Seismic Behavior of Cluster-Connected Prefabricated Shear Walls under Different Axial Compression Ratios

Abstract

This study analyzes the nonlinear seismic behavior of cluster-connected prefabricated shear walls under varying axial compression ratios. The investigation focuses on the connectivity of shear wall segments assembled using cluster connections rather than separate walls connected by beams. Using the finite element software ABAQUS, this study simulates monotonic horizontal displacement loading to evaluate the yield strength, peak strength, and deformation capacity of the shear walls. The results demonstrate that the horizontal load-bearing capacity of the shear wall significantly improves with an increase in axial compression ratio, while the axial compression ratio also influences ductility. The numerical simulations are validated against experimental data, confirming the accuracy of the model. These findings provide essential insights for optimizing the seismic design of precast shear walls.

1. Introduction

At present, prefabricated buildings, whose structure forms mainly incorporate prefabricated concrete shear wall structures and frame structures, play a key role in the industrialization of the building industry worldwide [1,2]. As an ideal lateral-force-resisting member, shear walls can effectively improve the overall stiffness of the structure. However, the connection between each precast shear wall poses challenges in the application of prefabricated concrete shear walls. Many research works have been conducted in the past to explore a suitable method to design a new connection structure with good mechanical performance and reasonable structural forms for construction convenience.

In past research, Chen et al. [3] conducted an experimental study on the shear resistance of precast reinforced concrete (RC) shear walls with novel bundled connections, showing that the new connection method significantly enhances the shear capacity of the walls. Their research provides valuable insights into the structural performance and potential applications of these innovative connections in seismic regions. Khaled et al. [4] conducted comparative tests on various steel bar connection methods of precast concrete shear walls, i.e., steel sleeve connection, welded connection between steel bar and embedded angle steel, steel sleeve plus shear key connection, and connection with or without bonded pre-stressed steel rods, etc. Can et al. [5] proposed a new type of precast shear wall connection structure that can prevent the force from moving the foundation to the upper wall, thereby protecting the wall from brittle failure. Xu et al. [6] demonstrated that precast shear walls with sleeve connections exhibit improved seismic performance, supported by both experimental and numerical studies. Lv and Guan [7] investigated the seismic behavior of precast shear walls with vertical reinforcements spliced by two different grouting methods. Their tests revealed significant differences in performance between the grouting techniques, highlighting the impact of connection methods on the overall seismic resistance of the walls. Although the connection joints have been extensively investigated, research has yet to systematically investigate the impact of the axial compression ratio on its seismic behavior.

The axial compression ratio is one of the main factors governing the seismic performance of shear walls. Li and Li [8] conducted tests on three groups of shear walls with axial compression ratios of 0.1, 0.2, and 0.3. Based on test results, they concluded that the horizontal bearing capacity of the shear wall increased accordingly as the axial compression ratio increased within a certain range. However, the body ductility, strength, and stiffness of the wall degraded significantly. Looi et al. [9] investigated the effects of axial load on the seismic performance of reinforced concrete walls with short shear spans, finding that higher axial loads improved the horizontal load-bearing capacity but reduced ductility. Their research highlights the trade-off between strength and ductility in such structural elements under varying axial loads. Pérez Gavilán et al. [10] conducted an experimental study on confined masonry walls with varying aspect ratios, revealing that aspect ratio significantly impacts the seismic performance of these walls. Their findings demonstrate that walls with lower aspect ratios exhibit better shear resistance and overall seismic behavior compared to those with higher aspect ratios. Zhang and Zeng [11] carried out quasi-static tests to examine the seismic performance of shear walls with axial compression ratios of 0.14, 0.28, 0.43, and 0.57. The results showed that as the axial compression ratio increased, the yield strength, ultimate strength, and ductility of the shear wall specimens all had a peak value, i.e., their value increased first and then dropped. Lefas et al. [12] showed that the axial compression ratio has a considerable influence on the ductility of shear wall specimens. They pointed out that as the axial compression ratio increased, the ductility of the specimen decreased accordingly, and when the axial compression ratio was greater than or equal to 0.25, the ductility of shear walls degraded rapidly.

