Experimental study on seismic behavior of precast wall panels with concealed bracings at both sides

Abstract:A new prefabricated wall panel with concealed bracings is proposed for multistory and highrise fabricated steel frame structures. Six 2/3 scale prefabricated wall panels with concealed bracings are tested under cyclic horizontal shear loading. Several typical failure models are analyzed. The test results show that the prefabricated wall panels with concealed bracings have good ductility and energy dissipation capacity. Decreasing the shear span ratio and increasing the wall thickness can effectively improve the shear capacity of prefabricated wall panels, but its peak displacement will be reduced to 1/2 of the original. The shear capacity of prefabricated wall panels can be effectively improved by setting concealed column frames. The upper connection of specimens, the lower weld between T-shaped pieces and concealed bracings are weak positions, so they should be strengthened locally. The upper connection should adopt oblong hole structure to prevent bolt shear failure. The results of finite element parametric analysis show that increasing concrete strength has little effect on the peak load of precast wallboard, but using concrete above C30 can effectively increase the yield load; when the steel content of steel support is the same, increasing the section width has more significant effect on the shear capacity of precast wallboard than increasing its section thickness.

Keywords:prefabricated wall panels with two sides connected; concealed bracings;quasi-static loading test; finite element analysis; seismic behavior

Steel frame structure has the advantages of assemblyand short construction period, but it also has weak lateral stiffness and large deformation under horizontal load. In design, lateral force resistance system is often added to meet the actual engineering requirements^[1]. As an excellent lateral force resistant system, prefabricated wall panels have been widely used in the seismic resistance and reinforcement of steel frame structures. However, their connection with the main structure and the internal skeleton construction method will directly affect the seismic performance of the structural system. Therefore, how to design the internal structure and connection mode of prefabricated wall reasonably plays a vital role in the safety performance and popularization of prefabricated steel frame structure.

Traditional prefabricated wall panels and steel frame columns are mostly connected with four sides, but a large number of experiments and theoretical analysis show that this kind of connection will cause columns to bear large bending moments, and may cause premature instability of columns, affecting the overall seismic performance of the structure^[2]. On this basis, scholars at home and abroad have carried out in-depth research on the connection mode of prefabricated wall panels. American scholars XUE and LU^[3]have proposed a two-side connection between prefabricated wall panels and upper and lower beams, and have carried out in-depth analysis through experimental and theoretical methods. The results show that the two-side connection can effectively transfer loads, avoid excessive bending moments of columns, and has good ductility and energy dissipation performance. Based on relevant foreign research results, Guo Lanhui et al.^[4]and Zhou Tianhua et al.^[5]analyzed the detailed structure of the connection between the two sides and proposed a series of beneficial improvement measures.

Combined with the two sides of the connection, domestic and foreign scholars have carried out research on the performance of steel plate shear wall considering the influence of this kind of connection. Steel plate shear wall is adopted as lateral force resisting system in steel frame structure, its stability problem is prominent, and it is often necessary to avoid premature instability of steel plate by setting stiffeners or increasing plate thickness, which makes steel plate shear wall complex and economical in actual processing^[6-7]. Therefore, on this basis, Guo Lanhui et al.^[8], Li Guoqiang et al.^[9], LI et al.^[10], JIN et al.^[11]proposed buckling-resistant steel plate shear walls constrained by two concrete slabs, which effectively solved the premature instability phenomenon of traditional steel plate shear walls. However, in order to ensure the coordinated working performance of built-in steel plates and concrete constraint plates, complex structures need to be set for steel plates, which makes the production process complex and is not conducive to the streamlined operation of prefabricated components.

Therefore, this paper designs a new type of concealed braced precast wall plate suitable for multi-story and high-rise fabricated steel frame structure based on the existing research results of composite steel plate shear wall with two sides connected, which has the characteristics of simple manufacturing process and easy installation. Six precast wall panels were designed and tested under cyclic loading to analyze the seismic performance of shear capacity, ductility and energy dissipation capacity. By establishing refined finite element model, parametric analysis is carried out to verify the accuracy of test results, and then theoretical guidance suggestions are provided for the application of this kind of wall plate in practical engineering.

1Test overview

1.1Specimen design and fabrication

According to the wall thickness (b_w), shear span ratio (λ), concrete type, side column structure and connection structure, 6 prefabricated wall panels with concealed bracing with scale ratio of 2:3 are designed, numbered SJ-1 to SJ-6, and their main design parameters are shown in Table 1.

