HD}}}} = \beta_{{\text{n}}} \cdot \sigma_{{{\text{HD}}}} \cdot A_{{{\text{HD}}}}$$\end{document}where \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$T_{{{\text{HD}}}}$$\end{document} is the axial force of HD, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\sigma_{{{\text{HD}}}}$$\end{document} is the tension stress of HD, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\beta_{{\text{n}}}$$\end{document} is the softening coefficient of the tension stress of the tie, and the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\beta_{{\text{n}}}$$\end{document} is suggested to be defined as 0.8 according to ACI31819. A uniaxial bilinear steel constitutive model is applied to define the bar stress and strain relationship. The strain–hardening ratio of the uniaxial bilinear steel constitutive model is 1%. Based on assumption (2), the strain of tie HD \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varepsilon_{{{\text{HD}}}}$$\end{document}) can be calculated according to the uniaxial bilinear steel constitutive model. Deformation compatibility and force equilibrium condition of crack–based STM Deformation compatibility of crack–based STM Figure 12 illustrates the deformation condition of the Crack–based STM of beam–column joint. The deformation compatibility formula of Crack–based STM is established based on the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\theta_{{{\text{FED}}}}$$\end{document} (angle of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\angle_{{{\text{FED}}}}$$\end{document}). The positive angle variation is associated with the increase of the angles, and the negative angle variation is associated with the decrease of the angles.Fig. 12Shear deformation of Crack–based STM. Shear deformation of Crack–based STM. According to the deformation coordination principle and assumption (3), the vertical displacement of the horizontal tension tie HD due to rotation is equal to the vertical displacement of the vertical tension tie AG. Thus, the axial compression deformation of the main strut AE and the substrut AD can be determined as follows.17\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\text{d}}l_{{{\text{AE}}}} \approx { - }\theta_{{{\text{VA}}}} \cdot l_{{{\text{HD}}}} \cdot \cos \theta_{{{\text{EAG}}}}$$\end{document}18\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\text{d}}l_{{{\text{AD}}}} \approx { - }\theta_{{{\text{VA}}}} \cdot l_{{{\text{HD}}}} \cdot \cos \theta_{{{\text{DAG}}}}$$\end{document} When the joint shear deformation occurred, the substrut experienced the axial compression. For \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\Delta_{{{\text{FED}}}}$$\end{document}, the geometric relationship can be determined as follow:19\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\cos \theta_{{{\text{BAD}}}} = \frac{{l_{{{\text{AB}}}}^{2} + l_{{{\text{AD}}}}^{2} - l_{{{\text{BD}}}}^{2} }}{{2 \cdot l_{{{\text{AB}}}} \cdot l_{{{\text{AD}}}} }}$$\end{document} Take the derivative on the both sides of Eq. (19), the following equation could be obtained.20\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$2 \cdot l_{{{\text{AB}}}} \cdot {\text{d}}l_{{{\text{AD}}}} \cdot \cos \theta_{{{\text{BAD}}}} + 2 \cdot l_{{{\text{AB}}}} \cdot l_{{{\text{AD}}}} \cdot ( - \sin \theta_{{{\text{BAD}}}} \cdot {\text{d}}\theta_{{{\text{BAD}}}} ) = 2l_{{{\text{AD}}}} \cdot {\text{d}}l_{{{\text{AD}}}} - 2l_{{{\text{BD}}}} \cdot {\text{d}}l_{{{\text{BD}}}}$$\end{document} Thus, the variation of the angle \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\angle_{{{\text{BAD}}}}$$\end{document} can be calculated as follow.21\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\text{d}}\theta_{{{\text{BAD}}}} = \frac{{l_{{{\text{BD}}}} \cdot {\text{d}}l_{{{\text{BD}}}} - l_{{{\text{AD}}}} \cdot {\text{d}}l_{{{\text{AD}}}} + l_{{{\text{AB}}}} \cdot \cos \theta_{{{\text{BAD}}}} \cdot {\text{d}}l_{{{\text{AD}}}} }}{{l_{{{\text{AB}}}} \cdot l_{{{\text{AD}}}} \cdot \sin \theta_{{{\text{BAD}}}} }}$$\end{document} Similarly, the variation of the angles \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\angle_{{{\text{AED}}}} ,\;\angle_{{{\text{FEH}}}} ,\;{\text{and}}\;\angle_{{{\text{AEH}}}}$$\end{document} can also be calculated.22\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\text{d}}\theta_{{{\text{AED}}}} = \frac{{l_{{{\text{AD}}}} \cdot {\text{d}}l_{{{\text{AD}}}} - l_{{{\text{AE}}}} \cdot {\text{d}}l_{{{\text{AE}}}} + l_{{{\text{DE}}}} \cdot \cos \theta_{{{\text{AED}}}} \cdot {\text{d}}l_{{{\text{AE}}}} }}{{l_{{{\text{DE}}}} \cdot l_{{{\text{AE}}}} \cdot \sin \theta_{{{\text{AED}}}} }}$$\end{document}23\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\text{d}}\theta_{{{\text{FEH}}}} = \frac{{l_{{{\text{HF}}}} \cdot {\text{d}}l_{{{\text{HF}}}} - l_{{{\text{HE}}}} \cdot {\text{d}}l_{{{\text{HE}}}} + l_{{{\text{EF}}}} \cdot \cos \theta_{{{\text{FEH}}}} \cdot {\text{d}}l_{{{\text{HE}}}} }}{{l_{{{\text{EF}}}} \cdot l_{{{\text{HE}}}} \cdot \sin \theta_{{{\text{FEH}}}} }}$$\end{document}24\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\text{d}}\theta_{{{\text{AEH}}}} = \frac{{ - l_{{{\text{AE}}}} \cdot {\text{d}}l_{{{\text{AE}}}} - l_{{{\text{HE}}}} \cdot {\text{d}}l_{{{\text{HE}}}} + l_{{{\text{HE}}}} \cdot \cos \theta_{{{\text{AEH}}}} \cdot {\text{d}}l_{{{\text{AE}}}} + l_{{{\text{AE}}}} \cdot \cos \theta_{{{\text{AEH}}}} \cdot {\text{d}}l_{{{\text{HE}}}} }}{{l_{{{\text{AE}}}} \cdot l_{{{\text{HE}}}} \cdot \sin \theta_{{{\text{AEH}}}} }}$$\end{document} The angle of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\angle_{{{\text{FED}}}}$$\end{document} after the deflection of compression struts can be determined geometrically. The angle of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\angle_{{{\text{FED}}}}$$\end{document} is taken as follow.25\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\theta_{{{\text{FED}}}} = \theta_{{{\text{FEH}}}} + \theta_{{{\text{AEH}}}} + \theta_{{{\text{AED}}}} + {\text{d}}\theta_{{{\text{FEH}}}} + {\text{d}}\theta_{{{\text{AEH}}}} + {\text{d}}\theta_{{{\text{AED}}}}$$\end{document} In addition, under assumption (2), considering the axial deformation of the tension tie HD, the angle of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\angle_{{{\text{FED}}}}$$\end{document} can also be calculated as follow.26\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\theta_{{{\text{FED}}}} = \theta_{{{\text{FED}}}} - {\text{d}}\theta_{{{\text{HGF}}}} + \frac{{{\text{d}}l_{{{\text{HD}}}} }}{{l_{{{\text{HG}}}} }}$$\end{document} In order to satisfy the deformation compatibility condition, the Eq. (23) and Eq. (24) should be equal. Therefore, when the calculation results satisfy the formula shown in Eq. (25), the Crack–based STM reaches the deformation compatibility condition.27\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left| {\theta_{FEH} + \theta_{AEH} + \theta_{AED} + d\theta_{FEH} + d\theta_{AEH} + d\theta_{AED} - \left( {\frac{\pi }{2} - d\theta_{HGF} + \frac{{dl_{HD} }}{{l_{HG} }}} \right)} \right| < 0.001$$\end{document} Force equilibrium condition of crack–based STM The force balance of each node determines the force equilibrium condition of Crack–based STM. For the four corner nodes (A, C, E and G), since the external force acting on the model can change with the change of internal force, the four corner nodes satisfy the force balance automatically. The equilibrium conditions of forces at nodes H, B, F, and D are illustrated in Fig. 13. If the compression force exists in the substrut BD, the tension force of tie BF and HD could be calculated according to the equilibrium condition of forces at node H, B and F. Then node D is taken as the calibration node, if the equilibrium condition of forces at node D and the compatibility condition of deformations of each compression struts are satisfied, the Crack–based STM reaches the force equilibrium condition.Fig. 13Force equilibrium condition of each node. Force equilibrium condition of each node. According to Fig. 13a and b, the equilibrium conditions of forces in vertical direction at node B and F are as follows.28\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$T_{{{\text{BF}}}} = F_{{{\text{nBD}}}} \cdot \cos \theta_{{{\text{DBF}}}}$$\end{document}29\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$T_{{{\text{BF}}}} = F_{{{\text{nHF}}}} \cdot \cos \theta_{{{\text{BFH}}}}$$\end{document} In Eq. (28) and (29), \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\theta_{{{\text{DBF}}}} = \theta_{{{\text{BFH}}}}$$\end{document}, thus \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$F_{{{\text{nHF}}}} = F_{{{\text{nBD}}}}$$\end{document}. Similarly, in accordance with the equilibrium condition of forces in vertical direction at node H and D (Fig. 13c and d), it can be concluded that \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$F_{{{\text{nHE}}}}$$\end{document} equals \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$F_{{{\text{nAD}}}}$$\end{document}. Thus, the force equilibrium condition at node D is satisfied automatically when the joint shear deformation occurred in the Crack–based STM. Consequently, if the deformation compatibility condition is satisfied, it indicates that the Crack–based STM reaches the force equilibrium condition. Solution algorithm for joint shear stress and strain relationship based on crack–based STM In the previous sections of this