Current finite element software still has difficulty in accurately simulating the entire stress process of the structure under cyclic loading since the cyclic loading test of shear walls involves many factors, such as local crushing of concrete, cracking and crack closure of wall concrete, the Bauschinger effect of internal steel bars, bond degradation between steel bars and concrete, stress stiffness of concrete and steel bars, and other nonlinear factors. However, in recent years, researchers [13,14,15,16,17,18,19,20] found that the monotonic loading simulation method could reflect the failure mode of shear walls, and the load–displacement curve generated was also consistent with cyclic loading test results.

This study aims to investigate the seismic performance of precast shear walls with novel bundled connections under cyclic loading. By employing a combination of experimental tests and numerical simulations, we seek to provide a comprehensive understanding of the behavior of these structures under seismic conditions. The primary objectives include evaluating the impact of different axial compression ratios on the load-bearing capacity, ductility, and failure modes of the shear walls. Our research contributes to the existing body of knowledge by offering new insights into the design and optimization of precast shear wall connections, which hold significant practical relevance in civil engineering.

2. Numerical Model

In this study, the finite element software ABAQUS 6.14 was employed to simulate and analyze the seismic behavior of the specimens. To accurately compare the laboratory test and simulation results, the average value of forward and reverse loads imposed on the specimen was used as the skeleton curve in modeling the specimen. The precast short-leg shear wall specimen (PSSW) and the precast ordinary shear wall specimen (POSW) were selected for simulation.

The detailed dimensions and reinforcement specifications of the POSW specimen are shown in Figure 1. The dimensions of the POSW specimen were 2600 mm in height, 1800 mm in width, and 200 mm in length. The PSSW specimen had dimensions of 2600 mm in height, 1000 mm in width, and 200 mm in length. The foundation beam had a length of 1900 mm, a width of 450 mm, and a height of 450 mm. The height of the outer preformed holes was 1320 mm, while one of the inner preformed holes was 720 mm. The length of the vertical steel bars extending into the preformed holes was 50 mm shorter than the height of the corresponding preformed holes. The anchor length of the additional steel bars in the preformed holes was 480 mm.

For the PSSW, indirect lap splices used HRB400 grade steel bars with a diameter of 14 mm. In the POSW, the central indirect lap splices used HRB400 grade steel bars with a diameter of 12 mm, while the preformed edge holes used HRB400 grade steel bars with a diameter of 14 mm. The foundation beam had a length of 1900 mm, a width of 450 mm, and a height of 450 mm. The height of the outer preformed holes was 1320 mm, while one of the inner preformed holes was 720 mm. The length of the vertical steel bars extending into the preformed holes was 50 mm shorter than the height of the corresponding preformed holes. The anchor length of the additional steel bars in the preformed holes was 480 mm.

The plastic damage model was adopted for the constitutive relationship of concrete. The uniaxial compression and tensile stress–strain curves of C35 concrete are shown in Figure 2. The elastic modulus E and Poisson’s ratio of the precast concrete in the model were set as 31 GPa and 0.2, respectively. Detailed parameters of the concrete damage plasticity model are shown in

Table 1. Model parameters of concrete.

Density [kg·m−3]	Poisson’s Ratio	Eccentricity [%]	Dilation Angle [°]	fb0/fc0	Kc	μ
2400	0.2	0.1	38	1.16	2/3	0.005

. Incorporating the damage factor, the stress–strain curve expressions of concrete under uniaxial tension and compression are as follows:

$$\sigma =(1-d)\overline{\sigma }$$

(1)

$$\overline{\sigma }=D_0^{el}(\epsilon -{\epsilon }^{pl})$$

(2)

$$\sigma =(1-d)D_0^{el}(\epsilon -{\epsilon }^{pl})$$

(3)

where $d$ is the damage factor, $\overline{\sigma }$ is the effective stress, $\epsilon$ is the strain of concrete, ${\epsilon }^{pl}$ is the plastic strain of concrete, and $D_0^{el}$ is the initial stiffness of the material.