Table 1 Parameters of specimen components

With reference to common engineering examples, the height of prefabricated wallboard is set as 2000mm, steel supports are made of steel plates with width of 50mm and thickness of 10mm, the section specifications of hidden columns are [63×40×48, I50×50×5×5] respectively, and double-layer bidirectional structural steel mesh is set in the slab. 5 or 9 Φ 16 shear bolts of 4.6 grade with length and spacing of 100mm are set according to different shear span ratio. In order to quickly and accurately position and install prefabricated wallboard, weld a vertical positioning welding piece with the same length as the width of wallboard at the upper beam and the lower beam respectively. After the wallboard is installed and positioned, use 12.9-grade M24 bolts to connect the connecting angle steel, embedded T piece and vertical positioning welding plate. At the same time, connecting angle steel and loading frame beam shall be connected by 12.9 grade M20 bolts, among which the upper T-shaped piece and angle steel connecting bolt hole shall be of oblong hole structure (test piece SJ-5 shall be of ordinary round hole), and the lower part shall be of ordinary round hole. The node plate size of each test piece, the weldment and the arrangement mode of connecting angle steel openings shall be consistent, and the detailed structure is shown in Figure 1.

In order to reduce the overall weight of wallboard, non-sintered fly ash ceramsite is used instead of traditional coarse aggregate, and the proportion design of C10 ceramsite concrete is carried out by bulk density method. The whole test process was adapted twice, and finally the ceramsite concrete proportion scheme was formed, as shown in Table 2.

Q345 steel shall be selected for steel support and concealed column in the specimen, and the concrete strength inspection and evaluation standard shall be adopted according to the specification.(GB50107 -2010) Metallic tensile test specimen (GB6397 -86), Sampling position and specimen preparation for mechanical property test of steel and steel products (GB/T 2975-2018) Material property test analysis is carried out on key parts of prefabricated wall panels, and the mechanical properties of concrete are obtained as shown in Table 3, and the mechanical properties of steel bars and steel plates are shown in Table 4.

Table 2 Composition of concrete materials

Table 3 Mechanical Properties of Concrete

Table 3 Mechanical properties of concrete

表4 钢筋及钢板力学性能

Table 4 Mechanical properties of steel bars and plates

1.2 加载方案与测点布置

The test was completed at the State Key Laboratory of Subtropical Building Science of South China University of Technology. The test pieces and loading device were installed on site as shown in Figure 2. In order to evaluate the shear resistance of the prefabricated wallboard and its connection performance with the steel frame, this test used shear loading. Frame, set the four points of the frame as hinged joints to ensure that horizontal shear is transmitted to the prefabricated wallboard, while restricting the rigid body rotation and vertical movement of the wallboard.

Fig. 2 Installation diagram of test piece

Considering the difference between positive and negative sides in the wallboard test, in order to record and describe the failure of the specimen, the side of the wallboard close to the loading section is the west side, and the side far away from the loading section is the east side. The horizontal displacement of the top of the precast wall panel is measured by placing a horizontal displacement meter on the upper east side of the precast wall panel. At the same time, one-way strain gages are arranged on the steel skeleton and reinforcement to monitor the deformation of key parts of prefabricated wall panels, and their positions are shown in Figure3a-b.

(a)Arrangement of strain measuring points of steel mesh	(b)Arrangement of strain measuring points of steel skeleton
Fig. 3 Measuring points of strain displacement of specimen

According to Code for Seismic Test of Buildings (JGJ/T 101-2015)^[12], horizontal force and displacement mixed control loading mode is adopted. The loading process can be divided into two stages: in the first stage, the horizontal force control is used to carry out low-cycle reciprocating loading (one load cycle per stage) before the concrete cracks occur, and the loading is transferred to the second stage; in the second stage, the displacement control is used to load (two load cycles per stage displacement), and the concrete cracking displacement Δ y is takenas the incremental step until the bearing capacity of the specimen decreases to 85% of the peak load. The test loading regime is shown in Figure 4.

Figure 4 Cyclic loading regime

2Test results and analysis

2.1Experimental phenomena

When the test load reaches 2Δ_y, only the specimen SJ-2 and SJ-4 begin to crack along the steel support direction; when the load reaches 4Δ_y, the other specimens appear inclined cracks along the steel support direction; with the increasing of the test load displacement, the load is finally stopped because the bearing capacity of the specimen decreases greatly or the connection failure occurs. Three failure modes were observed: buckling of steel brace, fracture of steel brace and lower connection of concealed column, shear failure of high strength bolt.

