Experimental Study on Seismic Behavior of Steel Plate Shear Walls with Connections in the Weak Axis of PEC Columns

2.1. The Design of Steel Plate Shear Walls with Connections in the Weak Axis of PEC Columns

2.1.1. Specimen Design

Based on a 1/3 scale model of a 16-story office building and referring to the Chinese “Code for Seismic Design of Buildings” (GB 50011-2010,2016) [17], three specimens of SPSW with connections in the weak axis of PEC columns with varying parameters were designed and numbered PEC-SPSW-2, PEC-SPSW-4, and PEC-SPSW-6. The specimens consisting mainly of edge columns, frame beams, thin steel plates, concrete and other ancillary components, with all components, were connected by welding. The frame beams of the three specimens were H-shaped steel with a cross-sectional size of H100 mm × 100 mm × 6 mm × 8 mm and a length of 1800 mm. PEC-SPSW-6 was welded with stiffeners at the connection between the web of the PEC column and the thin steel plate. The details of the specimens are shown in Fig. (1). The geometric parameters of the specimens are shown in Table 1.

The main difference between PEC-SPSW-2 and PEC-SPSW-4 was the section size of the edge column. Specifically, the section size of PEC-SPSW-2 was 120mm × 120mm × 4mm × 6mm, while the section size of PEC-SPSW-4 was 140mm × 140mm × 4mm × 6mm. Apart from this, all other parameters were the same. Similarly, the primary difference between PEC-SPSW-6 and PEC-SPSW-4 was that PEC-SPSW-6 included stiffening plates at the connection of PEC columns and thin steel plates. Again, all other parameters were identical. The PEC columns of the three specimens were primarily composed of H-shaped steel, concrete, and tie reinforcement.

Additionally, a stiffening plate was pre-installed within the PEC-SPSW-6 column. The design process for the PEC-SPSW-2 and PEC-SPSW-4 columns was consistent, with link reinforcements with a diameter of 6mm welded within the H-shaped steel columns. The distance between the link reinforcements and the edge of the steel column flange was 20mm, and the spacing between the link reinforcements was 60mm. Link reinforcements at the column base and top require denser spacing of 30mm, and a stiffening plate was installed within the steel column at the same height as the beam-column connection. The design process for PEC-SPSW-6 was mostly the same as the previous two specimens, with an additional stiffening plate pre-installed at the connection point between the steel column and thin steel plate, with a spacing of 60mm. The detailed design drawings of the PEC columns for the three specimens are shown in Fig. (2). The slenderness ratio of the PEC column was calculated as follows:

(1)

(2)

(3)

(4)

(5)

is the slenderness ratio of the PEC column, L is the length of the PEC column, I₀ is the converted section moment of inertia, A₀ is the converted sectional area, I_c is the moment of inertia of concrete section, I_a is the section moment of inertia of the steel, E_a is the elastic modulus of the steel, E_c is the elastic modulus of concrete

Table 1.

Number	Cross-sectional Size of PEC /mm	Thickness of Embedded Steel Plate /mm	Strength Grades of Concrete		Whether to Set Stiffeners	Slenderness Ratio of H-shape Steel	Flange Width-thickness Ratio of PEC Columns
PEC-SPSW-2	H120 × 120 × 4 × 6	2	C30		-	27.6	10
PEC-SPSW-4	H140 × 140 × 4 × 6	2	C30		-	23.7	11.67
PEC-SPSW-6	H140 × 140 × 4 × 6	2	C30		stiffener	23.7	11.67