It should be noted that the compressive strength of Grade C35 concrete should normally be greater than 35 MPa. It seems that the results shown in Figure 2 underestimate the strength of C35 concrete, which may have resulted from the sample disturbance during the transportation stage and the workmanship in preparing the test cubes. As Figure 2 represents the actual results from the experimental measurement, both curves shown in Figure 2 were adopted for the numerical analyses.

In ABAQUS, Drucker–Prager’s equation, as illustrated in Equation (4), is adopted to simulate the damaged plasticity model for the concrete.

$$G=\sqrt{{\left(\epsilon {\sigma }_{t0}\tan \psi \right)}^2+q^2}-p\tan \psi$$

(4)

where G is the flow potential, ε is the eccentricity, ψ is the dilation angle on the q–p plane, and σ_t0 is the initial yield stress. More details on the constitutive models adopted in ABAQUS can be observed in the ABAQUS Theory Guide online document. The parameters adopted in ABAQUS are illustrated in Table 1.

The ideal elastoplastic double broken line hardening model was adopted for the constitutive relationship of the steel bars. The yield strengths of the steel bars were determined based on material property tests, including 6 different types of steel bars with varying strength levels and diameters. The specific parameters are shown in

Table 2. Mechanical properties of rebar.

Rebar Specification	Diameter [mm]	Yield Strength [MPa]	Ultimate Strength [MPa]	Elongation [%]	Hardening Ratio
HPB300	6	392	479	21	0.82
HPB400	8	520	620	22	0.84
	10	491	613	24	0.80
	12	505	631	25	0.80
	14	479	633	23	0.76
	16	474	620	24	0.76

. The elastic modulus is 210 Gpa. Each type of steel bar includes three 600 mm long test specimens, and 18 specimens were tested in total.

In the finite element model, the eight-node hexahedron linear reduced integration unit C3D8R was adopted for the simulation of concrete, and the space truss unit T3D2 was employed for the simulation of steel bar. In order to simulate the co-deformation of steel bars and concrete, the embedded analysis model was adopted. The models of steel bars and concrete were built separately, and then the steel bar units were embedded into the concrete model through the embedded constraints of the interaction module. However, the model ignored the bond–slip effect between steel bars and concrete. The dimensions of the concrete solid model and the steel frame were determined according to the actual test dimensions. Figure 3 illustrates the finite element model of the shear wall.

The surface-to-surface contact method was adopted for the simulation of the contact interface between the prefabricated wall and the foundation beam. The contact in the normal direction of the interface was set as hard contact, which can serve as a constraint during contact but a free boundary while the solid bodies on the other side of the interface departed from each other (i.e., the pressure imposed is 0 or negative). In the tangential direction, the Coulomb friction model was employed. The friction coefficient is 0.6, and the maximum friction shear stress is 2.5 MPa. In addition, no failure occurred at the interface between the grouting material, corrugated pipe, and concrete during the test. Therefore, the corrugated pipe was not directly established in the model. Instead, the interface between grouting material and concrete to which the TIE model applied was employed to simulate it.

Since the cyclic loading test of shear walls involves many factors, such as local crushing of concrete, cracking and crack closure of wall concrete, Bauschinger effect of internal steel bars, bond degradation between steel bars and concrete, stress stiffness of concrete and steel bars and other nonlinear factors, current version of ABAQUS still have difficulty in accurately simulating the entire stress process of the structure under cyclic loading. This study is based on the skeleton curve of the cyclic loading test of the shear wall. The numerical simulation of the entire monotonic displacement loading process of the precast shear wall was carried out. Figure 4 illustrates the boundary conditions of the model.

