Seismic behavior and shear capacity calculation of a new type of self-centering steel-concrete composite joint

2.1 Conceptual proposal of novel self-centering joints

To resolve critical limitations in conventional joints – including inadequate energy dissipation, challenging post-earthquake repairs, and safety risks from high-altitude strand tensioning – this study proposes a novel self-centering steel-concrete composite joint (Figure 1). The system integrates precast RC columns with embedded square steel tubes, steel-concrete composite beams, steel beams, and specialized energy-dissipation assemblies comprising replaceable energy-dissipating plates, high strength bolt, and web connection plates with horizontally slotted holes. Unbonded post-tensioned strands provide self-centering capability [21], while high-strength bolts and web connection plates enable efficient dry connections that eliminate in-situ welding and facilitate rapid component replacement after seismic events [22].

Figure 1

Schematic representation of self-centering steel-concrete composite joint.

Anchorage plates are welded to the extremities of the steel beams to facilitate tensioning of the prestressed strands. The composite beams interface with steel beams via high-strength bolts securing both the energy-dissipating plates and web connection plates. Horizontally slotted holes in the web connection plates enhance the rotational capacity of the composite beams. Prestressed strands are anchored at one terminus to the anchorage plate of the steel beam and at the opposite terminus to the corresponding anchorage plate on the composite beam. Energy dissipation is primarily achieved through the replaceable energy-dissipating plates, while self-centering functionality is provided by post-tensioned unbonded strands. Post-earthquake serviceability can be restored through the replacement of damaged energy-dissipating plates.

The replaceable energy-dissipating plates utilize buckling deformation of energy-dissipating steel bars to enhance joint energy dissipation capacity while reducing damage to steel beams and preventing premature yielding of prestressed strands. Compared to core-region running through strand configurations, the proposed single-span tensioning system not only isolates strands from core-joint failure effects but also improves construction efficiency through pre-tensioning, eliminating high-altitude operations.

2.2 Design methodology for novel self-centering joints

To evaluate the seismic influence of energy-dissipating plates and prestressed strands on the novel self-centering steel-concrete composite joint (PSC joint), three precast interior joint specimens were designed:

PSC: with prestressed strands and energy-dissipating plates;

Post-tensioned concrete (PC): with prestressed strands only (no energy-dissipating plates); and

Reinforced self-centering (RSC): with energy-dissipating plates only (no prestressed strands).

Key design parameters are detailed in Table 1.

Reinforcement detailing remains identical across all joints’ precast concrete beams and columns. The specimens differ solely in their connection configurations:

The PC joint omits energy-dissipating plates compared to the PSC joint.

The RSC joint omits prestressed strands compared to the PSC joint.

Using the PSC joint as the reference configuration, detailed dimensions and reinforcement are specified (Figure 2). All reinforcing bars in the joints utilized HRB400 steel (hot-rolled ribbed bars, characteristic yield strength f _y = 400 MPa). The PSC joint configuration employs precast concrete columns with 400 mm × 400 mm cross-sections and prestressed steel-concrete composite beams spanning 1,630 mm on each side. Beam cross-sections measure 250 mm × 400 mm, cast from C40 concrete. Reinforcement configured as twelve 28 mm-diameter longitudinal bars and 12 mm-diameter stirrups spaced at 80 mm within plastic hinge zones and 100 mm elsewhere (4-legged arrangement).

Figure 2

Dimensional and reinforcement details of the PSC joint.

To enhance core-zone shear capacity: stirrups are densely spaced in the joint core region and a 400 mm × 400 mm × 14 mm square steel tube extends 750 mm vertically through the core. Energy-dissipating components utilize Q235 steel, while all other structural steel is Q345 grade. Connections employ M27 Grade 10.9 high-strength bolts passing through 30 mm-diameter holes in precast columns.

The prestressed steel-concrete composite beam system comprises four integrated subsystems:

1. Steel beam: H340 × 280 × 10 × 25 cross-section, 320 mm long. Anchorage plates (20 mm thick) and stiffening ribs are welded to its mid-section, with a 30 mm-thick end plate at the column interface.

