Structures 53 (2023) 749–763

Contents lists available at ScienceDirect

Structures
journal homepage: www.elsevier.com/locate/structures

Parametric study on the partially interconnected steel plate shear walls

with stiffeners
Nima Paslar a, Alireza Farzampour b, *, Navid Chalangaran a
a
Department of Civil Engineering, Islamic Azad University of Qeshm, Qeshm Island, Iran
b
Department of Civil and Environmental Engineering, Virginia Tech, Blacksburg, VA, USA

A R T I C L E I N F O A B S T R A C T

Keywords: Steel plate shear walls are a popular and effective lateral load-resisting system used in high-rise buildings.
Steel plate shear wall However, limited interconnection between the infill plate and boundary elements is recommended to avoid
Infill plate interconnection direct demands on the boundary elements and brittle modes of behavior. But previous studies have shown that
Boundary stiffeners
the absence of interconnection between infill plates and columns in steel plate shear walls with beam-connected
Energy dissipation
Finite element modeling
systems can reduce the system’s performance significantly. This study proposes procedures to improve the
performance of steel plate shear walls with limited infill plate interconnections. Sixty-two parametric compu­
tational models were investigated with and without boundary infill plate stiffeners and different infill plate
widths to find efficient procedures for eliminating the lack of interconnections. Results show that utilizing
boundary stiffeners can increase ultimate strength, energy dissipation, and stiffness by 15%, 20%, and 24% on
average. The width of the infill plate and boundary stiffeners are identified as key factors in the system’s per­
formance. Reducing the infill plate width without using stiffeners exacerbates the performance of the lateral
resisting system, resulting in a reduction of ultimate strength, energy absorption, and stiffness up to 17%, 24%,
and 56%, respectively. Analytical equations are proposed and verified to predict the behavior of steel plate shear
walls with partial infill plate interconnections with or without boundary stiffeners. Overall, the study provides
procedures and recommendations to improve the performance of steel plate shear walls with limited infill plate
interconnections.

1. Introduction same authors, a two-phase numerical study was conducted to investigate

the effect of column flexural demands caused by column rotations at
Steel Plate Shear Wall (SPSW) is a commonly used structural lateral floor levels due to inter-story drift gradient [11]. The results indicated
load-resisting system due to a variety of advantages such as low weight, that the columns of designed beam-connected web plates determined
fast and appropriate workability, significant ultimate strength, sufficient more desirable behavior, preventing buckling limit states under twenty
ductility, and stiffness. Several studies investigated the advantages of ground motions scaled to a seismic hazard level of 2% in 50 years [11].
these systems to be used with several structural limitations and space- Along the same lines, it was shown that the participation of the frame in
contained areas [1-4]. In addition, recent studies allocated to improve the shear strength of the whole system could vary from 25% to 43% of
the performance of this system by provision of sufficient interconnection the total strength [12].
interactions between the plate and the boundary elements [5,6], or Previously, studies investigated the effects of steel plate shear wall
different infill plate shapes and geometries such as corrugated infill systems with limited infill plate interconnection to the boundary ele­
plates and butterfly-shaped infill plates [7-9], to overcome previously ments. Zhang et al. (2019) investigated a beam-only infill plate con­
reported structural issues. nected to a prefabricated steel plate shear wall by discontinuous cover
For this purpose, previous studies focused on the modeling behavior plate connection (DCPC). The performance of this type of steel plate
of beam-connected web plates engaged in the beam-connected web shear wall under seismic criteria was shown to be heavily affected by the
plates system. The study showed a desirable performance for high en­ thickness of the infill plate. The ultimate capacity of the prefabricated
ergy dissipation capacity and base shear [10]. In a similar study by the steel plate shear wall connected only to the beam with DCPC system was

* Corresponding author.
E-mail address: [email protected] (A. Farzampour).

https://doi.org/10.1016/j.istruc.2023.04.100
Received 6 January 2023; Received in revised form 28 March 2023; Accepted 24 April 2023
Available online 4 May 2023
2352-0124/© 2023 Institution of Structural Engineers. Published by Elsevier Ltd. All rights reserved.
N. Paslar et al. Structures 53 (2023) 749–763