3. Comparisons of the Results

3.1. Results of the Specimen PSSW

3.1.1. Failure Mode for the Specimen PSSW

The equivalent plastic strain PEEQ can represent the accumulated plastic damage of the concrete specimen PSSW during the entire monotonic loading process, as shown in Figure 5. As the displacement increases, the plastic deformation of the shear wall gradually develops from the bottom of the compression side of the wall. The final failure mode is associated with the crushing of the concrete on the compression side, and the equivalent plastic deformation reached the maximum value, which is consistent with the test results. The experimental validation for the numerical models was based on the quasi-static cyclic loading tests described by our research team [21], which evaluated the seismic performance of novel bundled connections in precast shear walls.

3.1.2. Load–Displacement Curves for the Specimen PSSW

The comparison of load–displacement curves for the PSSW specimen from laboratory tests and simulations is shown in Figure 6. The curve obtained from the laboratory test aligns with the results from the simulation. Both curves have an elastic stage, a yielding stage, and a declining stage. The ultimate horizontal load calculated by ABAQUS is 365 kN, which is approximately of the same order as the measured one from laboratory tests (331 kN). However, the yield and ultimate displacement from the test results are slightly larger than those from finite element calculation due to ignoring the Bauschinger effect of internal steel bars and the bond–slip between steel bars and concrete in ABAQUS.

3.1.3. Crack Development for the Specimen PSWW

The distribution of tensile damage in the shear wall is shown in Figure 7. It can be observed that the tensile damage emerged from the left bottom of the shear wall specimen during the forward loading process. As the displacement increases, the tensile damage gradually expands to the compression zone and the end of the loading beam. Additionally, the compression damage distribution diagram of the shear wall (Figure 8) indicated that the compression damage at the bottom of the compression zone of the precast wall gradually develops as the horizontal displacement at the top of the wall increases. These results are consistent with the development and distribution of cracks during forward loading in field tests.

The influence of the axial force ratio on crack development is significant. At higher axial force ratios, cracks tend to initiate at lower loads and propagate more quickly compared to lower axial force ratios. This observation aligns with the findings from our numerical simulations and experimental results, which demonstrate that increased axial compression can constrain the development of the plastic zone, leading to a more brittle failure mode.

3.2. Comparison for Specimen PSSW

3.2.1. Failure Mode of the Shear Wall

The equivalent plastic strain cloud map of specimen POSW is shown in Figure 9. As the displacement increases, the plastic deformation of the shear wall gradually develops from the bottom of the compression side of the wall. The final failure mode is the crushing of the concrete on the compression side, and the equivalent plastic deformation reached the maximum value, which is consistent with the test results.

3.2.2. Load–Displacement Curves

Figure 10 indicates that the trend in the load–displacement curve of specimen POSW from the laboratory test is consistent with that calculated by ABAQUS. Both curves are supported with an elastic stage, a yielding stage, and a declining stage. The ultimate horizontal load from finite element analysis is 865 kN, which is in agreement with the result from the laboratory test (825 kN). However, the yield and ultimate displacement obtained from the laboratory test are slightly larger than those from finite element analyses. In general, the results of the test align with those from the simulation.

3.2.3. Crack Development

According to the tensile damage distribution diagram shown in Figure 11, tensile damage began to appear at the left bottom of the shear wall specimen POSW during forward loading. As the displacement increased, the tensile damage gradually spread to the compression zone and the end of the loading beam. Figure 12 illustrates the distribution of compression damage of the shear wall. As the horizontal displacement at the top of the wall increases, the compression damage at the bottom of the compression zone of the precast wall gradually develops, which is consistent with the development and distribution of cracks during forward loading in field tests.

