Experimental and numerical study on mechanical properties of PC shear wall bolted connection joints

Abstract

Steel-bolt connection joint with embedded steel is increasingly utilized for prefabricated concrete (PC) shear walls and plays an indispensable role in the shear force transmission process. To enhance the mechanical performance of the connection joint in PC walls, a new pre-embedded steel bolt connection is proposed in this paper. The new connection can quickly connect prefabricated shear walls, with a simple construction process and short construction period. The present study respectively designed and fabricated four specimens with different connection configurations to conduct shear performance tests on the steel connectors and hysteresis performance tests on the connection joint. A numerical model for the joint was then established based on the shear test data of the steel connectors, to further analyze the mechanical performance of the proposed connection joint. The results indicate that depending on the different steel connection configurations, shear failure modes can be categorized into bolt shear failure and embedded steel pullout failure. The shear studs have minimal impact on the shear performance of the steel connectors. The flanges on both sides of the joint effectively enhance the hysteresis performance of the connection joint. Among the tested specimens, the one with a steel configuration at the connection joint featuring ribbed channel steel exhibits superior hysteresis performance, with higher load-bearing capacity, gradual stiffness degradation, and excellent energy dissipation capability. It is recommended to use these connection details for embedded steel connections at the joint during the design.

Keywords

steel-bolt connection prefabricated concrete shear wall pre-embedded steel experimental study seismic performance finite element simulation

Introduction

With the rapid development of prefabricated building structures, prefabricated components are increasingly applied to practical engineering, especially for prefabricated concrete (PC) shear walls which have high lateral stiffness and load resistance as well as convenient assembly (Huang et al., 2018). The most crucial aspect of the PC shear wall system is the joint connection technology which significantly influences the load-carrying capacity, stiffness, and stability of the shear wall. Shear walls are thin-wall structures, which is a hot topic for prefabricated structures. Although numerous breakthroughs and accomplishments in the research of PC shear wall structural systems have been achieved, there are currently two prominent issues. On the one hand, extensive experimental research for innovative exploration is required to ensure equivalent overall performance for PC walls to cast-in-place walls. On the other hand, the connection joint between prefabricated shear wall structures needs to strike a balance between connection strength and economic cost. This dilemma has led to the prevalence of cast-in-place connections in current PC shear wall structures, with the second issue still struggling. Therefore, investigating novel connection joints for PC shear walls is of significant importance in further enhancing the overall seismic performance and promoting the engineering application.

The commonly employed wet connections for vertical and horizontal joints of shear walls are relatively mature, including horizontal loop connections, grouting sleeve connections, grout anchor lap connections, and so on (Cheng et al., 2023; Jiang et al., 2021; Liu et al., 2023; Zhi et al., 2017). With these connection forms, shear walls generally exhibit a load-carrying capacity comparable to that of cast-in-place shear walls. However, they face challenges such as complex on-site construction, long construction periods, and extensive wet operations. Currently, dry connections including mechanical connections, embedded steel welding, and embedded steel bolt connections are developed for shear walls (Fu et al., 2022; Guo et al., 2019; Sun and Qiu 2015, 2023; Yang et al., 2023). These connections can avoid extensive on-site wet operations, reduce exposed steel for ease of transportation, and eliminate the influence of weather conditions during construction. Mechanical connections have high installation difficulty and cost, while the performance of welding connections is vulnerable to the quality of welding, with the potential for residual stress and challenges in ensuring construction quality (Arandjelovic et al., 2024). In contrast, bolt connections are convenient to construct and unaffected by weather conditions, also can reduce environmental pollution, and offer advantages such as greater ductility and strong energy dissipation compared to welding (Liu et al., 2021). Nowadays, bolt connections are used in various aspects such as frame joints, shear wall joints, and frame-shear wall connection joints (Du et al., 2024; Ioannis et al., 2022; Lu et al., 2024; Pan et al., 2021; Zhai et al., 2024).

Currently, the common bolt connections for PC shear walls include externally wrapped steel bolt connections and embedded steel bolt connections. Q Han et al. utilized steel connectors to link the upper and lower shear walls, ensuring that wall failure was not induced by the connection points, while simultaneously enhancing the wall’s stiffness and load-carrying capacity (Han et al., 2020). W Li introduced a novel embedded steel bolt connection joint, which effectively improved the structural plastic deformation capacity and seismic performance (Li et al., 2021). H Yu proposed a new type of steel plate-bolted horizontal connection for precast concrete shear walls, which enhances the bearing capacity, energy dissipation capacity, and ductility of walls (Yu et al., 2024). W Wang designed an improved AHW connection in a PCSW structure, which exhibited satisfactory seismic performance (Wang et al., 2018).

Ramin Vaghei et al. increased the cumulative energy dissipation of bolt connections compared to hoop reinforcement dowel connections by adding rubber pads in the steel connection area of the vertical joint (Vaghei et al., 2014, 2017). R.S. Henry et al. discovered that shear walls assembled by embedding steel on both sides exhibited good seismic performance (Henry et al., 2016). S Sritharan et al. introduced a prefabricated wall system (Precast concrete wall with end columns) with easily replaceable connectors (Sritharan et al., 2015). T Guo et al. proposed a novel vertical joint device with friction connectors, which is capable of effectively transmitting internal forces, dissipating load energy, and exhibiting a pronounced self-resetting effect (Guo et al., 2018). T Ding et al. designed a shear wall with the bolted end-plate design for deconstruction (DfD) connections (Ding et al., 2020). S Jiang et al. proposed a new form of bolted connection assembled shear wall (BASW), which can improve the structural integrity of assembled shear walls (Jiang et al., 2023).