marginally lower than the corresponding fully welded connected sys­ dissipation, initial stiffness, and lateral strength. Hence, TSPSWs with
tem. In case of the marginal drifts, the capacity of energy dissipation in infill plates attached only to the beams could be useful for retrofitting
this system was the same. By studying the story drift increments, greater and strengthening steel frames that have inadequate stiffness and
energy dissipation capacity for the prefabricated SPSW compared to the strength, considering reduced labor and material efficiency due to the
fully welded SPSW was reported [13]. A comparative assessment of steel absence of the interconnections between columns and infill plate edges
plate shear walls (SPSWs) with a diverse variety of unbonded stiffeners [21]. Schumacher et al. (1999) investigated different forms of inter­
has been carried out by Yu et al. (2019). So as to prevent out of plane connection between infill plate and boundary elements considering the
bending and deformation in the steel plates, multiple ribs and precast interaction between corner connections. In one of the models in which
concrete panels were established. Cyclic loading were tested on three the interconnection was welded directly without fishplates, it is shown
specimens, namely unstiffened, partially restrained and completely that the reduction of ultimate strength was less than in the models with
restrained SPSW. Columns’ inward deformation, rotation of beam- fishplates. However, this type of interconnection was not practical due
column connections and infilled steel plates out-of-plane deformations to the high degree of accuracy required in construction. In a separate
were prevented in both completely and partially restrained specimens. model, a a square-shaped cut with a length of 60 mm was made at the
Overall, the partially restrained specimen with unbonded multiple ribs corners to decrease the tension at the boundary elements’ connection
exhibits an outstanding behavior and constructability [14]. with high tension regions. The results showed that although the tearing
The cyclic performance of low-yield-strength (LYS) innovative steel was delayed on the corner, it did not significantly affect the final results
plate shear walls was investigated experimentally and numerically [22].
conducted [15,16]. It is shown that the partial connection of the infill To fortify SPSWs for better performance and overcome several pre­
plate had a significant impact on the steel plate shear walls’ strength, viously reported issues, the use of stiffeners was studied. Far­
energy dissipation, and ductility, whereas the for initial stiffness negli­ ahbakhshtooli et al (2019) studied multi-story buildings (7-, 10-, and 13-
gible impact was expected [15,16]. The structural behavior of steel story) stiffened SPSWs under 44 artificial ground motions compatible
shear walls with corrugated infill plates was investigated under mono­ with the uniform hazard spectrum of Vancouver, Canada region. A nu­
tonic and cyclic loadings using nonlinear finite element models. It is merical model was developed accordingly to obtain the seismic pinching
shown that increasing the web thickness and corrugation angle to 90◦ behavior of SPSWs, degradation backbone curve, and capping and yield
improved the structural response of the system. Moreover, a perforation points. The results of the study showed that a large number of shear
and increasing its size can reduce the performance by reducing the forces were resisted by boundary columns. However, the upper bound
participation of the infill plate [4]. Cui et al. (2020) studied a specific value recorded for the drift in each story was considerably smaller than
type of SPSW with a high capacity of load-bearing with a Low-Yield- the allowable value of 2.5%, suggested by NBCC 2015 (National
Point (LYP) steel shear web plate. Cyclic test results showed that the Building Code of Canada 2015, [23]. From the results of nonlinear static
yielding load and ultimate load of the LSPSW (LYP SPSW) were more analyses, it was shown that the post-buckling capacity and high initial
than 3000kN and 5000kN respectively, satisfying the initial design stiffness are imporved by using stiffeners; however, the stiffeners could
criteria. Ultimately, the factors of equivalent viscous damping (EVD) make the wall more rigid and less ductile [24].
and maximum ductility of the specimen were determined, indicating the Recently, SPSWs are designed with reduced interconnection inter­
feasibility of adopted geometric configuration for LSPSWs [17]. Along action between columns and infill plate in order to decrease demands on
the same lines, Zhang and Zhan (2019) studied discontinuous cover- the columns [3,20,25,26]. Although this procedure makes columns
plate connections used to model two beam-only connected SPSWs and more economical, the previous studies [27] showed that these forms of
prefabricated SPSWs. Despite the system’s desirable performance, the connections decrease the strength, stiffness and energy absorption.
infill plate has not been able to control the out-of-plane displacements, Furthermore, the system does not utilize the whole capacity of the infill
although significant ductility, sufficient energy dissipation capability, plate, especially for the regions with reduced interconnection in­
shuttle-shaped hysteretic curve, and stable hysteretic performance were teractions and the out-of-plane behavior of the infill plate could cause
obtained [18]. several stability issues. The present study investigates the behavior of
Similarly, Hajimirsadeghi et al. (2019) have investigated a method the SPSWs with limited infill plate interconnection interaction and de­
comprised of a modular steel shear panel containing boundary elements. termines improvement procedures using various boundary stiffeners
This method separated the gravity transferring mechanism from the located at the sides of the infill plate. The designed models are presented
lateral load resisting system. A full-scale experimental model was used with different geometrical infill plate conditions, type of interconnec­
to evaluate the seismic performance [19]. The results of the hysteretic tion between infill plate, boundary elements, and thickness of infill plate
curves indicated ductile behavior, reduced pinching effect, and a crucial and boundary stiffeners for which predictions equations are derived
capacity for energy dissipation, with a large quantity of energy dissi­ separately, and compared to computationally verified results subse­
pation obtained from the large hysteresis loops [19]. However, it can be quently. Ultimately, the ultimate strength, energy absorption, and
observed that the presence of an interconnection between the infill plate stiffness were established for all models and compared with each other
and the boundary elements could lead to large stress concentration and to indicate applicable solutions addressing issues raised by previous
precocious tearing in some regions. studies for SPSWs with limited interconnection of infill plate and
Along the same lines, Guo et al. (2011) tested two experimental boundary elements interactions.
SPSWs connected to beams only. The hysteretic behavior of these
specimens with the existence of stiffeners was investigated. Significant 2. Model properties
energy dissipation capacity and ductility were obtained from the
experimental results. They concluded that the energy dissipation ca­ 2.1. Verfication of the modeling methodology
pacity for the model with stiffeners was larger than that of the without
stiffeners. It is reported that a larger ratio for span-to-height and a lower The S2 model, a single bay-single story experimental specimen with a
ratio for height-to-thickness resulted in higher and better energy dissi­ width and height of 4000 and 2000 mm (center to center), conducted by
pation capacity of the SPSWs [20]. In addition, Vatansever and Yardimci Vian et al. (2009) is considered for the verification of the modeling
(2011) examined quasi-static loading tests on two TSPSWs (Thin methodology using ABAQUS 6.14-2. The infill plate has a thickness of
SPSWs). The two specimens were comprised of an infill plate connected 2.6 mm, with a yield stress of 165 MPa and an ultimate stress of 300
to the boundary frame on all four edges for which the infill plate was MPa, and is connected to the beam and column elements made of ASTM
connected to the beam. The contribution of the beam connection in A572 steel. All members have a yield stress of 345 MPa and an ultimate
specimens’ performance was found to be significant in terms of energy stress of 448 MPa, a modulus of elasticity of 210 GPa, and a strain

750
N. Paslar et al. Structures 53 (2023) 749–763

Fig. 1. Model’s geometrical properties based on Vian et al. [28].