2. Composite beam assembly: Total length 1,280 mm, containing: Reinforced concrete beam (1,050 mm long) with six 28 mm-diameter longitudinal bars and 12 mm-diameter 2-legged stirrups spaced at 100 mm. Encased steel beam (400 mm long, H280 × 160 × 10 × 12). End steel beams (210 mm long, H340 × 280 × 10 × 25) welded to 20 mm-thick end plates. Longitudinal steel bars are welded to encased beam flanges over 370 mm.

3. Energy-dissipation system: Four 8 mm-diameter steel bars (60 mm long). Two 20 mm-thick flange connection plates

4. Connection system: 10 mm-thick web connection plates. M22 Grade 10.9 high-strength bolts.

Energy dissipation plates and 10 mm-thick web connection plates connect the steel beams to the steel-concrete composite beams employing Grade 10.9 M22 high-strength bolts and prestressing strands. Four 15.2 mm-diameter 1860-grade prestressing strands (1 × 7 standard type) are symmetrically arranged along the concrete beam length. Each strand is tensioned to 150 kN.

2.3 Loading procedure

The axial compression ratio at each joint is 0.3. The loading method employs column-end displacement control, with the loading point located 200 mm from the top of the column. The loading diagram is shown in Figure 3. The displacement angles for the first three loading stages are 0.25, 0.5, and 0.75%, with each stage cycled once; Subsequent loading displacement angles are 1.0, 1.5, 2%, etc. (i.e., each loading displacement angle increases by 0.5% sequentially), with each level cycled twice. Loading is stopped when any of the following conditions are met: the steel exhibits obvious buckling, the concrete suffers severe damage, or the prestressed steel strands reach their nominal yield strength.

Figure 3

Loading setup schematic.

3.1 Specimen details for validation

Based on experimental data on self-centering steel-concrete composite joints, this study validates the finite element modeling approach using the ZJD-3 prestressed joint. The details of Specimen ZJD-3 are illustrated in Figure 4 [23]. Specimen ZJD-3 shares identical beam-column dimensions and reinforcement details with the proposed joint, but it features prestressed strands passing directly through the joint core, whereas the novel joint employs a ground-level single-span post-tensioning approach (increasing susceptibility to adverse effects during core failure and requiring elevated-level strand tensioning during testing). The ZJD-3 system comprises:

1. Precast columns with embedded steel sleeves.

2. Steel-concrete composite beams with 400 mm-long Q235 steel beams (HN200 × 100 × 5.5 × 8).

3. Encased steel beams welded to 30 mm-thick Q345 end plates.

4. 440 mm-long main steel beam (HN350 × 175 × 7 × 11).

5. Connections via 10.9-grade high-strength bolts and four 15.2 mm-diameter 1860-grade prestressing strands.

Figure 4

Dimensional and reinforcement details of ZJD-3 [23].

Validation simulations replicated the experimental conditions: axial compression ratio of 0.3 and reversed cyclic loading at the column end.

3.2 Finite element modeling methodology

This study employs the plastic damage model to simulate the mechanical behavior of concrete, based on the uniaxial stress-strain relationship recommended in the Chinese national standard Code for Design of Concrete Structures [24] as its constitutive relationship, as illustrated in Figure 5(a). To simulate bond-slip behavior between reinforcing steel and concrete, a bilinear constitutive model (Figure 5(b)) was implemented for reinforcing bars, structural steel, and bolts; prestressed strands lacking a distinct yield plateau were modeled using a linear elastic constitutive model (Figure 5(c)). Except for reinforcing bars and steel strands, which use T3D2 elements, all other joint components employ C3D8R elements.

Figure 5

Constitutive relations of materials [24]. (a) Uniaxial stress-strain relationship. (b) Bilinear constitutive model. (c) Linear elastic constitutive model.

Mesh sensitivity critically governs computational accuracy, convergence behavior, and solution efficiency in finite element simulations. Following rigorous convergence studies, optimized element sizes were implemented: structural concrete at 50 mm; reinforcement systems (rebars and prestressing strands) at 75 mm; steel assemblies (sleeves, connection plates, structural shapes, end plates, and stiffeners) at 20 mm; and critical connectors (bolts and energy-dissipating bars) at 5 mm. Contact interactions employed hard contact pressure-overclosure for normal behavior and a penalty friction formulation for tangential behavior – with friction coefficients of µ = 0.35 for steel-to-bolt nut interfaces, µ = 0.6 for steel-to-concrete [25], and frictionless conditions (µ = 0) for bolt shank interactions with steel or concrete [26]. Constraint definitions utilized embedded region formulations to simulate concrete-rebar and concrete-encased steel interactions, while MPC-beam constraints governed the kinematic coupling between prestressing strands and concrete components.