(1) Buckling failure of steel brace (specimen SJ-1 and SJ-3): the concrete at the foot of the west side is crushed first and exits from work. With the increase of test load displacement, the concrete at the bottom is peeled off in a large area. The external constraint of steel brace by concrete is gradually reduced, resulting in the gradual increase of out-of-plane deformation. Finally, the specimen fails due to buckling of steel brace, as shown in Figure 5a.

(2) Fracture of steel support and lower connection of hidden column (specimen SJ-2, SJ-4 and SJ-6): cracks distributed symmetrically in the center gradually appear on both sides of the middle and lower parts of the specimen with increasing loading displacement, without penetrating cracks, and spalling of lower concrete occurs. Since deformation of steel support is restricted by hidden column, no obvious out-of-plane deformation occurs. The specimen finally fails due to fracture of lower connection of steel support and hidden column, as shown in Figure 5b.

(3) Shear failure of high-strength bolts (specimen SJ-5): When the upper part of the specimen is connected by angle steel with circular hole structure, it is difficult to release the deformation of the wall during loading, and then a large vertical shear force is formed, which causes the connection failure due to the shearing of lower bolts before the formation of through-cracks in prefabricated wall panels, as shown in Figure 5c.

(a) Buckling of steel braces
(b) Fracture of steel support and lower connection of hidden column
(c) Shear failure of high strength bolts
Fig. 5 Failure mode of each specimen of precast wall panel with concealed support

2.2Hysteresis curve

Take the thrust generated in the loading section as the positive loading direction and the tension as the negative loading direction, and the horizontal load-displacement (P-Δ) hysteresis curve of each specimen is shown in Figure 6. Before concrete cracking, hysteretic curves of each specimen basically present linear development. With the increase of force and displacement, cracks begin to appear, hysteretic curve area gradually increases, the structure shows a certain energy dissipation capacity, and with the continuous development and crushing of concrete cracks, hysteretic curve appears pinch effect. The shear capacity of precast wall panels rapidly adds up to less than 85% of the peak capacity due to buckling of supports or fracture of hidden columns. Among them, the shear capacity of the specimens without concealed column (SJ-1 and SJ-3) will suddenly decrease to less than 100kN when the support buckling occurs, and lose the ability to continue bearing; while the specimens with concealed column (SJ-2, SJ-4 and SJ-6) will still have a certain horizontal load although the shear capacity decreases to less than 85% of the peak capacity when the concealed column is broken or the lower connection weld is broken due to the incomplete spalling of concrete and the failure of steel support. The specimen SJ-5 does not appear obvious descending section, mainly because its upper connecting angle steel adopts circular hole structure, resulting in shear failure of high strength bolts, while the prefabricated wall plate itself does not appear through cracks, still has bearing capacity.

(a) Specimen SJ-1	(b) Specimen SJ-2
(c) Specimen SJ-3	(d) Specimen SJ-4
(e) Test piece SJ-5	(f) Test piece SJ-6
Figure 6 Test load-displacement curve

2.3Shear capacity and ductility analysis

Fig. 7 Skeleton curve of specimen

The skeleton curve of each specimen is shown in Figure 7. According to themethod for determining yield point in Literature [13], yield displacementΔ_y, yield loadP_y, peak displacementΔ_max, peak loadP_max, limit displacementΔ_u, limit loadP_uand ductility coefficientμof each specimen are obtained by equivalent energy method, as shown in Table 5. The ductility coefficient of specimen is the ratio of limit displacement to yield displacement.

It can be seen from the skeleton curve in Figure 7 and the calculation results of shear capacity in Table 5 that the initial stiffness of the specimens (SJ-3 to SJ-6) with concrete strength of C30 has little change; the initial stiffness of the specimens (SJ-1 and SJ-2) with concrete strength of C10 is less than that of the specimens with concrete strength of C30. By comparing the specimen SJ-4 with SJ-5, it is found that the peak displacement decreases to about 1/2 of the original value, but the peak load increases by 16% when the shear span ratio decreases and the wall thickness increases. The peak bearing capacity of the specimen is obviously increased by the concealed column, and the maximum is 57%. The shear capacity of precast wall panels without hidden columns decreases greatly. The peak displacements of all specimens are greater than 1% of the interstory displacement angle, which meets the requirement of displacement limit (1/250) of multi-story and high-rise steel structures in China.

Table 5 Measured values of characteristic points of skeleton curve

2.4Strain analysis

(a) Specimen SJ-1	(b) Specimen SJ-3
(c) Test piece SJ-5	(d) Test piece SJ-6
Figure 8 Horizontal load-strain hysteresis curve of specimen

In this paper, the load-bearing mechanism of precast wall panels with different structures is clarified through the strain of key parts and different failure modes of each specimen. Select the strain of lower steel support on the west side of test pieces SJ-1, SJ-3 and SJ-5 and the strain of lower hidden column on the east side of SJ-6 for analysis, corresponding to the hysteresis curve of horizontal load-strain (P-ε), as shown in Figure 8.