paper, the calculation method of the force and deformation of each strut and tie has been addressed for the Crack–based STM theoretically. The method to develop the Crack–based STM can be verified by predicting the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\tau { - }\gamma$$\end{document} relationship of RC beam–column subassemblages. The step–by–step solution process is presented in the flowchart shown in Fig. 14. The Matlab38 program is used to solve the flowchart.Fig. 14Flowchart showing the solution algorithm for joint shear stress and strain relationship. Flowchart showing the solution algorithm for joint shear stress and strain relationship. Verification of crack–based STM The validation of Crack–based STM is demonstrated by comparison with test results of four RC beam–column joints proposed in this paper and published experimental results of the nineteen RC beam–column joints with respect to \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\tau { - }\gamma$$\end{document} envelope curve. In addition, the computational results of Crack–based STM have also been evaluated against the computational results using diagonal compression strut model proposed by Zhao et al. (named as Zhao–Yang model)39. Details of the published RC beam–column subassemblages can be found in Table 4.Table 4Design information of the joint specimens.SpecimensConcrete strength (fc/MPa)BeamColumnJointLength/mmDimension (bb × hb)Top bars (fy/MPa)Bottom bars (fy/MPa)Stirrups (fy/MPa)Length/mmDimension (bc × hc)Bars in one side of column (fy/MPa)Bars in front/back of column (fy/MPa)Stirrups (fy/MPa)HoopSetsfy/MPaJ–24029.33050400 × 2503–D18 (397)3–D18 (397)D8@100 (298)3520350 × 3504–D22 (363)N/AD8@100 (298)2–D88298J–64023.33050400 × 2504–D18 (397)4–D18 (397)D8@100 (298)3520350 × 3504–D22 (363)N/AD10@100 (298)2–D106382J–94026.63050400 × 2504–D18 (376)2–D18 (376)D8@100 (359)3520350 × 3503–D20 (350)1–D20 (350)D10@100 (359)2–D104382J–104022.33050400 × 2504–D18 (376)2–D18 (376)D8@100 (359)3520350 × 3503–D16 (423)1–D16 (423)D10@100 (359)2–D104382J–134026.43050400 × 2506–D17 (390)4–D17 (390)D8@100 (390)3520350 × 3504–D18 (400)1–D18 (400)D12@100 (390)2–D126370BL–13221.92550500 × 3507–D16 (528)7–D16 (528)D10@80 (361)3520400 × 4005–D16 (528)1–D16 (528)D10@70 (298)2–D128407BL–23228.52700500 × 3006–D20 (540)4–D20 (540)D10@80 (379)3520400 × 4004–D20 (540)1–D20 (540)D10@60 (379)2–D129379BL–33222.62750400 × 2508–D16 (444)4–D14 (444)D8@60 (274)3520450 × 3503–D16 (444)1–D16 (444)D10@100 (274)2–D66465BL–43224.92550500 × 3008–D16 (553)5–D16 (553)D10@80 (375)3520400 × 4004–D20 (540)1–D20 (540)D10@60 (301)2–D129355J3A4128.73000300 × 1756–D12 (480)4–D12 (480)D3@100(298.5)1780350 × 2005–D12 (480)5–D12 (480)D3@100(298.5)4–D64372J3B4123.73000300 × 1756–D12 (480)4–D12 (480)D3@100(298.5)1780350 × 2005–D12 (480)5–D12 (480)D3@100(298.5)2–D104448JE–042272700300 × 18010–D10(387)10–D10 (387)D6@50 (364)1470280 × 3206–D13 (345)2–D13 (345)D6@50 (364)2–D63364S343282700300 × 2005–D16 (470)5–D16 (470)D6@50 (390)1470300 × 3004–D19 (450)2–D19 (450)D6@40 (390)2–D64390PH–104425.63000350 × 2005–D13 (363)5–D13 (363)D6@50 (366)1750300 × 3005–D16 (402)1–D16 (402)D6@50 (366)2–D66366PL–134426.43000350 × 20010–D10 (372)10–D10 (372)D6@50 (366)1750300 × 3006–D16 (402)1–D16 (402)D6@50 (366)2–D68366X24533.652496419 × 2794–D22 (331)3–D19 (345)D3@89 (337)2248362 × 3623–D25 (414)1–D25 (414)D13@89 (352)3–D133352S14638.35760500 × 3504–D25 (452)2–D25 (452)D10@100 (462)2100460 × 5004–D25 (452)2–D25 (452)D13@100 (496)4–D133496No44730.62960300 × 3005–D13 (388)5–D13 (388)D6@50 (354)1470300 × 2005–D13 (388)3–D13 (388)D10@75 (325)2–D61354D = Diameter. Table 4Design information of the joint specimens.SpecimensConcrete strength (fc/MPa)BeamColumnJointLength/mmDimension (bb × hb)Top bars (fy/MPa)Bottom bars (fy/MPa)Stirrups (fy/MPa)Length/mmDimension (bc × hc)Bars in one side of column (fy/MPa)Bars in front/back of column (fy/MPa)Stirrups (fy/MPa)HoopSetsfy/MPaJ–24029.33050400 × 2503–D18 (397)3–D18 (397)D8@100 (298)3520350 × 3504–D22 (363)N/AD8@100 (298)2–D88298J–64023.33050400 × 2504–D18 (397)4–D18 (397)D8@100 (298)3520350 × 3504–D22 (363)N/AD10@100 (298)2–D106382J–94026.63050400 × 2504–D18 (376)2–D18 (376)D8@100 (359)3520350 × 3503–D20 (350)1–D20 (350)D10@100 (359)2–D104382J–104022.33050400 × 2504–D18 (376)2–D18 (376)D8@100 (359)3520350 × 3503–D16 (423)1–D16 (423)D10@100 (359)2–D104382J–134026.43050400 × 2506–D17 (390)4–D17 (390)D8@100 (390)3520350 × 3504–D18 (400)1–D18 (400)D12@100 (390)2–D126370BL–13221.92550500 × 3507–D16 (528)7–D16 (528)D10@80 (361)3520400 × 4005–D16 (528)1–D16 (528)D10@70 (298)2–D128407BL–23228.52700500 × 3006–D20 (540)4–D20 (540)D10@80 (379)3520400 × 4004–D20 (540)1–D20 (540)D10@60 (379)2–D129379BL–33222.62750400 × 2508–D16 (444)4–D14 (444)D8@60 (274)3520450 × 3503–D16 (444)1–D16 (444)D10@100 (274)2–D66465BL–43224.92550500 × 3008–D16 (553)5–D16 (553)D10@80 (375)3520400 × 4004–D20 (540)1–D20 (540)D10@60 (301)2–D129355J3A4128.73000300 × 1756–D12 (480)4–D12 (480)D3@100(298.5)1780350 × 2005–D12 (480)5–D12 (480)D3@100(298.5)4–D64372J3B4123.73000300 × 1756–D12 (480)4–D12 (480)D3@100(298.5)1780350 × 2005–D12 (480)5–D12 (480)D3@100(298.5)2–D104448JE–042272700300 × 18010–D10(387)10–D10 (387)D6@50 (364)1470280 × 3206–D13 (345)2–D13 (345)D6@50 (364)2–D63364S343282700300 × 2005–D16 (470)5–D16 (470)D6@50 (390)1470300 × 3004–D19 (450)2–D19 (450)D6@40 (390)2–D64390PH–104425.63000350 × 2005–D13 (363)5–D13 (363)D6@50 (366)1750300 × 3005–D16 (402)1–D16 (402)D6@50 (366)2–D66366PL–134426.43000350 × 20010–D10 (372)10–D10 (372)D6@50 (366)1750300 × 3006–D16 (402)1–D16 (402)D6@50 (366)2–D68366X24533.652496419 × 2794–D22 (331)3–D19 (345)D3@89 (337)2248362 × 3623–D25 (414)1–D25 (414)D13@89 (352)3–D133352S14638.35760500 × 3504–D25 (452)2–D25 (452)D10@100 (462)2100460 × 5004–D25 (452)2–D25 (452)D13@100 (496)4–D133496No44730.62960300 × 3005–D13 (388)5–D13 (388)D6@50 (354)1470300 × 2005–D13 (388)3–D13 (388)D10@75 (325)2–D61354D = Diameter. Design information of the joint specimens. D = Diameter. Figure 15 depicts the calculated and tested results of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\tau { - }\gamma$$\end{document} envelope curve of parts of specimens. An evaluation is made comparing the experimental data with the results calculated using the Crack–based STM and the Zhao–Yang model. As shown in Fig. 15, the proposed Crack-based STM demonstrates higher accuracy in representing the joint shear deformation behavior of the specimens compared to the Zhao–Yang model. However, compared with the experimental results, the simulation results of the Crack–based STM still exhibit certain discrepancies. The main reason of the discrepancy is that the proposed model was developed based on a series of simplified assumptions, without accounting for the contribution of longitudinal reinforcement bond stress in the joint region to the joint shear deformation behavior. Table 5 shows the comparative analysis of peak joint shear stress between the calculated and test results of all the specimens. The peak joint shear stress predicted by Crack–based STM corresponds well with the experimental data. The MAE between the maximum joint shear stress calculated by the Crack-based STM and the experimental data is 0.4389, with a R2 of 0.7266. The mean value of the maximum joint shear stress determined by the Crack-based STM compared to the experimental data is 0.9987, with a coefficient of variation (CV) of 8.03%. Nevertheless, the MAE between the maximum joint shear stress calculated by the Zhao–Yang model and the experimental data is 2.1887, with a R2 of 0.0521, and the mean value of the maximum joint shear stress determined by the Zhao–Yang model compared to the experimental data is 0.6921, with a CV of 25.37%, indicating a substantial underestimation of joint shear stress.Fig. 15Joint shear stress versus strain envelope curves of calculated results and experimental results.Table 5Comparison between calculated and experimental results.ResearchersSpecimenτjt/MPaτjSTM/MPaτjZYM/MPaτjSTM/τjtτjZYM/τjtThis paperCL–17.00377.12064.73651.01670.6763CL–27.40537.25036.35050.97910.8576CL–36.18036.80116.58241.10031.0651CL–45.91776.03585.35011.01990.9041Fu40J–27.51717.45086.45510.99120.8587J–66.27585.87465.65510.93610.9011J–86.24706.80345.07841.08910.8129J–96.17976.44863.58201.04350.5796J–105.43595.59384.15381.0290.7641J–137.48716.91855.02560.92410.6712Yang et al.32BL–17.45326.89054.84470.92450.6500BL–27.92287.94676.42861.00300.8114BL–37.01076.61494.22990.94350.6034BL–46.81727.00505.30511.02760.7782Supaviriyakit et al.41J3A6.12916.22183.49631.01510.5704J3B6.28976.48904.54581.03170.7227Kusuhara et al.42JE–06.72965.83172.47400.86660.3676Zaid et al.43S310.15949.23474.81490.90900.4739Joh and Goto44PH–105.12985.50643.08031.07340.6005PL–135.75745.57123.16090.96770.5490Durrani45X27.33308.43124.05231.14980.5526Hwang et al.46S15.57396.17694.35791.10820.7818Noguchi and Kurusu47No48.42936.91493.08970.82030.3665MAE–0.43892.1887––R2–0.72660.0521––Average–––0.99870.6921COV–––8.03%25.37%τjt is the tested peak joint shear stress, τjSTM is the peak joint shear stress calculated by Crack–based STM, τjZYM is the peak joint shear stress determined by Zhao–Yang model. Joint shear stress versus strain envelope curves of calculated results and experimental results. Table 5Comparison between calculated and experimental results.ResearchersSpecimenτjt/MPaτjSTM/MPaτjZYM/MPaτjSTM/τjtτjZYM/τjtThis