In general, the construction of wet-connected shear walls is cumbersome as a result of extensive wet operations, while dry-connected shear walls exhibit lower overall integrity and uncertain seismic performance, in comparison to cast-in-place shear walls. Compared to welded connections and sleeve grouting connections, bolted connections allow for some slip and deformation, enhancing the overall stability of the structure, an also can accommodate on-site construction and installation errors, ensuring consistent mechanical properties of the structure. In addition, bolted connections are easy to disassemble, easy to maintain, and prefabricated components are relatively inexpensive, without the need for special equipment. Construction, installation, and disassembly are relatively fast. Therefore, an efficient connection joint with assembly convenience and excellent seismic performance is necessary for PC shear walls. To enhance the safety, ease of installation, and durability of the joint, it is essential to propose a bolted connection joint that requires minimal wet operations. During the assembly of shear walls, the two adjacent walls are connected at the joint using pre-embedded steel bolts, followed by the casting of concrete. This connection method ensures rapid assembly and reduces the extent of wet operations. This joint has high shear strength, high stiffness, and convenient on-site construction splicing. To delve deeper into the mechanical performance of this innovative joint, a series of shear performance tests, low-cycle reciprocating tests, and finite element simulations were conducted on four connection joints. The study examined the failure modes and hysteresis performance of the joint, validating the structural reliability and providing a basis for subsequent designs of overall connections in shear walls.

Description of new bolted connections

Bolted connections minimize on-site wet work and are easy to install. The use of grooves eliminates the need for external embedded parts on shear walls, resulting in smooth wall edges. This design prevents out-of-plane distortion of connection joints, avoids eccentric bending moments, and enhances bearing capacity, stiffness, and energy dissipation. Once the shear walls are installed, the joint areas filled with concrete in the cavities provide excellent fire resistance and corrosion protection.

The proposed novel bolted connection joint which adopts dry-wet combined technology (DWCBJ) consists of pre-embedded steel plates and high-strength bolts, forming a PC shear wall structural system as illustrated in Figure 1. Horizontal connection is similar to vertical connection, with the same embedded connectors. The vertical connection joint is depicted in Figure 2. In the horizontal connection area, the floor slab is secured by grouting through reserved holes for the exposed steel bars in the lower shear wall. High-strength grouting material is applied at the contact interface between the floor slab and shear wall to ensure compactness and effective force transmission. The floor slab is treated as a rigid diaphragm to transfer shear force and axial force on the shear wall, and the main force on the connecting joints is focused on the shear wall, without considering the floor slab. The pre-embedded steel connections at the joint between shear walls serve as a box-type structure to transmit both gravity and lateral loads. Considering factors such as joint force distribution, construction convenience, and future replacement and maintenance, it suggests to equidistantly arrange the connection joints along the central axis of the wall. In practical structural design and construction, the specific number and spacing of joints should be further optimized based on the wall length. This prefabricated assembly of the shear wall system allows for rapid on-site assembly, requiring only a minimal amount of concrete filling at the connection joint. This dry-wet hybrid connection approach not only ensures excellent mechanical performance of the connection joint but also significantly shortens the construction period.

Figure 1.

PC shear wall structural system with bolted connections.

Figure 2.

Prefabricated component connection joint.

Shear performance test of the bolt connection steel connector

To investigate the static performance of the connection joint and establish a shear-displacement model for bolted connection joints, monotonic shear tests were conducted on steel connectors. The shear-displacement model derived from these tests was utilized in subsequent chapters to develop a numerical model for bolted connection joints. Consequently, the parameters of the steel connectors in the monotonic shear tests are consistent with those of the PC shear wall connection joint.

Test specimens

During the test, a total of four specimens with various types of connection area steel were designed to investigate the mechanical performance of the connection joint. The selected specimens are along with PC walls within a shared 400 mm range of concrete above and below the joint. The concrete shell provides protection for the joints, which reduces the direct exposure of bolted connections to environmental factors that typically lead to fatigue and corrosion. In this experiment, the wall thickness is designed to be 200 mm, and the concrete strength is rated at C40. Longitudinal reinforcement adopts HRB400 grade, lateral ties adopt HPB300 grade, and the steel material is Q235 steel. The bolts utilized are high-strength M16 bolts with a grade of 8.8.

The basic parameters of the steel in the connection zone are detailed in Table 1. For SW-1, the steel in the connection zone was rectangular steel plate; for SW-2, H-shaped steel; for SW-3, T-shaped steel; and SW-4, ribbed channel steel. On the basis of standard specimen SW-1, SW-2 changed the rectangular steel plate to an H-shaped steel structure with flanges on both sides. And SW-3 utilized T-shaped steel and bolt connection without requiring the connection of steel plates, the stress mode of which is the same as that of H-shaped steel. The ribbed channel steel of SW-4 adds a steel plate surface to the contact surface of the H-shaped steel, and its bolts are subjected to different forces compared to the first three types. The bolts are subjected to both shear and axial tensile forces. The embedded sections of the steel all featured unequal-legged H-shaped steel with shear-resistant flanges and shear studs. The detailed construction of the joint connection is illustrated in Figure 3. The dimensional details of the specimens and the embedded steel are illustrated in Figure 4, by taking SW-1 as an example, which has dimensions of 300 mm high, 400 mm wide, and 200 mm thick. Moreover, strain gauges are positioned on the shear web plates on both sides of the steel embedded in the lower part of each specimen. These gauges are used for real-time monitoring of strain changes in the steel plates. Additionally, four strain gauges are placed on both the tensile and compressive sides of the shear studs to record the strain development in the shear studs (Figure 4(c)). SW-1 served as the control group. In comparison to SW-1, both sides of connection joints for SW-2, SW-3, and SW-4 featured “flanges”. Under seismic loading, the load-bearing capacity of the connection for SW-1∼SW-3 was provided solely by the shear-bearing capacity of the connecting bolts. However, due to differences in the construction of the connection zone, the load-bearing capacity of the connection for SW-4 was influenced by both the shear-bearing capacity of the bolts and the tensile resistance of the bolts and nuts.