Finite element boundary conditions were rigorously aligned with experimental protocols by constraining degrees of freedom consistent with the column-end loading methodology. The column base was fully fixed against translations in all directions and rotations about the x and y axes, while the column top was restrained vertically and in rotational degrees of freedom. Both beam ends were constrained against transverse displacements and rotations. To optimize computational efficiency while ensuring convergence, the analysis employed a quasi-static displacement protocol mirroring experimental sequences, stepping through drift ratios of 0.1, 0.2, 0.5, 0.75, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, and 4.0% with single-cycle implementation at each increment.

3.3 Analysis of finite element simulation results

The equivalent plastic strain (PEEQ) is the cumulative result of plastic strain during deformation and reflects the degree of plastic damage in the material throughout the loading process. Figure 6(a) shows the final failure mode of the simulated joint under low cyclic loading. As can be seen from the figure, plastic damage is primarily concentrated in the end plates and the upper and lower flanges of the steel beams near the end plates. No significant plastic damage occurs in other parts of the joint, which is consistent with the failure mode of the tested joint (Figure 6(b)).

Figure 6

Failure mode comparison: numerical simulation vs experimental test. (a) Numerical model. (b) Test specimen [23].

Hysteresis loop comparison and envelope skeleton curve comparison were illustrated in Figure 7. Figure 7(a) demonstrates strong agreement between experimental and finite element hysteresis curves, with marginal differences in load values at maximum displacements – confirming the simulation’s effectiveness in replicating specimen stress states. The ABAQUS model exhibits marginally higher initial stiffness due to idealized bound connections representing welded joints and unaccounted material imperfections. This simplification omits cyclic damage progression in steel components, resulting in moderately elevated structural stiffness compared to physical tests. Nevertheless, Figure 7(b) reveals closely matched skeleton curves, with only 1.8% variance in peak load capacity. This sub-2% discrepancy validates the simulation methodology’s accuracy and establishes its reliability for seismic performance assessment.

Figure 7

Hysteresis loop curves and envelope skeleton curves. (a) Hysteresis loop curves. (b) Envelope skeleton curves.

4.1 Failure modes

Figure 9 presents stress contours for structural steel and connection plates across all joint types. For visual clarity, energy-dissipating plates were computationally removed from the compression flange on the right side of PSC and RSC joints. Analysis reveals distinct damage patterns: PSC joints exhibit plastic deformation concentrated in energy-dissipating bars and the compression flange, while PC joints show localized yielding primarily in the compression flange. Crucially, neither joint type demonstrates significant flange buckling, with minimal plastic damage elsewhere. Conversely, RSC joints concentrate deformation exclusively in energy-dissipating bars, preserving flange integrity. Notably, PSC joints substantially reduce beam damage compared to PC variants through controlled energy dissipation, significantly enhancing post-earthquake recoverability by limiting structural steel yielding.

Figure 9

Stress contours of steel sections and connector plates. (a) PSC joint. (b) PC joint. (c) RSC joint.

Figure 10 compares prestressing strand stress-displacement responses in PSC and PC joints, revealing two critical findings:

1. PC joint strands exceed yield stress during loading, inducing failure through strand yielding that compromises structural safety. Conversely, PSC joint strands remain strictly within the elastic range, precluding yielding and maintaining structural integrity.

2. The addition of energy-dissipating plates increases the stiffness of the rotating section of the steel-concrete composite beam, reducing the rotational capacity of the PSC joint under the same displacement. This reduces the elongation of the steel strands, resulting in a slower rate of stress change in the PSC joint compared to the PC joint.

Figure 10

Stress-displacement curves of steel strands.

Overall, the failure modes of PSC joints and RSC joints mainly manifest as obvious buckling deformation of the energy-dissipating steel bars in the energy-absorbing plates, while the failure mode of PC joints mainly manifests as yielding of the prestressed steel strands. In all three types of assembled middle joints, the concrete in the joints remained largely undamaged, and the failure modes all manifested as bending failure at the beam ends.