2.2. Experimental Study on Steel Plate Shear Walls with Connections in the Weak Axis of PEC Columns

2.2.1. Material Properties

The steel parts of the specimen were made of Q235 steel, and samples of the steel materials were taken to conduct a tensile test in accordance with the Chinese standard “Metallic materials—Tensile testing at ambient temperature” (GB/T228-2002) [18]. First, it was necessary to prepare the steel specimens and calibrate the testing equipment. The specimen was properly installed onto the testing machine and then uniformly and slowly stretched until pulled apart. The relevant data was recorded. Fig. (3) shows the schematic diagram of the steel specimen and the tensile testing equipment. The material properties of the steel obtained from the test are shown in Table 2. Extra concrete was reserved when pouring PEC columns to make 6 concrete test blocks. These blocks were cured under the same conditions for 28 days and then placed in a hydraulic press with a capacity of 2000kN to conduct a concrete strength test. The concrete components of the specimens were made of C30 concrete, and six standard cube concrete blocks were produced. The compressive strength test of the concrete was carried out according to the Chinese “Code for Construction Quality Acceptance of Concrete Structure Engineering” (GB50204-2002) [19]. The compressive strength of concrete measured by the test is shown in Table 3.

Table 2.

Type	Yield Strength fy/MPa	Ultimate Strength Fu/MPa	Elastic Modulus E/MPa	Elongation δ/%
Column	246	439	206000	24
Beam	241	451	220000	26
Plate	238	423	232700	26

Table 3.

Test Block Number	Ultimate Load (kN)	Cubic Compressive Strength fcu,k(N/mm2)
1	758	33.35
2	742	33.17
3	761	33.80
4	786	35.21
5	769	34.51
6	775	34.72
Average Value	763	34.13

2.2.2. Loading Protocol and Instrument

The quasi-static test was performed using an elector-hydraulic servo-programmed actuator (MTS) that can apply a horizontal force of 1000 kN, as shown in Fig. (4b). The actuator was connected to the loading beam, which transmitted horizontal force to the specimen. The loading beam was connected to the top of the specimen columns with 50 mm diameter pins, and the bottom plate of the specimen was fixedly connected to the reaction floor with high-strength bolts of 60 mm in diameter. The device details are shown in Fig. (4a). To prevent the out-of-plane instability of the specimen during loading as shown in Fig. (4c), an overall scaffold was installed outside the specimen, as shown in Fig. (4d and 4e). A lateral support with a pulley was installed in the web area of the beam to constrain the out-of-plane development of the specimen during loading.

The loading method used for the specimen followed the “Specification of Testing Methods for Earthquake Resistant Building” (JGJT 101-2015, 2015) [20], which utilized a displacement-controlled loading method. The amplitude of the displacement loading is shown in Fig. (5), The specimen on each level was cycled three times, and the interval of the loading was 4mm until the bearing capacity of the specimen significantly decreased. The test was terminated once the specimen was damaged.

2.2.3. Stress Measurement and Analysis

Fig. (6) shows the arrangement of strain gauges at key points on the specimen. During the testing and loading process, these gauges collected data on the strain experienced by the specimen, which was used to calculate the stress. This information was then transmitted to the strain collection device.

Fig. (7) shows the stress value at key positions of PEC-SPSW-2. As the displacement reached 5mm, the stress on the specimen frame was relatively small, but the main diagonal area of the embedded thin steel plate had entered the elastoplastic state due to the lateral force causing shear and was susceptible to buckling deformation. The low-yield steel began to yield, reaching a maximum stress of 294.35 MPa. In addition, most areas of the specimen remained in the elastic stage. At a displacement of 24 mm, the minimum stress of T11-T13, H14, T15 and T16 regions in the main diagonal of the steel plate ranged from a minimum of 201.52 MPa to a maximum of 308.24 MPa, while most of the area outside the main diagonal of the steel plate was in the yield state. The stress on the edge column and the filled concrete of the specimen increased, and the steel at the column foot reached its yield strength. When the maximum stress of the concrete at the top of the column reached 33.28 MPa, the concrete began to peel off. The main frame of the specimen bore the additional tension generated by the tensile position of the thin steel plate, and the PEC column, with high stiffness and strength, provided a good anchoring effect for the steel plate. The stress development law of PEC-SPSW-4 and PEC-SPSW-6 was basically similar to PEC-SPSW-2. These results demonstrate that introducing the PEC column into the SPSW structure can prevent local buckling of the edge column, increase the anchoring effect on the thin steel plate, and effectively improve the seismic performance of the shear wall structure.