Fig. 2. Boundary elements properties based on Vian et al. [28].

hardening ratio of 17% for A572 and 26% for LYS 165. The beam sec­ analysis is conducted to obtain the fine mesh size of 100 mm for which
tions are W18 × 65 with the Reduced Beam Section (RBS), and the any extra refining did not improve the accuracy of the results and
column sections are W18 × 71 [28]. Fig. 1 shows the geometry of the increased the cost of analysis.
specimen and the details of the infill plate interconnection and boundary The loading is applied to the midpoint of the upper beam on the
elements, and Fig. 2 shows the properties of the boundary elements’ experimental under the loading protocol of ATC-24 [29], which is
geometry. shown in Fig. 3. The initial mode of buckling deformation is applied for
The type of elements used for modeling is four noded shell element all the computational models since all components are not ideally flat
(S4R) to accurately capture the buckling behavior of thin plates while due to various construction reasons such as improper transportation and
avoiding any shear locking and hourglass issues. The first order elements installation. The captured computational mode of behavior is shown in
could be prone to shear locking and hourglass effects for which the Fig. 4(a) and (b), respectively.
reduced integration point and enhanced hourglass control is deter­ Fig. 5 indicates the comparison between the hysteretic response
mined. The fishplates were not considered in the computational behavior for the experimental specimen and the computational model.
modeling process due to having negligible effects on the boundary The results determine that the computational model with the described
components’ moment of inertia. It is noted that the mesh sensitivity modeling methodology captured the peak ultimate strength values with

751
N. Paslar et al. Structures 53 (2023) 749–763

2.2. Model definition

Sixty two computational models were developed based on Vian

et al.’s experimental model [28], for which the differences in geometry
of the infill plate, the type of interconnection between infill plate and
boundary elements, the thickness of the infill plate, and the boundary
stiffeners are considered. It is noted that several models were considered
with reduced infill plate sections as is determined in Fig. 6 to investigate
the behavior of systems with partially connected infill plates with
boundary elements, and provision of procedures to compensate the
strength, stiffness and ductility degradation issues. It is noted that
loading is applied based on the ATC-24 protocol [29] (3% drift) at the
middle of the up beam for all models according to the Vian et al (2009)
verified test. Moreover, four L stiffeners have been added to both sides of
the infill plate as it is determined in Fig. 6 to fortify the system and
prevent instant degradation. All the studied geometries of the models
Fig. 3. Loading protocol based on the ATC-24 Protocol [28]. are summarized in Table 1. In Table 1 in which “S” stands for SPSWs,
and the ’P’ refers to the partial infill plate which has various values
more than 97% accuracy. In addition, the loading and unloading parts of based on the percentage of the fully connected infill plate with the width
the hysteric behavior of the FE computational models captured the of 3177 mm. T1, T2, and T3 indicate the infill plate thicknesses of 2.6
actual behavior with more than 95% accuracy. It is noted that the pre­ mm, 3.3 mm, and 4 mm, respectively. The letter “S” at the end of each
viously reported limit states and mode of behavior occurred over the model’s name represents the stiffener for which the related thickness is
reduced section at the ends of the beams due to high tension stresses, included in millimeters. For instance, No. 8, ‘s-P90T2-S5’ is the SPSW
which were accurately captured by the FE models as it is shown in Fig. 4 with 90% infill plate length compared to the full length of 3177 mm,
(c) and (d), respectively. with 3.3 mm infill plate thickness, and it has boundary stiffeners with a

Fig. 4. FE and experimental model at the drift of 3%.

752
N. Paslar et al. Structures 53 (2023) 749–763

Fig. 5. SPSW with boundary stiffeners details.

positive displacement side. Fig. 7 demonstrates the hysteresis and

backbone curves. Backbone curves were extracted through the
maximum force of each displacement in the positive direction.

3.2. Models with different infill plate sizes and boundary stiffeners

Fig. 8 illustrates the stress distribution and out-of-plane displacement

behavior of some of the models at 3% of loading. It is shown that, S-
P100T1 computational model exhibits higher ultimate strength and
initial stiffness due to complete tension field formation (in comparison
with verification and other models). However, the designed columns are
less economical, have significant stresses and plastic strain concentra­
tion at the joints. In addition, Fig. 8(c) and Fig. 8(d) (S-P100T1-BC)
exhibit the beam-connected steel plate shear wall with the lack of
complete formation of the tension field, and significantly higher out-of-
plane displacement compared to S-P100T1 at edge regions. The effect of
stiffeners implementation for the partially connected steel plate shear
wall for alleviating the structural issues is summarized in Fig. 8(e) and
Fig. 8(f). Although the boundary stiffeners could partially limit the
Fig. 6. Hysteresis curves and developed backbone curve for computational maximum out-of-plane displacement, they transfer the location with
post-processing. maximum out-of-plane displacements from the edges to the middle
resulting in improved tension field formations.
5 mm thickness. Furthermore, it is shown that reducing infill plate width could be
beneficial in preventing stress concentration at the beam-column
3. Discussion of the results connection area, leading to desirable transfer of the loads to the safer
zone within the infill panel and ductile behavior. The overall behavior of
3.1. Backbone behavior development based on the hysteric results the load-resisting system can be improved significantly through the
implementation of a thicker infill plate with partial beam interconnec­
Computational models with various lengths of infill plate, with or tion, as shown in Fig. 8(g) and (h). However, it’s noteworthy that partial
without boundary stiffeners, different thicknesses of infill plate and infill plate interconnection without stiffeners at both sides of the infill
stiffener, are classified and represented based on the percentage of the plate could lead to greater out-of-plane displacement. Although in
length of fully interconnected infill plate. Previous studies indicate that several studies it is indicated that, SPSWs with beam-only connected
the infill plate is typically out of service at drifts greater than 3%, and infill plates could lead to an economical design approach for columns
does not have a considerable effect on the behavior of the system, to [3,20,25,26], recent investigations show that the absence of the inter­
have an accurate examination, the loading was restricted for all models connection between infill plate and columns could negatively affect the
at 3% drift (Paslar et al., 2020). The backbone curves are obtained from performance of SPSWs as it is represented in Table 2.
the gradual displacements and force values of hysteresis results on the