4.2 Hysteretic loops and skeleton curves

Hysteresis loop area quantifies a specimen’s energy absorption capacity, with fuller loops indicating superior seismic energy dissipation. In Figure 11, flag-shaped hysteresis curves demonstrate self-centering capability with minimal residual displacement, enabling rapid post-earthquake recovery, while Z-shaped curves reveal structural vulnerabilities signaling inadequate energy dissipation and large residual deformations requiring mitigation. The skeleton curve traces peak load-bearing capacity versus lateral displacement, defining the load-deformation relationship throughout loading cycles. Figure 11 presents hysteresis and skeleton curves for all joint configurations, revealing three key performance distinctions. First, PSC and PC joints exhibit minimal residual deformation with hysteresis loops evolving from linear to flag-shaped configurations, while RSC joints display substantially larger residual deformations and spindle-to-Z-shaped hysteresis transitions. This behavioral divergence stems from RSC’s reduced initial stiffness, amplifying angular displacements at joint interfaces that intensify friction and sliding in web connection plates. Second, PSC joints demonstrate 18.1% higher yield load and 14.5% greater ultimate load than PC joints, confirming that energy-dissipating plates enhance load-bearing capacity beyond self-centering functionality. Most significantly, PSC joints achieve 93.1 and 142.1% improvements in yield and ultimate loads, respectively, over RSC joints – demonstrating that prestressing strands not only enable self-centering but fundamentally transform structural resistance mechanisms through active clamping forces.

Figure 11

Hysteretic behavior with envelope skeleton curves. (a) Hysteresis loops. (b) Envelope skeleton curves.

4.3 Ductility

The ductility of each joint was calculated using the tangent method [27]. The characteristic values of the skeleton curve and the ductility coefficients are shown in Table 2. As shown in the table, the displacement ductility coefficients of PC joints and RSC joints are greater than those of PSC joints. This is because PSC joints incorporate both energy-dissipating plates and prestressed steel strands, which increase the overall stiffness of the joint, thereby reducing its rotational capacity. This delays the joint’s entry into plastic deformation, resulting in an increase in the joint’s yield displacement and a corresponding reduction in ductility. The displacement ductility coefficients of all joints exceed 3.0, meeting the seismic code requirement for structural ductility coefficients greater than 3.0 [28], indicating that although the proposed PSC joints have reduced ductility, they still possess good deformation capacity.

4.4 Stiffness degradation and self-centering capability

The trend of stiffness degradation at each joint is represented using the secant stiffness [29]. The secant stiffness is defined as the ratio of the sum of the absolute values of the peak loads under the same positive and negative loading displacements to the sum of the corresponding absolute values of the displacements. The stiffness degradation curves for each joint are shown in Figure 12(a). As shown in Figure 12(a):

1. The stiffness of each joint decreases with increasing loading displacement, and the rate of stiffness degradation gradually slows down.

2. The initial stiffness of PSC joints is greater than that of PC joints, and the tangent stiffness of PSC joints at all loading displacements is also greater than that of PC joints, indicating that the addition of energy-dissipating plates can enhance the initial stiffness of joints.

3. The initial stiffness of PSC joints is approximately 1.52 times that of RSC joints, indicating that the addition of prestressed steel strands can significantly enhance the initial stiffness of joints, with a relatively large contribution to the initial stiffness.

Figure 12

Stiffness degradation curves and residual deformation angle curves. (a) Stiffness degradation curves. (b) Residual deformation angle curves.

Residual deformation angle – defined as residual displacement divided by the column height from loading point to base – serves as the key metric for joint self-centering capability, with values ≤0.2% indicating fully recoverable structures [30]. Figure 12(b) demonstrates that all joints exhibit progressively increasing residual angles with displacement amplitude, though PSC joints show marginally higher residual angles than PC joints across loading stages due to energy-dissipating plate inclusion. Crucially, both remain well below the 0.2% recoverability threshold, confirming effective self-centering functionality advantageous for post-earthquake repair, with PSC joints demonstrating superior resilience. Conversely, the residual deformation angles of RSC joints are significantly larger than those of PSC and PC joints at all stages, and the residual deformation angle at the end of loading is far greater than 0.2%, indicating that RSC joints lack self-centering capability due to the absence of prestressed steel strands.