3.1. Experimental Phenomenon

The experimental set-up was an electro-hydraulic servo-programmed actuator (MTS) capable of applying a horizontal force of 1000 kN. The test piece loading mode was bidirectional cyclic loading mode.

The loading direction of the test piece and the definition of the edge column: the left column of the test piece was S1, and the right column was S2. The push load was positive, and the pull load was negative. The arrangement of the strain gauge of the test piece is shown in Fig. (8). The strain gauges were carefully positioned and fixed onto the surface of the object being tested in order to accurately measure strain or deformation.

PEC-SPSW-2: During the initial loading period, the components of the specimen were in an elastic state, and the steel plate had a slight bulging defect during the processing, which made it sound, but there was no obvious deformation. When the displacement reached an 8mm cycle, the embedded (Fig. 9) steel plate was in an elastic-plastic state. The above area of the main diagonal appeared to uplift the phenomenon of the plate and gradually developed into a tension band because the top half of the steel plate was close to the action point of the force. When the displacement reached 12 mm cycle, the thin steel plate had formed an obvious tension strip of the main diagonal of the “X-shape” under cyclic load because it will occur breathing effect under cyclic load, and the frame was no phenomenon. When the displacement reached 20 mm, the development of the main diagonal tension strip of the steel plate became more and more obvious, and a small range of tension strip appeared near the main diagonal tension strip. Due to the cyclic load, the steel plate made a violent sound during the push-pull load alternate period, then the concrete outside of the S2 column cracked with the surface peeled off. At this time, the steel plate was in the post-buckling strengthening stage. When the displacement reached 24 mm, the main diagonal tension strip of the steel plate was particularly obvious, and the difference at the maximum deformation was about 10 mm. Thus, the steel plate reached post-buckling strength under the continuous alternation of push-pull loads. There were the phenomena of paint peeling off and the steel plate tearing at the intersection of the main diagonal tension strip of the steel plate because of the continuous alternation of push-pull loads. The embedded steel plate of the specimen finally produced an obvious tension strip. The phenomenon of weld fracture occurred at the connection of the stiffener of the PEC column. In addition, the frame beam had a slight out-of-plane offset, and the concrete cracks of the PEC column gradually increased. When the displacement reached 28 mm, the steel plate generated a huge noise, then the tension strip area of the steel plate increased and was in the yield state. The end flange of the frame beam occurred at the local buckling, and the concrete fragments of the PEC column fell off in large quantities. At this time, the specimen was basically in a failure state, and the bearing capacity no longer increased with the displacement. The ultimate bearing capacity of the specimen was 470 kN.

During the elastic stage of the first two loading cycles, the steel plate exhibited slight bulging and produced a faint sound due to the initial defect. As the specimen was loaded to 4mm in the third cycle, the sound from the steel plate gradually increased. By the time the displacement reached 8mm, a small primary diagonal tension strip had formed. When the displacement reached 12mm, the primary diagonal tension strip on the steel plate became significant, and several secondary tension strips appeared. As the displacement increased to 16mm, the tension strip on the steel plate expanded, and cracks appeared on the concrete at the top of the PEC column. Some concrete had even fallen from the contact area between the concrete and steel column. Although the concrete cracks continued to develop as the displacement reached 20mm, there was no obvious deformation on the PEC column. However, at 28mm, the end flange of the frame beam buckled locally, and there was a slight out-of-plane displacement. Concrete debris fell off from the PEC column, and the load-bearing capacity of the structure began to decrease. The test was terminated, and the ultimate bearing capacity of the specimen was determined to be 468KN. The test phenomenon is shown in Fig. (10).