paperCL–17.00377.12064.73651.01670.6763CL–27.40537.25036.35050.97910.8576CL–36.18036.80116.58241.10031.0651CL–45.91776.03585.35011.01990.9041Fu40J–27.51717.45086.45510.99120.8587J–66.27585.87465.65510.93610.9011J–86.24706.80345.07841.08910.8129J–96.17976.44863.58201.04350.5796J–105.43595.59384.15381.0290.7641J–137.48716.91855.02560.92410.6712Yang et al.32BL–17.45326.89054.84470.92450.6500BL–27.92287.94676.42861.00300.8114BL–37.01076.61494.22990.94350.6034BL–46.81727.00505.30511.02760.7782Supaviriyakit et al.41J3A6.12916.22183.49631.01510.5704J3B6.28976.48904.54581.03170.7227Kusuhara et al.42JE–06.72965.83172.47400.86660.3676Zaid et al.43S310.15949.23474.81490.90900.4739Joh and Goto44PH–105.12985.50643.08031.07340.6005PL–135.75745.57123.16090.96770.5490Durrani45X27.33308.43124.05231.14980.5526Hwang et al.46S15.57396.17694.35791.10820.7818Noguchi and Kurusu47No48.42936.91493.08970.82030.3665MAE–0.43892.1887––R2–0.72660.0521––Average–––0.99870.6921COV–––8.03%25.37%τjt is the tested peak joint shear stress, τjSTM is the peak joint shear stress calculated by Crack–based STM, τjZYM is the peak joint shear stress determined by Zhao–Yang model. Comparison between calculated and experimental results. τjt is the tested peak joint shear stress, τjSTM is the peak joint shear stress calculated by Crack–based STM, τjZYM is the peak joint shear stress determined by Zhao–Yang model. It is well known that the mechanical behavior of the beam-column joint region is highly complex, with three force transfer mechanisms existing in the joint region: the diagonal compression strut mechanism, truss mechanism, and confined mechanism, as shown in Fig. 1. The Crack-based STM directly incorporates both the diagonal compression strut mechanism and the truss mechanism by analyzing the interaction between tension ties and compression struts. Additionally, when determining the parameters for the concrete compression struts, the confinement mechanism is indirectly accounted for through Mander confined concrete model. It can be concluded that the proposed model comprehensively accounts for multiple force transfer mechanisms, thereby providing a more rational representation of the stress states of both reinforcement and concrete within the joint region. Consequently, the calculated results of the proposed model show good agreement with the experimental measurements. In contrast, the Zhao-Yang model is exclusively based on the diagonal compression strut mechanism. The effects of both joint horizontal transverse reinforcement and column vertical reinforcement on joint shear transfer are not considered. However, experimental evidence clearly indicates that the reinforcements within joint region significantly influence the joint shear capacity and joint deformation capacity. Consequently, the calculated joint shear stress is underestimated by Zhao-Yang model. Application of crack–based STM Description of the joint finite element analysis model The BeamColumnJoint Element (BCJE) modified by Mitra and Lowes is idealized in detail in Fig. 1648. The BCJE is designed to integrate smoothly with standard 2D beam-column elements. It is equipped with four external nodes that each have three degrees of freedom (DOF) and four internal nodes with one DOF each. This element comprises of three kind of components (Fig. 16). Bar-slip components represent the strength and stiffness degradation attributable to bond slip damage, while interface-shear components represent the reduced shear transfer capacity at the joint perimeter resulting from crack widening. Additionally, the shear-panel component represents the strength and stiffness degradation because of joint shear failure. Constitutive relationships of the components are constructed to ensure that the BCJE can accurately reproduce the inelastic response of the joint. In current application, an elastic load-deformation relationship with a high degree of stiffness is employed to simulate interface-shear response29.Fig. 16BCJE modified by Mitra and Lowes48 (Adobe Photoshop version 12.0.1; https://www.adobe.com/). BCJE modified by Mitra and Lowes48 (Adobe Photoshop version 12.0.1; https://www.adobe.com/). Shear panel response The Crack-based STM proposed above is applied to predict the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\tau { - }\gamma$$\end{document} relationship of the shear panel in BCJE. The Pinching4 material model, depicted in Fig. 17, is suggested for characterizing the shear behavior of the joint48. This model is characterized by a response envelope, an unload-reload path, and a set of three damage rules that shape the trajectory of the response path.Fig. 17Definition of Pinching4 uniaxial material model and damage