Table 1.

Basic parameters of the specimens.

No.	Buried depth of the section steel/mm	Type of connection area steel	Eccentricity of connecting steel/mm
SW-1	150	Rectangular steel	0
SW-2	150	H-shaped steel	0
SW-3	150	T-shaped steel	6
SW-4	150	Ribbed channel steel	0

Figure 3.

Detailed construction of connections.

Figure 4.

General arrangement of SW-1.

Material properties

Standard monotonic tensile tests were conducted on steel and reinforcing bars in compliance with the testing methods specified in the national standard (Metallic materials-Tensile testing-Part 1: Method of test at room temperature, China, 2010), and the results are presented in Table 2. The concrete strength grade for both the walls and the hidden columns was C40. Simultaneously with casting the specimens, three standard cubic specimens with a side length of 150 mm were prepared. These cubic specimens were subjected to the same curing conditions as the specimens. The average compressive strength of the cubes was 41.5 MPa.

Table 2.

Rebar and steel material properties.

Category	Diameter/Thickness/mm	Yield strength f_y/Mpa	Ultimate strength f_u/Mpa	Elastic modulus E/(×10⁵ N/mm²)
HRB400 rebar	12	426.7	582.2	2.10
Q235 steel	12	258.5	430.6	2.06
Bolt	16	640	800	2.06

Loading program

The test device and on-site layout are shown in Figure 5. The base is fixed to the ground by the anchor bolts. To ensure that the joint can not deviate from the plane, two steel plates connected by screws are installed at both ends of the upper wall. The horizontal load is applied by welding the MTS servo actuator with the left-end steel plate. The specimen uses a loading system that combines pre-yield load control and post-yield displacement control. The pre-yield loading rate is 0.5 kN/s, and the post-yield loading rate is 1 mm/min until the load drops to 85% of the peak load.

Figure 5.

Test setup.

Test results

The failure modes of the four specimens primarily encompass bolt shear failure and embedded steel pull-out failure. The failure characteristics of each specimen are depicted in Figure 6. In the case of SW-1, due to the absence of flanges on both sides of the steel, the deformation in the joint area enlarges as the load increases, ultimately causing the connecting bolts to yield and resulting in bolt shear failure. The failure modes of SW-2 to SW-4 are essentially similar, with significant deformations in the steel plate connection zone as the presence of flange plates on both sides of the joint constraints, culminating in the eventual pull-out failure of the embedded steel. Therefore, it can be observed that SW-1, due to the absence of flanges, primarily bears shear forces at the connecting joints during loading, making bolt shear failure more likely. In contrast, SW-2 to SW-4, with the presence of flanges, enhance the lateral stiffness of the connecting joints and primarily bear tensile forces during loading, making the embedded steel more susceptible to pull-out failure. Based on the rationality of the specimen design and relevant construction measures, it is necessary to prevent the occurrence of shear failure of bolts similar to SW-1 connectors and enhance the maximum bearing capacity of the connection joints.

Figure 6.

Failure characteristics of specimens.

The load-displacement curves for different connection joints are presented in Figure 7(a), and the specific experimental characteristic values are detailed in Table 3. SW-2, SW-3, and SW-4 illustrate an increased load-carrying capacity compared to SW-1. The curve for SW-1 is exceptionally smooth, attributed to the absence of flange constraints on both sides of the connection zone, concentrating the failure primarily on the connecting bolts. Although the peak load of SW-2 is lower than that of SW-4, it exhibits superior ductility, with a gradual decline in the failure phase. SW-3 shares similarities with SW-2 in terms of peak load and corresponding displacement, but it experiences a sudden drop in load-carrying capacity during the elastic-plastic phase and a rapid decline in load during the failure phase. SW-4 has the highest peak load, smallest peak displacement, and maximum stiffness. The cracking loads for SW-2 and SW-3 are similar, with increases of 35.42% and 32.91%, respectively, compared to SW-1. SW-4 has the highest cracking load, representing a 56.03% increase over SW-1. In comparison to SW-1, the peak loads for SW-2, SW-3, and SW-4 are increased by 40.17%, 29.19%, and 47.62%, respectively.

Figure 7.

Comparison of data from the specimens.

Table 3.

Main measurement data.

No.	Crack load/kN	Peak load/kN	Peak displacement/mm
SW-1	20.83	37.53	33.41
SW-2	26.19	53.63	32.17
SW-3	25.58	49.15	32.01
SW-4	31.57	55.83	24.63

The comparison of the mean strains on the embedded steel web plates (P1, P2) for the four specimens is depicted in Figure 7(b). Except for SW-1, the shear ribbed slab in the specimens has either yielded or approached yielding. In the case of SW-1, the maximum strain value on the shear ribbed slab is 367.89 με. Due to the absence of flange constraints on both sides of the connection joint, the failure zone is concentrated at the connection joint, resulting in minimal deformation in the lower concrete region. Consequently, the strain values in the lower concrete region for the shear ribbed slab in SW-1 are relatively low. The maximum strain values of SW-2 and SW-3 shear ribbed slabs are similar, which is related to the similarity of their connection joints. In the case of SW-4, the strain on the shear ribbed slab experiences a sudden increase in the mid-phase, possibly attributed to the relatively higher stiffness of the connection joint in SW-4. In the initial stage, with increasing displacement, defects in the joint area gradually adjust to stability with a “creaking” sound. Following the downward shift of the primary deformation zone, the load-bearing capacity of the embedded area increases, leading to a sudden spike in strain on the shear ribbed slab.