753
N. Paslar et al. Structures 53 (2023) 749–763

Table 1 Table 1 (continued )

Geometrical properties of computational models. No. Name Column Beam Plate Plate’s
No. Name Column Beam Plate Plate’s stiffener
stiffener
37 S-P70T3- W18 × W18 × 2471 × 1534 × L 50 × 97 ×
1 S-P100T1 W18 × W18 × 3530 × 1534 × S15 71 65 4 mm 15 mm
71 65 2.6 mm 38 S-P70T3- W18 × W18 × 2471 × 1534 × L 50 × 97 ×
2 S-P100T1- W18 × W18 × 3530 × 1534 × S25 71 65 4 mm 25 mm
BC*1 71 65 2.6 mm 39 S-P60T1 W18 × W18 × 2118 × 1534 ×
3 S-P90T1 W18 × W18 × 3177 × 1534 × 71 65 2.6 mm
71 65 2.6 mm 40 S-P60T1-S5 W18 × W18 × 2118 × 1534 × L 50 × 97 × 5
4 S-P90T1-S5 W18 × W18 × 3177 × 1534 × L 50 × 97 × 5 71 65 2.6 mm mm
71 65 2.6 mm mm 41 S-P60T1- W18 × W18 × 2118 × 1534 × L 50 × 97 ×
5 S-P90T1- W18 × W18 × 3177 × 1534 × L 50 × 97 × S15 71 65 2.6 mm 15 mm
S15 71 65 2.6 mm 15 mm 42 S-P60T1- W18 × W18 × 2118 × 1534 × L 50 × 97 ×
6 S-P90T1- W18 × W18 × 3177 × 1534 × L 50 × 97 × S25 71 65 2.6 mm 25 mm
S25 71 65 2.6 mm 25 mm 43 S-P60T2 W18 × W18 × 2118 × 1534 ×
7 S-P90T2 W18 × W18 × 3177 × 1534 × 71 65 3.3 mm
71 65 3.3 mm 44 S-P60T2-S5 W18 × W18 × 2118 × 1534 × L 50 × 97 × 5
8 S-P90T2-S5 W18 × W18 × 3177 × 1534 × L 50 × 97 × 5 71 65 3.3 mm mm
71 65 3.3 mm mm 45 S-P60T2- W18 × W18 × 2118 × 1534 × L 50 × 97 ×
9 S-P90T2- W18 × W18 × 3177 × 1534 × L 50 × 97 × S15 71 65 3.3 mm 15 mm
S15 71 65 3.3 mm 15 mm 46 S-P60T2- W18 × W18 × 2118 × 1534 × L 50 × 97 ×
10 S-P90T2- W18 × W18 × 3177 × 1534 × L 50 × 97 × S25 71 65 3.3 mm 25 mm
S25 71 65 3.3 mm 25 mm 47 S-P60T3 W18 × W18 × 2118 × 1534 ×
11 S-P90T3 W18 × W18 × 3177 × 1534 × 71 65 4 mm
71 65 4 mm 48 S-P60T3-S5 W18 × W18 × 2118 × 1534 × L 50 × 97 × 5
12 S-P90T3-S5 W18 × W18 × 3177 × 1534 × L 50 × 97 × 5 71 65 4 mm mm
71 65 4 mm mm 49 S-P60T3- W18 × W18 × 2118 × 1534 × L 50 × 97 ×
13 S-P90T3- W18 × W18 × 3177 × 1534 × L 50 × 97 × S15 71 65 4 mm 15 mm
S15 71 65 4 mm 15 mm 50 S-P60T3- W18 × W18 × 2118 × 1534 × L 50 × 97 ×
14 S-P90T3- W18 × W18 × 3177 × 1534 × L 50 × 97 × S25 71 65 4 mm 25 mm
S25 71 65 4 mm 25 mm 51 S-P50T1 W18 × W18 × 1765 × 1534 ×
15 S-P80T1 W18 × W18 × 2824 × 1534 × 71 65 2.6 mm
71 65 2.6 mm 52 S-P50T1-S5 W18 × W18 × 1765 × 1534 × L 50 × 97 × 5
16 S-P80T1-S5 W18 × W18 × 2824 × 1534 × L 50 × 97 × 5 71 65 2.6 mm mm
71 65 2.6 mm mm 53 S-P50T1- W18 × W18 × 1765 × 1534 × L 50 × 97 ×
17 S-P80T1- W18 × W18 × 2824 × 1534 × L 50 × 97 × S15 71 65 2.6 mm 15 mm
S15 71 65 2.6 mm 15 mm 54 S-P50T1- W18 × W18 × 1765 × 1534 × L 50 × 97 ×
18 S-P80T1- W18 × W18 × 2824 × 1534 × L 50 × 97 × S25 71 65 2.6 mm 25 mm
S25 71 65 2.6 mm 25 mm 55 S-P50T2 W18 × W18 × 2118 × 1534 ×
19 S-P80T2 W18 × W18 × 2824 × 1534 × 71 65 3.3 mm