(1) According to Fig. 8a-b, steel brace is in elastic stage before reaching yield load without hidden column specimen, and its hysteretic curve basically develops linearly, but completely opposite deformation occurs in middle and lower parts, indicating that steel brace and lower gusset plate are mainly compression-shear deformation, and gusset plate deformation is prior to steel brace; After the specimen reaches the yield load, the joint between steel brace and gusset plate has yielded, and the middle part of steel brace enters the yield state with the gradual increase of force and displacement, which is consistent with the buckling failure phenomenon of steel brace at the foot when specimen SJ-1 and SJ-3 fail.

(2) It can be seen from Figure 8c-d that the horizontal load-strain hysteretic curve of the lower hidden column on the east side and the lower steel support on the west side basically develops linearly before the yield load is reached, indicating that it is in the elastic stage; with the test loading to the peak load stage, the deformation of the steel support and hidden column increases continuously and enters the plastic stage. When the welding strength of the lower part of the specimen is insufficient, the lower connection will break, indicating that it should be locally strengthened compared with the traditional welding design.

(3) By comparing Figure 8b-c, it can be seen that when the specimen is provided with hidden column structure, it will restrict the steel support and lower connection, and it will still be in elastic stage before the specimen reaches the peak load stage, and it has certain bearing capacity; however, if the specimen is not provided with hidden column structure, it will lead to premature buckling deformation of steel support and gusset plate, resulting in reduction of shear capacity of prefabricated wall panel.

2.5Energy consumption capacity

Refer to Code for Seismic Test of Buildings (JGJ/T 101-2015) and take equivalent damping coefficient eta eq under cyclic load of each levelas reference index. The calculation results are shown in Figure 9. The equivalent damping coefficients of specimens SJ-1 and SJ-2 (prefabricated ceramsite concrete wall panels) are greater than 0.2, and the maximum is 0.23. The equivalent damping coefficients of specimens SJ-3 to SJ-6 (prefabricated commercial concrete wall panels) are between 0.14 and 0.17. The results show that the energy dissipation performance of prefabricated wall panels made of ceramsite concrete is better. Compared with theequivalent damping coefficient of RC wall plate (0.082-0.115) in reference [1] andequivalent damping coefficient of steel plate shear wall (0.272-0.287) in reference [14], the precast wall plate with concealed bracings proposed in this paper is in the middle, and the precast ceramsite concrete wall plate is close to the steel plate shear wall plate, which indicates that the specimen designed in this paper has good energy dissipation performance.

Figure 9 Equivalent damping coefficient curve of each specimen

3Finite element analysis

3.1Establishment of finite element model and verification of results

ABAQUS software is used to establish a refined finite element model of precast wall panels and analyze the internal force variation of main components in different stages. In the model, T3D2 element is used to simulate the structural reinforcement, and C3DR8 element is used to model the other members. The constitutive model recommended in Code for Design of Concrete Structures (GB50010 -2012)^[15]is adopted for commercial concrete, the constitutive model in Reference [16] is adopted for ceramsite concrete, and the constitutive model in Reference [17] is adopted for Q345 steel, high strength bolt and HPB300 steel bar.

In this paper, the accuracy of finite element model is verified by analyzing the stress and plastic strain development of specimen at different stages and comparing it with experimental results. Take specimen SJ-6 as an example, plastic strain of each component at different stages is shown in Figure 10. From the beginning of loading to the cracking displacement stage of concrete, the steel mesh and embedded steel skeleton are in the elastic stage, and the concrete part first produces cracks at the west foot; in the yield stage, the plastic strain range of concrete expands, the cracks continue to expand, and the steel mesh and steel skeleton begin to produce plastic strain; with the increasing of loading displacement, the lower concrete is crushed out of work, and the steel skeleton, as the main load-bearing component, produces the maximum plastic strain at the west foot. According to the above analysis results and plastic development process of different components in Figure 10, it can be seen that the force transmission mechanism and failure process of each component of the finite element model are consistent with the test results.