PEC-SPSW-6: PEC-SPSW-6 exhibited similar phenomena to PEC-SPSW-4 and PEC-SPSW-2. The initial behavior was the same until a displacement of 12mm was reached. At 16mm, the main diagonal tension strip began to develop and enlarge, and concrete started to fall off at the contact point between the concrete and steel column. When the displacement reached 20mm, the main diagonal tension strip alternated, and the paint peeled off in the joint area of the tension strip. Diagonal cracks appeared in the concrete at the base of the column. As the displacement increased to 24mm, the frame beam experienced out-of-plane displacement, and numerous cracks appeared in the concrete. The thin steel plate exhibited large-scale yielding, and multiple secondary tension strips appeared. Concrete debris fell off, and the load capacity of the specimen began to decline until the test was terminated. The test phenomenon is shown in Fig. (11).

3.2. Failure Mode

The failure modes of the three specimens under cyclic loading were similar. As shown in Fig. (12a), the displacement of the specimen reached 5.6mm (the value of the drift angle is 1/250), and the entire structure was in an elastic state with no obvious deformation, but the embedded thin steel plate exhibited slight buckling. As shown in Fig. (12b), the displacement of the specimen reached 15mm (the value of the drift angle is 1/100), the thin steel plate entered the elastoplastic state, forming an obvious tension strip, while the edge frame remained in the elastic state without significant change. As shown in Fig. (12c), as the displacement of the specimen reached 24 mm, most of the structure entered the plastic state, and the thin steel plate formed multiple “X-shaped” tensile strips. The steel plate folded and exhibited slight tearing at the intersection of the tensile strips under push-pull load. As shown in Fig. (12d), the displacement of the specimen reached 28 mm, the foot and top area of the PEC column began to yield concrete cracks, and some concrete fragments fell. The end and mid-span of the frame beam showed obvious bending deformation because the fragments were falling. The end and mid-span of the frame beam showed obvious bending deformation due to the increase in the flange width-to-thickness ratio of the PEC column, resulting in increased stiffness of the edge column and mismatched stiffness of the beam, column, and embedded thin steel plate.

3.3. Hysteresis Curve

The energy dissipation coefficient E was calculated by the hysteretic curve of the specimen under cyclic load, and it reflects the energy dissipation of the structure. The expression for the energy dissipation coefficient expression is as follows:

(6)

As shown in Fig. (13), S_ABC+CDA is the area enclosed of the hysteresis loop and S is the area of the OBE and ODF triangles.

Fig. (14) displays the hysteresis curves of three specimens, namely PEC-SPSW-2, PEC-SPSW-4 and PEC-SPSW-6. During the loading process, the hysteresis curve of all three specimens was fusiform and followed a similar trend. The bearing capacity of the specimen did not decrease significantly. Except for PEC-SPSW-2, which experienced a significant decrease in stiffness and bearing capacity due to the formation of tension strips and cracks at 24 mm, ultimately resulted in failure at 28 mm. As a result, the specimen was destroyed. In comparison, the hysteresis curve of PEC-SPSW-4 was fuller, and the stiffness and bearing capacity were significantly improved. And the bearing capacity was significantly improved. This could be attributed to the increased flange width-to-thickness ratio of the PEC column, which enhanced the anchoring effect of the edge frame on the thin steel plate, allowing the post-buckling strength of the steel plate to play a more significant role. The stiffness and bearing capacity of the specimen did not decrease significantly under the push load, it began to decrease when the displacement of the tensile load reached 28 mm, and the tensile and compressive loads were basically the same. The mechanical properties of PEC-SPSW-4 were better than those of PEC-SPSW-2 because PEC-SPSW-4 increased the flange width-to-thickness ratio of PEC columns, while the increase of the flange width-to-thickness ratio of PEC column increased the stiffness and strength of the edge column and enhanced the anchoring effect of the edge frame on the thin steel plate that makes the post-buckling strength of the steel plate play a significant role. Thus tension strip distribution of the thin steel plate was more uniform. Observing the hysteresis curve of PEC-SPSW-6, it was found that the initial stiffness of the specimen was significantly improved, but the “pinch” in the curve was more evident. The overall bearing capacity of the structure was greatly improved because the PEC-SPSW-6 had stiffeners at the connection between the PEC column web and the embedded thin steel plate. The setting of stiffeners improved the stiffness of the edge column and the restraint effect on the end of the thin steel plate and increased the initial stiffness of the structure. The stiffness of the edge column significantly improved, but the stiffness of the frame beam did not increase, resulting in a lower stiffness of frame beam. Finally, there was a serious mismatch in beam-column stiffness. The frame beam could not withstand the load change, it caused the out-of-plane instability prematurely during the loading process, the structure failed prematurely, and the energy dissipation could not be fully utilized.