71 65 3.3 mm 56 S-P50T2-S5 W18 × W18 × 2118 × 1534 × L 50 × 97 × 5
20 S-P80T2-S5 W18 × W18 × 2824 × 1534 × L 50 × 97 × 5 71 65 3.3 mm mm
71 65 3.3 mm mm 57 S-P50T2- W18 × W18 × 2118 × 1534 × L 50 × 97 ×
21 S-P80T2- W18 × W18 × 2824 × 1534 × L 50 × 97 × S15 71 65 3.3 mm 15 mm
S15 71 65 3.3 mm 15 mm 58 S-P50T2- W18 × W18 × 2118 × 1534 × L 50 × 97 ×
22 S-P80T2- W18 × W18 × 2824 × 1534 × L 50 × 97 × S25 71 65 3.3 mm 25 mm
S25 71 65 3.3 mm 25 mm 59 S-P50T3 W18 × W18 × 2118 × 1534 ×
23 S-P80T3 W18 × W18 × 2824 × 1534 × 71 65 4 mm
71 65 4 mm 60 S-P50T3-S5 W18 × W18 × 2118 × 1534 × L 50 × 97 × 5
24 S-P80T3-S5 W18 × W18 × 2824 × 1534 × L 50 × 97 × 5 71 65 4 mm mm
71 65 4 mm mm 61 S-P50T3- W18 × W18 × 2118 × 1534 × L 50 × 97 ×
25 S-P80T3- W18 × W18 × 2824 × 1534 × L 50 × 97 × S15 71 65 4 mm 15 mm
S15 71 65 4 mm 15 mm 62 S-P50T3- W18 × W18 × 2118 × 1534 × L 50 × 97 ×
26 S-P80T3- W18 × W18 × 2824 × 1534 × L 50 × 97 × S25 71 65 4 mm 25 mm
S25 71 65 4 mm 25 mm
*1, BC = beam connection.
27 S-P70T1 W18 × W18 × 2471 × 1534 ×
71 65 2.6 mm
28 S-P70T1-S5 W18 × W18 × 2471 × 1534 × L 50 × 97 × 5 3.2.1. Models with the infill plate of 100% length
71 65 2.6 mm mm Fig. 9 represents the backbone curves of S-P100T1 and S-P100T1-BC.
29 S-P70T1- W18 × W18 × 2471 × 1534 × L 50 × 97 ×
S15 71 65 2.6 mm 15 mm
As shown in Table 2, the model with beam-only infill plate intercon­
30 S-P70T1- W18 × W18 × 2471 × 1534 × L 50 × 97 × nection (S-P100T1-BC) has a lower ultimate strength, energy absorp­
S25 71 65 2.6 mm 25 mm tion, and stiffness compared to SPSW with a fully connected infill plate
31 S-P70T2 W18 × W18 × 2471 × 1534 × with boundary elements (S-P100T1).
71 65 3.3 mm
The implementation of the beam-connected SPSWs without stiffener
32 S-P70T2-S5 W18 × W18 × 2471 × 1534 × L 50 × 97 × 5
71 65 3.3 mm mm decreased the ultimate strength by 10.45%, energy absorption by 15%
33 S-P70T2- W18 × W18 × 2471 × 1534 × L 50 × 97 × and, 23.54% stiffness on average. In addition, the maximum out-of-
S15 71 65 3.3 mm 15 mm plane displacement of the infill plate is significant in comparison with
34 S-P70T2- W18 × W18 × 2471 × 1534 × L 50 × 97 × the model of S-P100T1, which promotes the utilization of boundary
S25 71 65 3.3 mm 25 mm
35 S-P70T3 W18 × W18 × 2471 × 1534 ×
stiffeners to improve the overall performance of the system.
71 65 4 mm
36 S-P70T3-S5 W18 × W18 × 2471 × 1534 × L 50 × 97 × 5 3.2.2. Models with the infill plate of 90% length
71 65 4 mm mm Fig. 10 displays the backbone curves of models with 90% width,
different thicknesses of infill plate, and boundary stiffeners. The results
show that adding four-five millimeter thick stiffeners at the sides of the