states48. Definition of Pinching4 uniaxial material model and damage states48. The crucial points on the backbone curves (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{{\text{j}}} ,\theta_{{\text{j}}}$$\end{document}) can be determined by Eq. (30), corresponding to four damage states of the shear panel. The four damage states of shear panel, depicted in Fig. 17, are the initial cracking of the concrete, the yielding of the longitudinal beam reinforcement, the peak joint shear strength, and the shear failure of the joint area. The multi-linear response envelope for the specimens CL–1, CL–2, CL–4, CL-5, and S3, which is calculated using the Crack-based STM, is shown in Fig. 18. The hysteresis rule is characterized utilizing the approach developed by Mitra and Lowes48. 30a\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\theta_{{\text{j}}} { = }\gamma_{{\text{j}}}$$\end{document}30b\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{{\text{j}}} { = }\tau_{{\text{j}}} \cdot h_{{\text{c}}} \cdot h_{{\text{b}}} \cdot b_{{\text{j}}}$$\end{document}where \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\tau_{{\text{j}}}$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\gamma_{{\text{j}}}$$\end{document} are obtained from the joint shear stress versus strain envelope curves calculated by the Crack-based STM. The damage state I is taken as the point where the stiffness of the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\tau { - }\gamma$$\end{document} envelope curve begins to change. The damage state II is taken as the point where the stiffness of the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\tau { - }\gamma$$\end{document} envelope curve significantly decreases. The damage state II is taken as the peak point of the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\tau { - }\gamma$$\end{document} envelope curve. The damage state IV is taken as the point where the joint shear stress decreases to 80% of the peak joint shear stress.Fig. 18\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\tau - \gamma$$\end{document} envelope curves of beam-column joint specimens. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\tau - \gamma$$\end{document} envelope curves of beam-column joint specimens. Bond-slip response The data from the pull-out test of rebars are utilized to define the bar-slip material parameters in the BCJE model. The relationship between bar stress and slip for the longitudinal beam and column reinforcement is calculated using the assumed distribution of bond stress in a yielded anchored bar (Fig. 19)48. The Pinching4 material model, depicted in Fig. 16, is suggested for capturing the pinching characteristics of the bar-slip component48. Additionally, the parameters for hysteresis could be determined according to Lowes and Altoontash1. Fig. 19Bar stress and bond stress distribution of rebar48. Bar stress and bond stress distribution of rebar48. Numerical–experimental comparative analysis for the joint specimens The numerical modeling was performed using the OpenSees framework. The specimens CL–1, CL–2, CL–4, CL-5, and S3 have been simulated. Figure 20 illustrates the numerical analysis model of the beam–column subassemblages. In this model, the beam and column were modeled using the Nonlinear Beam Column Element (NBCE) with three integration points (IP) for the beam and four IP for the column. The cross-sections for beam and column were idealized through Fiber Section (FS). The joint region was represented by BCJE48. Member dimensions, cross-sectional measurements, and reinforcement detailing were specified consistent with the information previously recorded in Table 2 and Table 3. The behavior of concrete in the beam and column was modeled using Concrete01 Material, while the Steel02 Material was employed to simulate the longitudinal reinforcement. Support conditions, pinned support at the bottom of the column and simple supports in the beams, were also consistent with those used in the cyclic tests. Fig. 20Simulated model of beam–column joint subassemblages. Simulated model of beam–column joint subassemblages. Figure 21 depicts the comparison between the simulated load–displacement diagrams and test results of joint specimens. In general terms, the numerical model offers an accurate representation of the experimental results, specifically regarding failure modes, initial stiffness, unloading stiffness, strength degradation, and pinching phenomenon. However, the maximum lateral load is slightly less than that in the experimental response.Fig. 21Comparison of load–displacement responses of analytical results with experimental test results. Comparison of load–displacement responses