4.5 Energy dissipation capacity

Cumulative energy consumption and equivalent viscous damping coefficient were employed as metrics to evaluate the energy dissipation capacity of each joint type, as illustrated in Figure 13(a). Key observations from Figure 13(a) are as follows:

1. Comparison between self-centering joints (PSC vs PC): The cumulative energy dissipation of the PSC joints, incorporating energy dissipation plates, consistently exceeded that of the PC joints without such plates. The disparity in cumulative dissipation progressively increased with loading cycles, culminating in the PSC joints achieving approximately 1.66 times the cumulative dissipation of the PC joints. This demonstrates that the integration of energy dissipation plates significantly enhances joint energy dissipation capacity.

2. Comparison between joints with energy dissipation plates (PSC vs RSC): Conversely, the cumulative energy consumption of the PSC joints (featuring prestressed strands) was consistently lower than that of the RSC joints lacking prestressed strands. Ultimately, the cumulative energy consumption of the PSC joints was 16.4% less than that of the RSC joints. This indicates that the inclusion of prestressed steel strands moderately reduces the joint’s energy dissipation capacity; however, the magnitude of this effect is notably smaller than the enhancement provided by energy dissipation plates.

Figure 13

Energy dissipation curves. (a) Cumulative energy dissipation curves. (b) Equivalent viscous damping ratios curves.

The equivalent viscous damping coefficients for each joint type are presented in Figure 13(b). Analysis of Figure 13(b) reveals the following key findings:

1. Trends in damping evolution: The equivalent viscous damping coefficients for all joint types exhibited generally consistent trends, increasing progressively with displacement amplitude. Notably, the coefficients for both the PSC joints and the RSC joints (both incorporating energy-dissipation plates) consistently surpassed those of the PC joints (without plates).

2. Damping values at ultimate displacement: At the ultimate displacement level, the equivalent viscous damping coefficients, calculated from hysteresis loops, were 0.128 for the PSC joints, 0.093 for the PC joints, and 0.248 for the RSC joints. In comparison, conventional reinforced concrete joints typically exhibit an equivalent viscous damping coefficient of approximately 0.1 [31]. These results demonstrate that joints equipped with energy-dissipation plates (PSC and RSC) exhibit superior energy dissipation capacity, whereas the PC joint (lacking energy-dissipation plates) demonstrates limited energy dissipation capability.

Based on the above analysis, it can be concluded that the proposed new joint type can ensure self-centering while balancing energy dissipation and ductility. The presence of energy dissipation plates can enhance the joint’s energy dissipation capacity, while the presence of prestressed steel strands can enhance the joint’s self-centering capacity. Overall, the new joint type meets the energy dissipation requirements of seismic design.

5.1 Calculation method for the shear bearing capacity of the joint core area

Joints serve as critical load-transferring components in structures, where joint failure precipitates simultaneous failure of beams and columns. The joint core region typically experiences a triaxial stress state combining axial compression, bending, and shear. Shear failure represents a prevalent failure mode in RC frame joint cores, characterized by diagonal cross-cracks or inclined concrete crushing. Hence, particular attention must be devoted to rational assessment of shear capacity in joint core design to mitigate brittle failure characteristics associated with core shear failure.

The shear capacity formula for beam-column joints is therefore a core theoretical tool in structural design, serving to quantify the shear resistance of the joint core region. This quantification is fundamental to ensuring the seismic design principle of strong-joint weak-component is achieved under extreme loading, particularly seismic actions. By providing a rational assessment of whether the joint can effectively transfer bending moments and shear forces from the beam ends, the formula plays a vital role in preventing shear failure within the core region. This prevention, in turn, safeguards the structural integrity, ductility, energy dissipation capacity, and overall collapse resistance of the structure.

Consequently, this formula finds wide application in engineering practice, including the seismic design of frame structures, the strengthening evaluation of existing buildings, and the connection design of prefabricated structures. Its importance extends beyond safety considerations, such as preventing catastrophic collapses triggered by brittle joint failure, to encompass economic efficiency. Through precise calculation, the formula enables engineers to balance necessary safety margins with material cost optimization.