754
N. Paslar et al. Structures 53 (2023) 749–763

Fig. 7. Computational results at 3% displacement loading.

Fig. 8. Backbone curves of s-p100t1 and s-p100t1-bc.

755
N. Paslar et al. Structures 53 (2023) 749–763

Table 2 3.2.4. Models with the infill plate of 70% length

Summary of the results for classic SPSW and beam-connected SPSW. Similarly, Fig. 12 determines the backbone curves of models with
No. Name Ultimate Energy Stiffness Max out-of-plane 70% interconnection width, various thicknesses of infill plate and
strength absorption (kN/ boundary stiffeners.
MIDDLE EDGE
(kN) (kN.mm) mm)
(mm) (mm) The results for the computational models with 70% of infill plate
illustrate that the 5 mm stiffener would improve the ultimate strength,
1 S-P100T1 2106.42 103696.9 127.23 99.2 17
2 S- 1886.27 87306.8 97.27 70.7 192.2
energy absorption and stiffness by 5.3%, 7.13% and, 13.5% on average
P100T1- for all the considered thickness of infill plate. Furthermore, the 15 mm
BC stiffener indicates a 15.9% increase in ultimate strength, 14.23% in
energy absorption, and 21.54% in stiffness on average for all the
considered thicknesses of the infill plate. In addition, the 25 mm stiff­
infill plate would increase ultimate strength, energy absorption and
ener increases the ultimate strength, energy absorption, and stiffness by
stiffness up to 6.73%, and 6.81% respectively, as it is shown in Table 3.
19.8%, 19.54%, and 26.11% on average for various infill plates’ thick­
Moreover, utilizing the boundary stiffeners decreased the maximum out-
nesses (Table 5).
of-plane displacements by about 86% at the edges, and transfers the
maximum out-of-plane displacements far from the joint areas to the
3.2.5. Models with the infill plate of 60% length
middle of the infill plate resulting in ductile behavior, and the control of
Fig. 13 represents the backbone curves of models with 60% width
the stress concentration at the joint area.
and different thicknesses of the infill plate and boundary stiffeners. The
results of each model are mentioned in Table 6.
3.2.3. Models with the infill plate of 80% length
From Table 6, it can be observed that the infill plate with a thickness
Fig. 11 represents the backbone curves for the computational models
of 3.3 mm (T2) instead of 2.6 mm (T1), would improve the ultimate
with 80% width, different infill plate thicknesses, and boundary stiff­
strength by 3.31%, the energy absorption by 5%, and stiffness by
eners. Although all results for computational models with 80% of the
10.15%; however, for the infill plate with a thickness of 4 mm (T1) the
infill plate interconnection width exhibit a decrease in all parameters
increase values were 6.93%, 12%, and 25.9%, respectively.
compared to those with 90% infill plate interconnection width however,
using boundary stiffeners (Table 4) can eliminate this decrease. The
3.2.6. Models with the infill plate of 50% length
boundary stiffeners caused an increase of 4.80% in ultimate strength,
Along the same lines, Fig. 14 shows the backbone curves of models
6.6% in energy absorption, and 4.55% in stiffness averagely for the
with 50% interconnection between the infill plate and the boundary
models with 80% of infill plate equipped with boundary stiffeners.

Fig. 9. Envelope backbone curves of models with 90% width of infill plate.

Fig. 10. Envelope backbone curves of models with 80% width of infill plate.

756
N. Paslar et al. Structures 53 (2023) 749–763

Table 3 added to improve the overall behavior of SPSWs with beam-connected

Summary of results for computational models with 90% width of infill plate infill plates. Subsequently, all computational models are compared
interconnection. with S-P100T1-BC which is fully connected to the beam only without
No. Name Ultimate Energy Stiffness Max out-of- any stiffeners. Fig. 15 illustrates the comparisons of the results of all
strength absorption (kN/mm) plane models with the S-P100T1-BC model.
(kN) (kN.mm) From Fig. 15(a) can be inferred that SPSW with T1 thickness and
Middle Edge
(mm) (mm) without stiffener and 50% infill plate length (S-P50T1) has gained about
3 S-P90T1 1842.78 86248.37 96.50 85.4 117.6 9% less ultimate strength in comparison with the S-P100T1-BC model,
4 S-P90T1- 1987 91397.92 104.06 84.9 31.5 and SPSWs with T1 thickness and without stiffener and 90% plate length
S5 (S-P90T1) has gained about 2% less ultimate strength compared to the S-
5 S-P90T1- 2072.44 95055.79 106.90 79.2 20 P100T1-BC model. Regardless of infill plate length, increasing the
S15
6 S-P90T1- 2189.56 99315.57 110.71 88.3 19.6
thickness of the stiffener or the infill plate results in increased ultimate
S25 strength, energy absorption, and stiffness values. For instance, with the
7 S-P90T2 2043.35 93365.78 111.40 103.1 97.2 increase of 10 mm in stiffeners’ size, 7.82%, 6.25% and 5.30% increases
8 S-P90T2- 2160.43 99061.23 114.3 101 44.6 in ultimate strength, energy absorption and stiffness are obtained.
S5
Fig. 15 illustrates stiffeners enrich the overall behavior of SPSWs. The
9 S-P90T2- 2317.09 105583.9 123.0431 99.3 24.3
S15 majority of computational models with stiffeners remained in the posi­
10 S-P90T2- 2377.43 109199.2 139.2593 90.7 10.2 tive regions of the graph indicating a better performance compared with
S25 models without stiffeners. It is noted that the regions showing values
11 S-P90T3 2125.73 99732.42 129.38 95.2 157.9 under zero have less desirable performance compared to S-P100T1-BC
12 S-P90T3- 2267.07 108091.1 141.75 102.2 30.2
S5
13 S-P90T3- 2429.36 113228.9 148.92 107.3 30.7
S15 Table 4
14 S-P90T3- 2513.77 116250.7 150.93 90.4 16.7 Summary of results for computational models with 80% width of infill plate
S25 interconnection.
No. Name Ultimate Energy Stiffness Max out-of-
strength absorption (kN/ plane
elements. From Table 3 and Table 7, it can be concluded that utilizing
(kN) (kN.mm) mm)
the 50% width plate decreases the ultimate strength, energy absorption, Middle Edge
(mm) (mm)
and stiffness by 12.19%, 17.1%, and 31.3% on average in comparison
with models with 90% interconnection infill plate. It was expected that, 15 S-P80T1 1821.59 82736.25 91.24 59.9 133.2
with increasing in thicknesses of the infill plate or boundary stiffeners, 16 S-P80T1- 1952.33 86592.43 94.71 92.1 50.8
S5
the overall stiffness of the system would increase. With the increase of
17 S-P80T1- 2064.36 92036.44 99.63 84 15.6
the stiffener’s thickness (e.g. from 5 mm to 15 mm) the overall initial S15
stiffness of the system would increase 11.66% on average; however, 18 S-P80T1- 2186.61 94509.38 108.33 86.6 12.8
with 50% increase in infill plate’s thickness (e.g. from 2.66 to 4 mm), the S25
19 S-P80T2 1984.3 88083.89 100.93 108.2 147.1
overall stiffness increases 25.76% on average, which means the thick­
20 S-P80T2- 2053.62 95274.05 110.61 104.3 45.8
ness of infill plate is fairly more effective in improving the performance S5
of lateral load resisting system compared with the thicknesses of 21 S-P80T2- 2297.08 101733.1 113.39 85.1 28.7
boundary elements. S15
22 S-P80T2- 2369.97 105245 130.49 86.3 15.8
S25
3.3. Investigation of the structural solution for improving the behavior of 23 S-P80T3 2040.86 94932.7 126.42 105.2 171.1
the steel plate shear walls with partially connected infill plate and 24 S-P80T3- 2116.41 101663.4 126.76 116.8 47
boundary stiffeners S5
25 S-P80T3- 2386.91 107699.4 142.08 89.5 29.4
S15
Several structural solutions are considered and investigated for 26 S-P80T3- 2484.91 112524.4 144.21 80.8 13.3
improving the behavior of the steel plate shear walls with partially S25
interconnected infill plates. For this purpose, various stiffeners are