3.4. Skeleton Curve and Stiffness Analysis

As shown in Fig. (15), the skeleton curves of the three specimens exhibited similar development trends, and the push-pull loads were nearly identical. However, there were significant differences in their mechanical properties:

(1) PEC-SPSW-2 had a yield displacement of 7.92 mm and a yield load of 240.65 kN. The load reached a peak of 344.38 kN at a displacement of 15 mm and then gradually decreased until the specimen failed at 28 mm. The stiffness and bearing capacity of this specimen were significantly lower than those of the other two specimens due to the formation of tension strips and cracks in the steel plate at 24 mm displacement.

(2) PEC-SPSW-4 exhibited a significantly higher initial stiffness and ultimate bearing capacity than PEC-SPSW-2. Its ultimate bearing capacity was 404.43 kN, 17.4% higher than that of PEC-SPSW-2. This improvement was mainly attributed to the increasing flange width-thickness ratio of the PEC column, which enhanced the stiffness and strength of the edge column and improved the anchoring effect of the edge frame on the thin steel plate. Thus, the post-buckling strength of the steel plate was fully utilized, resulting in more uniform distribution of the tension strips.

(3) The ultimate bearing capacity of PEC-SPSW-6 is 450.09 kN, which was 11.3% higher than that of PEC-SPSW-4. It showed that the stiffeners were set at the connection between the web of the PEC column and the embedded thin steel plate, which effectively avoided the local buckling of the thin steel plate due to stress concentration, enhanced the integrity of the steel plate with the edge column, and significantly improved the initial stiffness and bearing capacity of the structure. However, the stiffness of the frame beam did not increase, resulting in a lower stiffness of the frame beam, which caused premature out-of-plane instability during the loading process, leading to premature failure of the structure and incomplete utilization of energy dissipation.

Fig. (16) shows the stiffness degradation curve of three specimens; they all exhibited a similar pattern of stiffness degradation. During the initial loading period, the PEC column provided sufficient stiffness for the structure, and the stiffness of the specimen increased. In the middle stage of loading, the steel plate had a tension strip, and the PEC column needed to bear the additional tension generated by the horizontal force and the post-buckling strength of the thin plate. At this time, the PEC column began to yield locally, and the stiffness of the specimen began to degrade. At the end of the specimen loading, most areas of the structure into plasticity; thus, the stiffness degradation was more serious. Comparing the stiffness of the three specimens, it was found that the larger the width-thickness ratio of the flange of the PEC column, the greater the stiffness of the structure.

3.5. Energy Dissipation Performance

The energy dissipation coefficient of the three specimens is shown in Table 4. Each specimen had a large area enclosed of the hysteresis loop and good energy dissipation. While comparing PEC-SPSW-2 and PEC-SPSW-4, it was found that the larger the width-to-thickness ratio of the flange of the PEC column, the larger the area enclosed of the hysteresis loop and the energy dissipation coefficient at 12 mm and 24 mm cycles. The area enclosed of the hysteresis curve of the PEC-SPSW-4 at 24 mm was about 16% higher than that of the PEC-SPSW-2, indicating that when the flange width-thickness ratio of the PEC column increased, the stiffness and bearing capacity of the structure also increased, which enhanced the anchoring effect of the edge column on the embedded steel plate. The post-buckling strength of the thin plate was more fully exerted. The PEC-SPSW-6 had stiffeners at the connection between the flange of the PEC column and the steel plate; thus that the area enclosed of the hysteresis curve was about 15% higher than that of PEC-SPSW-4, indicating that the connection between the weak axis of the PEC column and the thin plate was set and the stiffeners could effectively improve the energy dissipation of the structure.