The strain variation trends of the shear bolts in the four specimens are similar, as illustrated in Figure 8. It can be observed that the strain values of strain gauges Q1 and Q3 placed on the shear bolts at the ends are relatively higher, while the strain values of strain gauges Q2 and Q4 positioned on both sides of the shear bolts at the tail end are smaller. Considering the presence of the shear rib plates and the tail flange, the shear bolts bear less shear force. Hence, they consistently remain in the elastic stage.

Figure 8.

Strain-displacement curve of shear bolts.

Hysteresis performance test of connecting joints

Specimens of connecting joints

Four specimens were manufactured with steel bar parameters, dimensions, and connection structures identical to the design parameters outlined in Section 3.1. Cyclic loading tests were conducted on the specimens to study the hysteresis performance of the connecting joints. Specimens are named SJ. The wall dimensions are 300 mm high, 400 mm wide, and 200 mm thick, with the joint area cast with filling concrete. As established in the preceding chapter, the shear force borne by the shear studs can be neglected compared to the shear rib plates. Therefore, the shear studs are not considered in the embedded steel of this connection joint.

Loading program

After filling the connection joint with concrete, a quasi-static experimental study under cyclic load is carried out. The loading equipment is shown in Figure 9(a). The specimen is fixed to the foundation using four anchor bolts to resist horizontal shear. The horizontal load is supplied by the MTS hydraulic servo actuator. It is difficult to determine the yield displacement of the specimen, so the interlaminar displacement angle of 0.01 rad (1 mm) is adopted as the control parameter of cyclic loading. The Loading scheme is shown in Figure 9(b). When the load drops below 85% of the peak load, the test is terminated.

Figure 9.

Loading program.

Failure process

In general, the four specimens failed with a similar mode. According to the test records, the failure process of the specimen is shown in Table 4. In the table, I is the initial joint cracks, II is the base diagonal cracks, III is the yield of the rebar, IV is the shedding of concrete, V is the reduction of the slope of base cracks on the base, VI is the extension of base cracks, VII is the Shear failure. The contact interface between the new and old concrete of the specimens in the hysteresis test showed partial cracks, which may be due to construction issues such as incomplete cleaning of the contact interface or insufficient roughness of the surface. Besides, there is no damage at the connection joint. The connection joint maintained its integrity throughout the entire testing process and withstood the applied loads without failure. This indicates that, although the concrete interface needs improvement, the performance of the connection joints meets the necessary strength requirements.

Table 4.

The failure process of specimens.

Specimen	Failure mode
SJ-1	I, III, II, V, VI, VII
SJ-2	I, II, III, V, VI, VII
SJ-3	I, II, IV, V, VI, VII
SJ-4	I, III, II, IV, V, VI, VII

SJ-1 specimen

When the loading displacement was pushed towards 2 mm (1.96 mm), a subtle crack appeared on the front of the connection joint (Figure 10(a)). At the displacement was pulled towards -3 mm (−2.89 mm), a transverse crack appeared on the right side of the connection joint (Figure 10(b)), and the corresponding horizontal load was 7.34 kN. With the loading displacement pushed towards 4 mm (3.32 mm), a significant transverse crack appeared on the back of the connection joint (Figure 10(c)). As pushing further to 3.87 mm, the maximum horizontal load was 27.10 kN, and the reinforcement yielded a strain value of 2156.72 με. When the loading displacement was within a period of 5 mm, fine cracks appeared on the back of the base (Figure 10(d)), which made it difficult to close the cracks during this cycle. As the increase of loading displacement, new cracks spread from both ends to the middle on the left side of the connection joint, while the cracks on the back of the base continued to expand. Under a load with a displacement of 6 mm, the cracks on the back of the base gradually extended inward, which caused damage to the concrete. The load-bearing capacity of the specimen is reduced to 85% of the peak load, and the test ends.

Figure 10.

The failure characteristics of the SJ-1 specimen at different loading stages. (a) The subtle crack on the front of the connection joint. (b) The transverse crack on the right side of the connection joint. (c) The significant transverse crack on the back of the connection joint. (d) The fine cracks on the back of the base.

SJ-2 specimen

When the loading displacement was pushed towards 2 mm (1.89 mm), the first crack appeared on the left side of the connection joint (Figure 11(a)). At 2.89 mm, several upward diagonal cracks appeared at the right end of the front of the connection joint (Figure 11(b)), and there were slight cracks on the left side of the base. When the loading displacement was pushed towards 4 mm (3.92 mm), the maximum horizontal load was 33.40 kN, and the strain generated by the lower rebar was 2099.38με. When the loading displacement was pulled towards -5 mm, a longer crack appeared at the left end of the back of the base (Figure 11(c)), and the diagonal cracks around the connection joint did not change. At 6 mm, the cracks gradually extended on the base and the base suffered serious damage (Figure 11(d)). In the loading with a displacement of 6 mm, it can be observed that the whole concrete began to crack, and the bearing capacity decreased by 15%. The loading was stopped and the test ended.

Figure 11.

The failure characteristics of the SJ-2 specimen at different loading stages. (a) The first crack on the left side of the connection joint. (b) The upward diagonal cracks at the right end of the front of the connection joint. (c) The longer crack appeared at the left end of the back of the base. (d) The base suffered serious damage.

SJ-3 specimen

The failure process of SJ-3 was similar to that of SJ-2. When the loading displacement was pushed towards 2 mm (1.95 mm), a transverse crack appeared on the back of the connection joint (Figure 12(a)), and the horizontal load was 27.83 kN. At −1.92 mm, a transverse crack appeared on the right side of the connection joint (Figure 12(b)), and the horizontal load was −30.1 kN. When the loading displacement was pushed towards 3 mm (2.99 mm), the maximum horizontal load reached 35.60 kN. Under a load with a displacement of 5 mm, an internal extension crack appeared at the right end of the back of the base, and the surrounding concrete was partially destroyed (Figure 12(c)). As the horizontal load continued, the crack on the front of the base continued to extend (Figure 12(d)). The bearing capacity drops to 85% of the peak load, and the test ends.