Fig. 11. Envelope backbone curves for computational models with 70% width of infill plate interconnection.

757
N. Paslar et al. Structures 53 (2023) 749–763

Fig. 12. Envelope backbone curves of models with 60% width of infill plate.

Table 5
Summary of results for computational models with 70% width of infill plate
interconnection. Table 6
No. Name Ultimate Energy Stiffness Max out-of- Summary of results for computational models with 60% width of infill plate
strength absorption (kN/ plane interconnection.
(kN) (kN.mm) mm)
Middle Edge No. Name Ultimate Energy Stiffness Max out-of-
(mm) (mm) strength absorption (kN/ plane
(kN) (kN.mm) mm)
27 S-P70T1 1792.29 79848.87 76.37 89.3 90.7 Middle Edge
28 S-P70T1- 1931.88 84553.15 86.48 100.3 29.8 (mm) (mm)
S5
39 S-P60T1 1780.09 76634.01 72.77 79.2 99.2
29 S-P70T1- 2052.01 90148.28 98.36 82.4 20
40 S-P60T1- 1861.33 81861.08 82.6 101.6 27.4
S15
S5
30 S-P70T1- 2161.47 93722.13 105.95 81.4 22.8
41 S-P60T1- 1990.1 85863.19 85.18 79.4 19.4
S25
S15
31 S-P70T2 1904.88 84874.03 90.46 86 164.8
42 S-P60T1- 2118.76 91499.19 88.88 80.8 6.6
32 S-P70T2- 1979.93 91493.5 102.5 85.1 45.3
S25
S5
43 S-P60T2 1839.09 80483.87 80.16 90.1 142.1
33 S-P70T2- 2204.09 96911.98 108.39 83.5 9.5
44 S-P60T2- 1956.8 87712.96 94.39 113.3 35.1
S15
S5
34 S-P70T2- 2309.1 101737.1 111.1 88.8 4.5
45 S-P60T2- 2107.56 92679.26 96.06 86.2 16.4
S25
S15
35 S-P70T3 1952.36 89637.36 108.56 84.6 166
46 S-P60T2- 2195.34 97043.22 100.23 92.1 2.31
36 S-P70T3- 2049.86 96571.69 123.91 77.8 38.9
S25
S5
47 S-P60T3 1903.6 85872.87 91.63 70.7 154.2
37 S-P70T3- 2318.96 103656.2 125.97 88.9 30
48 S-P60T3- 1983.32 91479.98 106.98 107.9 44.8
S15
S5
38 S-P70T3- 2430.9 108829.3 126.82 85.4 19.1
49 S-P60T3- 2215.44 99227.21 110.75 92.5 18
S25
S15
50 S-P60T3- 2300.94 102874.7 111.88 93 6.6
S25

Fig. 13. Envelope backbone curves of models with 50% width of infill plate with boundary elements.

758
N. Paslar et al. Structures 53 (2023) 749–763

Fig. 14. Comparisons of the results with S-P100T1-BC model.