of analytical results with experimental test results. Figure 22 compares the numerical and experimental joint shear deformation diagrams of specimens. As only the joint shear stress–strain skeleton curve of specimen S3 is reported in reference43, the simulated hysteresis curve for this specimen is compared against its experimental skeleton curve. Figure 22 shows that the numerical model can reproduce the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\tau { - }\gamma$$\end{document} relationship to a certain degree. The numerical values of peak joint shear stress are slightly lower than the test data. The numerical and experimental values of peak joint shear strain of specimen CL-1, specimen CL-2 in negative direction, and specimen CL-3 in positive direction are similar. According to Figs. 21 and 22, the analytical model can capture the general response of joint shear deformation history and the lateral load and displacement history with reasonable accuracy. Although the Crack-based STM demonstrates improved accuracy compared to existing model, it still has certain limitations. This model is specifically developed for interior beam-column joints and cannot be directly applied to calculate the shear performance of exterior beam-column joints, knee joints, and interior top-column joints. Therefore, in subsequent research, this model could be refined to enable its application for analyzing different types of joints. Furthermore, with advancements in computer technology, machine learning is increasingly being applied to engineering calculations. In future studies, machine learning could be utilized to analyze the shear stress–strain relationship of beam-column joint, thereby improving computational accuracy and enabling more precise predictions of joint shear behavior. Fig. 22Comparison of the joint shear deformations of analytical results with experimental test results. Comparison of the joint shear deformations of analytical results with experimental test results. Conclusion The present investigation focuses on the analysis of the shear performance of RC interior beam-column joint. Six full-scale interior beam-column subassemblages were tested and analyzed. A theoretical strut-and-tie model was proposed to predict the joint shear strength and joint shear deformation of beam–column subassemblages. The following main conclusions were obtained:Experimental studies of the beam-column subassemblages have shown that under cyclic loading, high shear stress is concentrated in the beam-column joint region. The measured joint shear deformation increased after reaching the maximum load and was followed by significant degradation in strength and stiffness. At the end of the tests, severe damage was observed in the joint cores. Increasing the joint horizontal transverse reinforcement characteristic value can effectively mitigate the rate of joint strength degradation and reduce the joint shear deformation.Based on the force–deformation characteristics and crack distribution patterns of RC interior beam-column joints, a Crack-based Strut-and-Tie Model was developed. The force equilibrium conditions and deformation compatibility requirements of this model were formulated. A calculation method for the joint shear stress–strain skeleton curve was established based on the Crack-based STM.The joint shear stress–strain curves calculated by the Crack-based STM were compared against published experimental data and those obtained from the Zhao-Yang model. The results indicate that there is good correlation between the joint shear stress calculated by the Crack-based STM and the published experimental data, with a mean value of 0.9987 for the predicted joint shear stress ‌relative to‌ the tested values from 23 RC beam–column joints. Moreover, the MAE between the maximum joint shear stress calculated by the proposed STM and the experimental data is significantly lower than that of the Zhao-Yang model. These findings demonstrate‌ that the Crack-based STM can accurately predict the shear performance of the beam-column joints.The numerical analysis of five beam-column joint specimens was performed based on geometry of joint, material properties of concrete and reinforcing bar and reinforcement detailing on OpenSees platform. In the numerical analysis, the Crack-based STM was used to determine the main parameters of the shear-panel component of the Beam Column Joint Element. The simulated load–displacement relationships and joint shear deformation responses of the beam–column subassemblages agreed fairly well with the experimental results. This proposed STM provides practical tool for obtaining the joint shear strength and deformation relationship of RC