Figure 12.

The failure characteristics of the SJ-3 specimen at different loading stages. (a) The transverse crack on the back of the connection joint. (b) The transverse crack on the right side of the connection joint. (c) Local destruction of foundation concrete. (d) The crack on the base continued to extend.

SJ-4 specimen

With the loading displacement pushed towards 1 mm (0.96 mm), a microcrack appeared on the back of the connection joint (Figure 13(a)), and the horizontal load was 16.52 kN at this time. There were no other obvious phenomena in the early stage. The maximum horizontal load reached 38.22 kN When the loading displacement was pushed towards 4 mm (3.98 mm). As the loading displacement was pulled towards -4 mm (−3.97 mm), a transverse crack extending inward appeared at the right end of the back of the base (Figure 13(b)), and the horizontal load was −38.26 kN at this time. Under a load with a displacement of -6 mm, the width of the crack on the base significantly increased, and a diagonal crack extended along a 45° direction at the right end (Figure 13(c)). In the loading with a displacement of 6 mm, the right side of the base gradually pulled away from the ground beam, and the inclined crack on the front of the base connected with the crack on the right side of the base (Figure 13(d)). The bearing capacity dropped to less than 85% of the peak load, and the text ended.

Figure 13.

The failure characteristics of the SJ-4 specimen at different loading stages. (a) The microcrack on the back of the connection joint. (b) The transverse crack at the right end of the back of the base. (c) The diagonal crack at the right end. (d) The diagonal crack on the right side is connected with the diagonal crack on the front of the base.

The failure phenomena of the four specimens are similar. In the initial stage of loading displacement, fine cracks appear in the front of the connecting joint, indicating local stress concentration due to poor bonding or microstructural defects at the interface between new and old concrete. As the loading displacement increases, cracks on the back of the base continue to expand, indicating increased stress and damage, ultimately reducing the structure’s load-bearing capacity. In order to achieve better and more realistic damage effects, we will apply this type of joint to the overall shear wall in future experiments.

Hysteretic behavior

The hysteresis curves for the four specimens are depicted in Figure 14. During the initial stages of loading, the specimens exhibit elastic behavior, with a relatively small area enclosed by the hysteresis loop and a substantial initial stiffness. The load-displacement relationship is primarily linear. As the loading displacement increases, microcracks formed in the early stages expand, and the specimens transition into the elastic-plastic phase. During this phase, the residual deformation of the specimens increases, and the stiffness gradually degrades. Compared with specimen SJ-1, other specimens demonstrate higher load-carrying capacity in the early loading stage. Specifically, the load-carrying capacities for SJ-2, SJ-3, and SJ-4 are increased by 23.25%, 31.37%, and 41.40%, respectively, relatively to SJ-1. As internal damage accumulates within the specimens, cracks expand significantly, leading to a rapid degradation in the specimens’ stiffness. After the load surpasses the inherent peak load of the specimens, the curves exhibit a descending segment, ultimately resulting in a decrease in load-bearing capacity to 85% of the peak load.

Figure 14.

Hysteretic curves of the four specimens.

The skeleton curves for the four specimens are illustrated in Figure 15(a), and Table 5 summarizes their seismic characteristic parameters. The results indicate that adding flanges at both ends of the steel enhances the initial stiffness of the specimens. This enhancement is attributed to the flanges restricting the deformation of the connecting steel plates, compelling them to function within a specified range of limitations. In comparison to SJ-1, both the yield load and peak load of SJ-2, SJ-3, and SJ-4 are significantly higher. While SJ-3 exhibits the highest initial stiffness, it enters the failure phase after 3 mm, displaying the least deformation capacity. SJ-4’s yield load is increased by 41.45% compared to SJ-1, and the peak load is elevated by 41.40%. The seismic performance of the SJ-4 connection joint is provided by the shear bearing capacity of the bolt, the tensile strength of the bolt and nut, and the flange at the connection joint. Therefore, compared to other connection joints, the SJ-4 connection joint has better resistance to bending deformation.

Figure 15.

(a) Skeleton curves of the four specimens (b) Stiffness degradation curves of the four specimens. (c) Energy dissipation capacity of the four specimens.

Table 5.

Seismic parameters of the specimens.

No.	SJ-1	SJ-2	SJ-3	SJ-4
Initial stiffness (kN/mm)	9.85	12.17	15.45	15.16
Yield displacement (mm)	2.56	2.37	2.09	2.23
Yield load (kN)	21.24	27.33	31.91	33.63
Peak displacement (mm)	4.12	4.04	3.03	4.02
Peak load (kN)	27.03	33.66	35.70	38.22
Ultimate displacement (mm)	6.04	5.89	5.26	7.39
Ultimate load (kN)	21.8	28.06	28.20	31.43
Ductility factor	2.36	2.49	2.51	3.31

Generally, the experimental results indicate that the SJ-4 connection joint can significantly improve the lateral stiffness and horizontal load-carrying capacity of the wall.

Displacement ductility

In this paper, the displacement ductility coefficient is selected for evaluation. The displacement ductility coefficient is calculated by:

μ = \frac{Δ_{u}}{Δ_{y}}

(1)

Where μ is the average displacement ductility factor; Δ_u is the ultimate displacement of the specimen; Δ_y is the yield displacement of the specimen.