(a) Stress and plastic strain nephogram of different components in concrete cracking stage

(b) stress and plastic strain nephogram of different components at yield stage

(c) stress and plastic strain nephogram of different components at peak stage

Figure 10 Stress and plastic strain nephogram of each component in different stages of specimen SJ-6

Figure11shows the comparative analysis results of skeleton curves between finite element simulation and test measurement. It can be seen from the figure that the finite element simulation results of test pieces SJ-3 to SJ-6 are basically consistent with the test measurement values, which can reflect the rigidity and strength variation law of the structure. The finite element simulation results of specimen SJ-1 and SJ-2 are larger than the experimental values, mainly due to the addition of foam in C10 ceramsite concrete, which makes the interface bond force between steel skeleton and concrete worse, while the finite element simulation directly adopts "Embedded" contact, which magnifies the interface bond force between steel skeleton and concrete. Therefore, in order to ensure the accuracy of the subsequent finite element parametric study, the concrete part is analyzed with commercial concrete.

(a) Specimen SJ-1	(b) Specimen SJ-2
(c) Specimen SJ-3	(d) Specimen SJ-4
(e) Test piece SJ-5	(f) Test piece SJ-6
Fig. 11 Skeleton curve checking

3.2Parametric analysis of finite element models

In this paper, the specimen SJ-5 is taken as the basic model, and the concrete strength, steel brace thickness and steel brace width are selected as the variable parameters (as shown in Table 6) to explore the influence of each parameter on the seismic performance of the wall panel. Among them, the parameters of the foundation model are C30 commercial concrete, the steel strength of channel steel side column and embedded steel skeleton is Q345, the width of steel support is 50mm and the thickness is 10mm, and the upper angle steel of the wall plate is connected with the loading beam through the oblong hole structure.

Table 6 Finite element parametric analysis model

Table 6 Details of finite element parametric analysis

3.2.1Concrete strength

By changing the concrete material parameters in the finite element model, the variation law of yield load and peak load of different specimens is obtained, as shown in Figure 12. When C10 commercial concrete is used, the shear capacity of precast wall panel is increased by 2.27 times than that of pure steel frame specimen, which indicates that the combination of concrete and embedded steel frame can greatly improve the shear capacity of precast wall panel. When the concrete strength is from C10 to C30, the yield load of each grade concrete specimen changes little (within 2%). When the concrete strength is above C40, the yield load of each grade concrete increases by 11.6%, 7.2% and 11.7% respectively. For the peak load of specimen, the variation range of concrete strength of each grade is about 5%. In order to give full play to the load-bearing performance of precast wall panels, it is suggested that concrete above C30 grade should be used in practical projects.

Fig. 12 Relationship between concrete strength and shear capacity

3.2.3Steel support plate thickness

By changing the steel brace thickness parameter in the finite element model, the influence of the steel brace thickness parameter on the seismic performance of prefabricated wall panels is clarified, and the variation law of yield load and peak load of different specimens is obtained, as shown in Figure 13. When the thickness of concealed braced steel plate is 4mm, the yield load and peak load increase by 1.25 and 1.28 times respectively compared with plain concrete wall plate. In the yield stage, the thickness of the concealed support steel plate is 10mm as the cut-off point. When the thickness is greater than 10mm, the yield load of each step increases by more than 15%. When the thickness is less than 10mm, the yield strength of each step increases by less than 5%. At the same time, the peak load of precast wall panels is increased by 5%-10% with the increase of thickness of each stage of concealed bracing steel plates.

Fig. 14 Relation between thickness of steel brace and shear capacity

3.2.4Steel support plate width

Based on the finite element simulation results of different steel support plate thicknesses, the yield load and peak load variation law of different specimens are obtained by changing the steel support width parameter under the condition of the same steel consumption, as shown in Figure 14. When the width of concealed braced steel plate is larger than 50 mm, the yield load increases by more than 15%, but the peak shear capacity increases by about 5%. When the thickness and width of steel plate change, the effect of increasing the width of steel plate on the shear capacity is more significant.

Figure 15 Relationship between steel support width and shear capacity

4Conclusions

(1) The failure modes of prefabricated wall panels with concealed bracings on both sides include buckling of steel bracings, fracture of lower connections of steel bracings and concealed columns, shear failure of high strength bolts.

(2) The equivalent viscous damping coefficient of precast wall panels with concealed bracings at both sides is between 0.082-0.115 and 0.272-0.287, which shows excellent energy dissipation performance; the peak displacement is greater than 1%, which meets the limit requirements in the code and has good ductility characteristics.

(3) Ceramsite concrete can enhance the energy dissipation capacity of prefabricated wall panels; setting hidden columns can effectively improve the bearing capacity of prefabricated wall panels and prevent buckling failure of steel supports.

(4) With the increase of concrete strength grade, taking C30 as the cut-off point, the yield load of precast wallboard changes gradually at first and then increases significantly; when the steel content of steel supporting plate is the same, the increase of width is more significant than that of thickness.