4. Parametric Analysis

To investigate the influence of the axial compression ratio on the mechanical properties of cluster-connected shear wall specimens, finite element calculations on specimens PSSW (precast short-leg shear wall) and POSW (precast ordinary shear wall) under axial compression ratios (n) of 0.1 and 0.3 were conducted. The results were compared with those corresponding to an axial compression ratio of 0.2.

4.1. Horizontal Bearing Capacity

According to Figure 13a, the horizontal bearing capacity of the precast shear wall PSSW improves as the axial compression ratio n increases from 0.1 to 0.3. Similarly, the horizontal load-bearing capacity of specimen POSW also increases with the increase in axial compression ratio, as shown in Figure 13b. Specifically, the yield load of the precast short-leg shear wall PSSW increases from 215 kN to 311 kN, and the horizontal peak load increases from 306 kN to 410 kN (

Table 3. Bearing capacity of specimens under different axial compression ratios.

Number of Specimens	Axial Compression Ratio	Axial Force [kN]	Yield Load [kN]	Peak Load [kN]
PSSW	0.1	468	215	306
	0.2	936	276	365
	0.3	1404	311	410
POSW	0.1	842	460	679
	0.2	1684	620	865
	0.3	2526	722	960

) when the axial compression ratio rises from 0.1 to 0.3. Additionally, for the precast ordinary shear wall POSW, the yield load grows from 460 kN to 722 kN, and the horizontal peak load increases from 679 kN to 960 kN as the axial compression ratio increases from 0.1 to 0.3. This trend suggests that the axial compression ratio has a significant impact on the yield strength and peak strength of the shear wall.

4.2. Ductility and Failure Mode

While the axial compression ratio can increase the lateral strength of shear walls up to a certain point, it generally has a diminishing effect on ductility. Higher axial compression ratios typically lead to a more brittle failure mode, characterized by rapid crack propagation and localized crushing at the base of the wall. This observation aligns with well-established knowledge in the field. Our analysis focused on the specific range of axial compression ratios examined in this study. We clarified that beyond the examined range, higher axial compression ratios typically lead to a decrease in lateral strength and ductility, consistent with previous research findings.

4.3. Damage Progression and Internal Force Redistribution

The progression of damage in the shear walls was monitored through the distribution of tensile and compressive damage. At higher axial compression ratios, damage initiated at lower load levels and propagated more quickly. This was evident from the earlier onset of tensile damage and the accelerated development of compressive damage at the bottom of the wall. The damage patterns observed in the numerical simulations aligned well with the experimental results, confirming the model’s accuracy.

Higher axial compression ratios led to increased axial forces in the vertical reinforcement, affecting the horizontal load-bearing capacity of the wall. The redistribution of internal forces was observed through the strain profiles of the reinforcement, indicating higher stress concentrations at the base and corners of the wall as the axial compression ratio increased.

5. Conclusions

This study employed the large-scale finite element software ABAQUS to perform nonlinear finite element analysis on the precast short-leg shear wall specimen PSSW and the precast ordinary shear wall specimen POSW under monotonic horizontal displacement loading. The following conclusions can be drawn:

(1) Based on the finite element equivalent plastic strain PEEQ cloud map, the accumulation of plastic damage of the precast wall concrete during the entire monotonic loading process was analyzed. The results from the numerical simulation were also compared with those from the laboratory test to verify the accuracy of the proposed model.

(2) The load–displacement curves under monotonic loading and the cyclic loading test skeleton curve were compared and analyzed; the results showed that the simulation results align with those of the test, indicating that numerical simulation could effectively emulate the mechanical behavior during the test process.

(3) Finite element calculations were conducted for specimens PSSW and POSW under axial compression ratios of 0.1 and 0.3, respectively. The results were then compared with those corresponding to the axial compression ratio of 0.2. The results showed that the axial compression ratio had a significant influence on the yield strength and peak strength of the shear wall specimens. As the axial compression ratio increases, the horizontal bearing capacity of the shear wall increases accordingly.