beam–column joints. Experimental studies of the beam-column subassemblages have shown that under cyclic loading, high shear stress is concentrated in the beam-column joint region. The measured joint shear deformation increased after reaching the maximum load and was followed by significant degradation in strength and stiffness. At the end of the tests, severe damage was observed in the joint cores. Increasing the joint horizontal transverse reinforcement characteristic value can effectively mitigate the rate of joint strength degradation and reduce the joint shear deformation. Experimental studies of the beam-column subassemblages have shown that under cyclic loading, high shear stress is concentrated in the beam-column joint region. The measured joint shear deformation increased after reaching the maximum load and was followed by significant degradation in strength and stiffness. At the end of the tests, severe damage was observed in the joint cores. Increasing the joint horizontal transverse reinforcement characteristic value can effectively mitigate the rate of joint strength degradation and reduce the joint shear deformation. Based on the force–deformation characteristics and crack distribution patterns of RC interior beam-column joints, a Crack-based Strut-and-Tie Model was developed. The force equilibrium conditions and deformation compatibility requirements of this model were formulated. A calculation method for the joint shear stress–strain skeleton curve was established based on the Crack-based STM. Based on the force–deformation characteristics and crack distribution patterns of RC interior beam-column joints, a Crack-based Strut-and-Tie Model was developed. The force equilibrium conditions and deformation compatibility requirements of this model were formulated. A calculation method for the joint shear stress–strain skeleton curve was established based on the Crack-based STM. The joint shear stress–strain curves calculated by the Crack-based STM were compared against published experimental data and those obtained from the Zhao-Yang model. The results indicate that there is good correlation between the joint shear stress calculated by the Crack-based STM and the published experimental data, with a mean value of 0.9987 for the predicted joint shear stress ‌relative to‌ the tested values from 23 RC beam–column joints. Moreover, the MAE between the maximum joint shear stress calculated by the proposed STM and the experimental data is significantly lower than that of the Zhao-Yang model. These findings demonstrate‌ that the Crack-based STM can accurately predict the shear performance of the beam-column joints. The joint shear stress–strain curves calculated by the Crack-based STM were compared against published experimental data and those obtained from the Zhao-Yang model. The results indicate that there is good correlation between the joint shear stress calculated by the Crack-based STM and the published experimental data, with a mean value of 0.9987 for the predicted joint shear stress ‌relative to‌ the tested values from 23 RC beam–column joints. Moreover, the MAE between the maximum joint shear stress calculated by the proposed STM and the experimental data is significantly lower than that of the Zhao-Yang model. These findings demonstrate‌ that the Crack-based STM can accurately predict the shear performance of the beam-column joints. The numerical analysis of five beam-column joint specimens was performed based on geometry of joint, material properties of concrete and reinforcing bar and reinforcement detailing on OpenSees platform. In the numerical analysis, the Crack-based STM was used to determine the main parameters of the shear-panel component of the Beam Column Joint Element. The simulated load–displacement relationships and joint shear deformation responses of the beam–column subassemblages agreed fairly well with the experimental results. This proposed STM provides practical tool for obtaining the joint shear strength and deformation relationship of RC beam–column joints. The numerical analysis of five beam-column joint specimens was performed based on geometry of joint, material properties of concrete and reinforcing bar and reinforcement detailing on OpenSees platform. In the numerical analysis, the Crack-based STM was used to determine the main parameters of the shear-panel component of the Beam Column Joint Element. The simulated load–displacement relationships and joint shear deformation responses of the beam–column subassemblages agreed fairly well with the experimental results. This proposed STM provides practical tool for obtaining the joint shear strength and deformation relationship of RC beam–column joints.