The novel joint proposed in this study enhances core shear capacity through integrated square steel tubes within the joint core region. Analysis based on strut-and-tie theory reveals its load-resisting mechanism. During the initial loading stage, when the joint is in the elastic phase, shear force is primarily resisted by the inclined compressive struts formed by the concrete in the core region of the joint. At this stage, the shear deformation of the concrete in the core region is essentially consistent with that of the web of the square steel tube. As loading increases, multiple diagonal cracks develop in concrete, reducing the shear resistance contribution of concrete struts while inducing deformation incompatibility between concrete and square steel tubes. This leads to progressively increased shear resistance contribution from the square steel tubes. When further plastic deformation occurs, concrete’s strut action significantly diminishes due to crack propagation and spalling, concurrently mobilizing greater shear resistance from stirrups. Consequently, the total shear capacity of the novel joint comprises three components: core concrete contribution, square steel tube contribution, and stirrup contribution.

5.1.2 Shear capacity contribution of a square steel tube

According to the research findings in reference [33], the contribution of the flanges of the square steel tube to the shear capacity of the joint is relatively small; therefore, the influence of the flanges may be neglected in the shear capacity calculation. The web of the square steel tube in the core region is subjected to a combined shear-compression stress state under the concurrent action of compressive stress and shear stress, as shown in Figure 14.

Figure 14

Stress state of the square steel tube web.

Based on the distribution of shear stress at yield in the square steel tube web [34], illustrated in Figure 15, the shear capacity of the web can be determined. The calculation formula is given as follows:

(7) V s = 2 ∫ 0 b − 2 t τ ( x ) d x t = 1.8 t ( b − 2 t ) τ y ,

Figure 15

Shear stress distribution in the square steel tube web.

where t is the thickness of the square steel tube web; b is the side length of the square steel tube; and τ y is the shear stress of the square steel tube web when yielding.

In the elastic stage, the principal stresses at each point on the web are:

(8) σ 1 = σ s 2 + σ s 2 2 + τ s 2 ,

(9) σ 2 = 0 ,

(10) σ 3 = σ s 2 − σ s 2 2 + τ s 2 .

When the joint reaches the ultimate bearing capacity, the steel reaches the plastic flow state, which conforms to the Mises yield rule:

(11) f y = 1 2 [ ( σ 1 − σ 2 ) 2 + ( σ 2 − σ 3 ) 2 + ( σ 3 − σ 1 ) 2 ] ,

where f y is the tensile yield strength of the web.

Substituting the principal stress expression into Eq. (11) yields the shear stress at web yield:

(12) τ y = f y 2 − σ c 2 3 .

The research shows that the axial pressure has little effect on the shear capacity of the square steel tube web [35]. Therefore, the shear stress expression when the web yields can be simplified as:

(13) τ y = f y 3 .

At joint failure, the web of the square steel tube does not yield. Based on recommendations from the Architectural Institute of Japan (AIJ) [31], a reduction factor of 0.5 is applied to derive the calculation formula for the shear capacity of the square steel tube web, as shown in Eq. (14).

(14) V s = 0.9 t ( b − 2 t ) 3 f y .

5.2 Verification of shear capacity calculation accuracy

Analysis of the failure mode in the aforementioned PSC joint reveals that failure manifested at the beam ends, concentrating on the energy-dissipating steel bars. The joint core concrete exhibited no significant damage, and no shear failure occurred in the core region. To evaluate the accuracy of the proposed shear capacity calculation formula, two additional joint specimens were designed based on the prototype PSC joint. Key parameter variations in their core regions are detailed in Table 3. By reducing the strength grades of the square steel tubes and stirrups, along with decreasing the core region stirrup reinforcement ratio, the shear capacity of the joint core was intentionally compromised. Through refined finite element simulations, this study examines whether the proposed formula accurately assesses the actual shear capacity when core region shear failure is induced.

Comparison between calculated and simulated shear capacity values revealed certain discrepancies. This occurred because the designed joint model lost its load-bearing capacity prior to failure in the core region, leading to termination of loading. Therefore, to further verify the accuracy of the proposed shear capacity calculation formula, two joint models were designed following the weak-joint strong-component principle. The shear capacity of the joint core region was intentionally reduced by decreasing the strength grades of the square steel tube and stirrups, along with reducing the stirrup reinforcement ratio in the core region. This approach ensured the occurrence of shear failure in the joint core region. Variations in core region parameters for the joint models are presented in Table 3.

A comparison between the calculated shear capacity values and finite element simulation results for the joint models with intentionally weakened core regions is presented in Table 4. The ratios of calculated to simulated values are consistently close to unity, demonstrating that the proposed shear capacity calculation formula can serve as a reliable reference for practical engineering applications of this novel joint.