As per Table 5, the ductility factors of SJ-1, SJ-2, SJ-3, and SJ-4 were calculated as 2.36, 2.49, 2.51, and 3.31, respectively. Compared to SJ-1, the ductility factors of SJ-2, SJ-3, and SJ-4 are increased by 5.51%, 6.36%, and 40.25%, respectively. The increase in the ductility factor of SJ-4 is particularly remarkable. Therefore, it can be concluded that SJ-4 exhibits more superior plastic deformation capacity.

Stiffness degradation

Figure 15(b) shows the stiffness degradation curves of the four specimens, where the secant stiffness K_i is adopted. In this paper, K_i is expressed by:

K_{i} = \frac{| + F_{i} | + | - F_{i} |}{| + X_{i} | + | - X_{i} |}

(2)

Where ±F_i is the corresponding load when the displacement reaches the maximum at the time; ±X_i is the corresponding displacement when the displacement reaches the maximum at the time.

The secant stiffness curves of SJ-2, SJ-3, and SJ-4 are all higher than that of SJ-1, indicating that the flanges on both sides can provide significant lateral stiffness to strengthen the connection joint. The most notable increase in initial stiffness is observed in SJ-3 and SJ-4. However, after loading, the curve of SJ-3 descends rapidly, suggesting that the design of this connection joint may be less favorable for seismic resistance, increasing the structural brittleness. In contrast, SJ-4 exhibits a larger initial stiffness with a more gradual degradation of stiffness.

Energy dissipation

The structural energy dissipation capacity is a crucial metric for evaluating a structural or component’s ability to absorb and dissipate seismic energy. Figure 15(c) illustrates the cumulative energy dissipation at each loading stage for the specimens. In the early loading stages, the energy dissipation capacities of the specimens are essentially similar. As the loading displacement increases, the energy dissipation continually rises, with no apparent slowdown in the growth rate. At the loading displacement of 6 mm, SJ-1 exhibits a cumulative energy dissipation of 475.74 kN·mm. In comparison, SJ-4, at the same displacement level, shows a cumulative energy dissipation of 569.98 kN·mm, representing a 19.81% increase over SJ-1. This indicates that in the later stages of displacement loading, the flanges located at both ends of the steel in SJ-4 share the horizontal load, enhancing the specimen’s energy dissipation capacity. Moreover, at this stage, SJ-4 continues to bear the load, and its energy dissipation capacity continues to increase, which indicates that DWCBJ exhibits superior energy dissipation capacity.

Numerical simulation

Finite element model

The finite element model (FEM) of the connection joint is established using ABAQUS, as shown in Figure 16. The concrete walls, joint area steels, connecting steels, and bolts are all modeled by eight-joint hexahedral reduction element (C3D8R). The rebars are built by a three-dimensional truss element (T3D2). The grid size of the concrete walls, steel plates, and rebars is 20 mm, and the steel around the bolt hole and joint area is 5 mm. The bolts are sized with a grid of 2 mm.

Figure 16.

FEM of the connection joint.

The interface contact behaviors between the steel components and concrete components are also considered in the model. For the interface between bolts and connecting steels, the hard contact is used in normal constraints, while a Mohr-Coulomb friction model with a friction coefficient of 0.3 is adopted in the tangential direction. Similarly, the hard contact is used between the interface of the screw stem and the steel, but there is no friction behavior defined in the tangential direction. To simulate the interfacial contact between the pre-embedded steel and concrete, hard contact and the Mohr-Coulomb friction model with a friction coefficient of 0.6 is employed. Subsequently, the normal hard contact and tangential friction with a friction coefficient of 0.8 are defined for the concrete interface. In addition, the contact relationship between the nut and concrete is not considered in the model. Bolt preload determines the shear bearing capacity of the connection, thereby it is the critical parameter for the precast wall to transfer shear force. If the preload is not enough, slippage on the interface and wall uplift may be generated, and thus affect the overall performance of the wall. However, the bearing capacity of the walls may not be influenced by the preload, because the failure mode is dominated by the crush of wall concrete or the facture of anchor components in the connections. During the design, it is ensured that the connectors are not damaged as much as possible. Through the final experiment, it was found that the failure mode of the joint was caused by the bottom concrete, which verified the rationality of the specimen. Therefore, it can be concluded that the bolt preload shows few effects on the bearing capacity of the shear wall.

During the analysis, the bottom of the base is restrained in all degrees of freedom (U1 = U2 = U3 = UR1 = UR2 = UR3 = 0) to simulate the experimental boundary conditions. Four loading steps including three steps for applying bolt preload and one step for applying horizontal displacement loads are defined. Among the analysis steps for bolt preload, steel is constrained to limit out-of-plane deformation. Moreover, the cross-section at the top of the wall is coupled with the reference point RP-1 to apply the displacement loads.

Material parameters

In ABAQUS, the constitutive parameters for steel are input based on the material test data, with the Poisson’s ratio set to 0.3. The bolts employ a bilinear constitutive model, and the steel’s fracture behavior is not accounted for. The Concrete Damaged Plasticity (CDP) model is utilized for the PC walls. The stress-strain relationship of the model is calculated based on the constitutive curve under uniaxial loading specified in the standard, and the damage factor is computed using the Sidoroff energy equivalent theory (Symposium Senlis, France, 1981). The constitutive models of steel and concrete are shown in Figures 17 and 18, respectively. The values for the expansion angle, eccentricity, initial yield strength ratio of biaxial to uniaxial loading (f_b0/f_c0), the ratio of the second stress invariant on the meridional plane to the axial plane (K), and the viscosity parameter are provided in Table 6.

Figure 17.

Concrete stress-strain curves.

Figure 18.

The constitutive curve of steel.

Table 6.

Parameter of the damaged plastic model of concrete.

Values of expansion angle	Eccentricity	f_b0/f_c0	K	Viscosity parameters
30	0.1	1.16	0.667	5 × 10⁻⁴

Validation of the FEM

Figure 19 presents a comparison between the results of finite element analysis and experimental outcomes. The finite element analysis results closely align with the experimental curves, affirming the model’s ability to accurately predict the hysteretic behavior of the specimens. The failure mode for the four specimens is also obtained from the numerical simulation, as depicted in Figure 20. All four specimens experienced failure due to excessive tensile damage in the lower concrete, with no discernible damage on the surface of the filled concrete at the connection joint, which closely match the experimental failure distribution, confirming the validity of the presented finite element modeling approach. From the numerical results, the stress state of the connections can also be obtained. When displacements were applied to 3.98 mm, 4.62 mm, and 3.58 mm, respectively, the embedded steel plates in SJ-1, SJ-2, and SJ-3 initiated yielding. The yielding zone propagated from the inner side of the bolt hole towards the shear rib plates, as depicted in Figure 21(a) and (b), and Figure 21(c). In contrast, SJ-4 remained unyielded throughout the loading process, with the embedded steel plate almost maintaining an elastic state, as illustrated in Figure 21(d). The plastic deformation of the steel plates is undersized, and the structure primarily dissipates energy through the plastic damage of the concrete shear walls. In subsequent research, consider further optimizing the structure in terms of connector deformation. The connecting steel undergoes sufficient plastic deformation under safe conditions, thereby further increasing the energy consumption of the joint through the plastic deformation of the steel.

Figure 19.

Comparison of hysteretic curves of the numerical simulation and test results.

Figure 20.

Failure modes of the specimens by numerical simulation.

Figure 21.

Stress distribution of the embedded steel.

Conclusions

In this paper, a novel assembly-type joint utilizing pre-embedded steel-bolt connections is presented for PC shear walls. Four meticulously designed and crafted specimens underwent shear performance testing, low-cycle reciprocation testing, and finite element model verification for the corresponding connection joint. The test results encompassed various aspects, including failure modes, hysteresis performance, ductility, stiffness degradation, and energy dissipation. Numerical analysis was used to validate the designs of the four distinct connection joints. The main conclusions derived from the study can be drawn:

(1) The shear failure modes of the DWCBJ can be classified into bolt shear failure and pre-embedded steel pullout failure, with significant differences between the two modes: the bolt shear failure mode has a lower load-carrying capacity, primarily due to the absence of flanges, resulting in concentrated failure in the joint area, while the pre-embedded steel pullout failure mode exhibits superior load-bearing capacity. The connection joint with flanges on both sides of the steel experiences a substantial increase in ultimate shear capacity. The flange baffles enable the connection joint’s ultimate moment to exceed the ultimate flexural strength of the concrete end, restricting the relative displacement between the upper and lower parts of the concrete-embedded steel and sharing the shear load of the bolts.

(2) The DWCBJ improves the hysteretic behavior of the connection joint. Notably, specimen SW-3 displays higher stiffness, but with an excessively small maximum failure displacement, rendering it unsuitable for recommendation. SJ-4 boasts the highest ultimate load-carrying capacity, with increases of 41.40%, 13.55%, and 7.06% compared to SJ-1, SJ-2, and SJ-3, respectively. Furthermore, SJ-4 exhibits good ductility, gradual stiffness degradation, and optimal energy dissipation. The ribbed channel steel in SJ-4 provides high lateral stiffness and bearing capacity. The bolts in SJ-4 are subjected to both shear and axial tension, enhancing the stiffness and strength of the joints. Consequently, SJ-4 can maintain a higher load-bearing capacity under larger deformations, thereby improving the overall seismic performance of the structure.

(3) Subject to cyclic loading, did not exhibit significant damage in the midsection of connection joints in all four specimens. In the finite element analysis, the wall-foot elements, connection steel, and bolts were within the elastic range, and the embedded steel did not undergo yielding. This affirms the reliability of the DWCBJ. In addition, further experiments and simulation studies are necessary to verify and optimize the performance of connection joints under various load conditions. To advance the implementation of these findings in design codes or construction practices, the next phase of research will concentrate on performance verification testing of full-scale shear walls under real-world conditions, evaluating the performance of bolted connections in actual working environments, and enhancing the overall safety and reliability of the structure.

Footnotes

Acknowledgments

The authors would like to express their gratitude for the support received from various sources, including the National Natural Science Foundation of China (No.52308309), the Guangdong Basic and Applied Basic Research Foundation (No.2022A1515110191; 2023A1515011914).

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Natural Science Foundation of China (No.52308309), the Guangdong Basic and Applied Basic Research Foundation (No.2022A1515110191; 2023A1515011914).

ORCID iD

Yindong Gong

References

Arandjelovic

Djordjevic

Sedmak

, et al. (2024) Failure analysis of welded joint with multiple defects by extended finite element method and engineering critical analysis. Engineering Failure Analysis 160: 108176. DOI: 10.1016/j.engfailanal.2024.108176.

Cheng

Luo

Cheng

, et al. (2023) Seismic performance of precast concrete walls with grouted sleeve connections using large-diameter bars. Soil Dynamics and Earthquake Engineering 169: 107905. DOI: 10.1016/j.soildyn.2023.107905.

Ding

Xiao

Wei

, et al. (2020) Seismic behavior of concrete shear walls with bolted end-plate DfD connections. Engineering Structures 214: 110610. DOI: 10.1016/j.engstruct.2020.110610.

Amer

Chen

, et al. (2024) Seismic behaviors of CFT-column frame-four-corner bolted connected buckling-restrained steel plate shear walls using ALC/RAC panels. Thin-Walled Structures 195: 111365. DOI: 10.1016/j.tws.2023.111365.

Fan

Tao

, et al. (2022) Seismic behavior of prefabricated steel reinforced concrete shear walls with new type connection mode. Structures 37: 483–503. DOI: 10.1016/j.istruc.2021.10.055.

Guo

Wang

, et al. (2018) Experimental and numerical investigation of jointed self-centering concrete walls with friction connectors. Engineering Structures 161: 192–206. DOI: 10.1016/j.engstruct.2018.02.028.

Guo

Zhai

Cui

, et al. (2019) Seismic performance assessment of low-rise precast wall panel structure with bolt connections. Engineering Structures 181: 562–578. DOI: 10.1016/j.engstruct.2018.12.060.

Han

Wang

Zhang

, et al. (2020) Experimental investigation and simplified stiffness degradation model of precast concrete shear wall with steel connectors. Engineering Structures 220: 110943. DOI: 10.1016/j.engstruct.2020.110943.

Henry

Sritharan

Ingham

(2016) Finite element analysis of the PreWEC self-centering concrete wall system. Engineering Structures 115: 28–41. DOI: 10.1016/j.engstruct.2016.02.029.

10.

Huang

Zhang

Yang

(2018) A comparative study on seismic performance of precast shear walls designed with different variables. KSCE Journal of Civil Engineering 22: 4955–4963. DOI: 10.1007/s12205-016-0765-1.

11.

Ioannis

Mohammad

Iman

, et al. (2022) Cold-formed steel beam-to-column bolted connections for seismic applications. Thin-Walled Structures 172: 108876. DOI: 10.1016/j.tws.2021.108876.

12.

Jiang

Luo

Xue

, et al. (2021) Experimental study on hysteresis behavior of precast concrete double skin shear wall with horizontal loop connection and spiral-confined vertical lap connection. Journal of Building Engineering 45: 103526. DOI: 10.1016/j.jobe.2021.103526.

13.

Jiang

Song

Bian

, et al. (2023) Experimental and numerical study on the shear strength of new bolted connection assembled shear wall. Journal of Building Engineering 69: 106139. DOI: 10.1016/j.jobe.2023.106139.

14.

Gao

Xiang

, et al. (2021) Experimental study of seismic performance of precast shear wall with a new bolt-plate connection joint. Structures 34: 3818–3833. DOI: 10.1016/j.istruc.2021.10.009.

15.

Liu

Chen

, et al. (2021) Seismic behavior of bolted truss-to-column joint with oversized or slotted bolt hole. Engineering Structures 247: 113110. DOI: 10.1016/j.engstruct.2021.113110.

16.

Liu

Jiang

Lin

(2023) Seismic performance of precast concrete shear wall using grouted sleeve connections for section steels reinforced at wall ends. Structures 57: 1050688. DOI: 10.1016/j.istruc.2023.105068.

17.

Fan

Tian

, et al. (2024) Seismic performance analysis of steel frame installed with prefabricated external walls and new connection type. Journal of Building Engineering 84: 108471. DOI: 10.1016/j.jobe.2024.108471.

18.

Pan

Cai

, et al. (2021) An experimental study of the seismic behaviour of precast concrete shear walls with bolted-plate connections. Engineering Structures 248: 113203. DOI: 10.1016/j.engstruct.2021.113203.

19.

Sidoroff

(1981) Description of anisotropic damage application to elasticity. In: Physical Non-linearities in Structural Analysis: Symposium Senlis. Berlin, Heidelberg: Springer Berlin Heidelberg, 237–244, France May 27–30, 1980. DOI: 10.1007/978-3-642-81582-9_35.

20.

Sritharan

Aaleti

Henry

, et al. (2015) Precast concrete wall with end columns (PreWEC) for earthquake resistant design. Earthquake Engineering & Structural Dynamics 44(12): 2075–2092. DOI: 10.1002/eqe.2576.

21.

Sun

Qiu

(2015) Seismic behavior and mechanism analysis of innovative precast shear wall involving vertical joints. Journal of Central South University 22(4): 1536–1547. DOI: 10.1007/s11771-015-2670-z.

22.

Sun

Wang

Sun

, et al. (2023) Parametric study on lateral behavior of totally precast RC shear wall with horizontal bolted connection. Structures 57: 105100. DOI: 10.1016/j.istruc.2023.105100.

23.

Vaghei

Hejazi

Taheri

, et al. (2014) Evaluate performance of precast concrete wall to wall connection. APCBEE procedia 9: 285–290. DOI: 10.1016/j.apcbee.2014.01.051.

24.

Vaghei

Hejazi

Taheri

, et al. (2017) Development of a new connection for precast concrete walls subjected to cyclic loading. Earthquake Engineering and Engineering Vibration 16: 97–117. DOI: 10.1007/s11803-017-0371-3.

25.

Wang

(2018) Seismic performance of precast concrete shear wall structure with improved assembly horizontal wall connections. Bulletin of Earthquake Engineering 16(9): 4133–4158. DOI: 10.1007/s10518-018-0348-2.

26.

Yang

Sun

, et al. (2023) Experimental study on seismic behavior of a new precast shear wall system with angle steel connectors. Structures 52: 30–41. DOI: 10.1016/j.istruc.2023.03.166.

27.

Zhang

, et al. (2024) Research on seismic behavior of precast concrete shear wall with new steel plate-bolted horizontal connection. Journal of Building Engineering 85: 85108719. DOI: 10.1016/j.jobe.2024.108719.

28.

Zhai

Liu

Mercan

, et al. (2024) A hybrid buckling-restrained brace for enhancing the seismic performance of steel moment resisting frames. Soil Dynamics and Earthquake Engineering 178: 108464. DOI: 10.1016/j.soildyn.2024.108464.

29.

Zhi

Guo

Xiao

, et al. (2017) Quasi-static test and strut-and-tie modeling of precast concrete shear walls with grouted lap-spliced connections. Construction and Building Materials 150: 190–203. DOI: 10.1016/j.conbuildmat.2017.05.183.