759
N. Paslar et al. Structures 53 (2023) 749–763

Table 7 shown in Fig. 16, the ultimate strength of SPSWs with an infill plate
Summary of results for computational models with 50% width of infill plate partially connected to beams or columns can be derived based on the
interconnection with boundary elements. infill plate connectivity ratio. For this purpose, the lateral resisting
No. Name Ultimate Energy Stiffness Max out-of- strength of the frame and the corresponding contributions of the steel
strength absorption (kN/ plane plate under post-buckling conditions are summarized in Eq. (1) and Eq.
(kN) (kN.mm) mm) (2) based on plate-frame interaction [30].
Middle Edge
(mm) (mm)
4Mfp
51 S-P50T1 1723.84 72462.14 69.03 72.8 134.3 Ffu = (1)
d
52 S-P50T1- 1828.04 77243.59 69.91 8.3 87.8
S5 ( )
1
53 S-P50T1- 1916.15 82906.48 77.54 78.4 14.5 Fpu = bt τcr + σty sin(2θ) (2)
S15 2
54 S-P50T1- 1979.74 85626.33 82.07 73.9 15.2
Where Ffu and Fpu are the lateral load-resisting contributions of the
S25
55 S-P50T2 1741.93 74773.01 75.21 76.7 156.3 frame and steel plate, respectively. Here, b represents the width of the
56 S-P50T2- 1868.58 81788.66 83.99 102.2 16.3 panel, while d and t represent the length and thickness of the steel infill
S5 plate. ϭty is the yield stress of the tension field action, θ is the post-
57 S-P50T2- 1992.64 87846.44 85.89 8 13.2 buckling angle of the tension field action, and τcr is the critical shear
S15
58 S-P50T2- 2099.06 92098.11 89.93 68.1 8.8
buckling stress, as represented in Eq. (3).
S25
K π4 E ( t )2
59 S-P50T3 1804.03 79029.19 75.23 98 144.6 τcr = (3)
60 S-P50T3- 1932.87 85995.17 96.44 92.4 42.4 12(1 − μ2 ) b
S5
61 S-P50T3- 2071.63 92245.77 97.90 83.2 14
In which μ is the poison ratio of the steel and E is the elastic modulus.
S15 K represents the buckling geometrical coefficient [30], which is derived
62 S-P50T3- 2156.95 95,759 106.59 88 12.9 based on the geometrical properties of the plate, as determined in Eq.
S25 (4).
( )2
b b
reference model. K = 5.35 + 4
d
for ≥ 1
d
Considering computational models with thicker infill plates, the S-
P50T3 model has gained 5% less ultimate strength in comparison with ( )2
b b
the S-P100T1-BC model while the others have gained more due to either K = 5.35 + 4for < 1 (4)
d d
the use of thicker infill plates or the use of boundary stiffeners. Also,
results illustrate that boundary stiffeners compensate for the loss of ul­ The computational regression analysis on the ultimate strength re­
timate strength, energy absorption, and stiffness in beam-connected sults of the parametric study is conducted, and the total strength of the
only SPSWs. Using boundary stiffeners in beam-connected only SPSWs SPSW based on the infill plate length ratio is derived from Eq. (5)
have more desirable performance in comparison with the fully con­ [
( )
( )]
Ffu
nected model. In addition, increasing the thickness of the infill plate Fα = Ffu + Fpu 1 − (1 − α) (5)
Ffu + Fpu
could be assumed as a major factor. For example, with a 26.9% increase
(from 2.6 mm to 3.3 mm) in the thickness of the infill plate, there is an Where Fα represents the ultimate strength of reduced infill plate
average increase of 6.1%, 5.48%, and 12.72% in ultimate strength, en­ SPSWs without boundary stiffeners, and α is the ratio of the new infill
ergy absorption, and stiffness, respectively. plate length compared to the full size of the infill plate. Fig. 16 sche­
matically shows the plate interconnection ratio (α) and its relationship
3.4. Behavior prediction of steel plate shear walls with partially connected to the ultimate strength of SPSWs with reduced infill plates and no
infill plates and stiffeners boundary stiffeners.
A separate computational regression analysis was conducted on the
Based on the comparison of the SPSWs with various infill plate results of SPSWs with reduced infill plates in the absence of boundary
interconnection types, it is concluded that partially connected infill plate stiffeners, in order to compare the different models. The total strength of
computational models with stiffeners could have desirable structural SPSWs based on the utilization of boundary stiffener thickness is derived
performance with the capability to reduce demands on the side columns, accordingly from the following regression analysis on the models with
concentration of high strain at the connection, and installation cost. In stiffeners, as indicated in Eq. (6).
addition, Based on further regression analysis of computational results

Fig. 15. α factor schematic representation.

760
N. Paslar et al. Structures 53 (2023) 749–763

Fig. 16. Comparison of the proposed analytical equations with validated FE computational models’ results.

761
N. Paslar et al. Structures 53 (2023) 749–763

9.86Est Validation, Visualization, Writing – original draft, Writing – review &

Fst = 1.36(ast +bst ) (6)
editing. Navid Chalangaran: Conceptualization, Methodology.
tst

Fst demonstrate the strength that the boundary stiffeners provide to the Declaration of Competing Interest
system. ESt shows the Young’s modulus of the stiffener material, and aSt
and bst represent the two side lengths of the stiffeners, as shown in Fig. 6 The authors declare that they have no known competing financial
tSt represents the thickness of the stiffeners. Therefore, the total strength interests or personal relationships that could have appeared to influence
of the SPSW with partially interconnected plate length, either with or the work reported in this paper.
without stiffeners, can be derived using Eq. (7).
[ ( )]
( ) Ffu 9.86Est tst Acknowledgment
Fa,st = Ffu + Fpu 1 − (1 − α) + (7)
Ffu + Fpu 1.36(ast + bst )
This study is conducted by the authors and has not been supported by
Which Fα.st is the ultimate strength of the reduced infill plate SPSWs
any other organizations.
either with or without boundary stiffeners. Fig. 16, shows the accuracy
of the proposed equation in predicting the beam-only SPSW systems
with partially infill plate length, and either with or without boundary Funding
stiffeners. Results indicate that the proposed equation can predict the
ultimate strength of the SPSWs with partial infill plate’s interconnection, This research did not receive any specific grant from funding
with or without boundary stiffeners with more than 93% accuracy. The agencies in the public, commercial, or not-for-profit sectors.
R2 values are added to indicate the accuracy of the regression study on
the computational results. Appendix A. Supplementary data

4. Conclusions Supplementary data to this article can be found online at https://doi.

org/10.1016/j.istruc.2023.04.100.
The present study investigates the performance of SPSWs with
different ranges of infill plate widths, and with or without boundary References
stiffeners. The computational studies are established according to the
verification test specimen initially. Subsequently, models are developed [1] Ali MM, Osman SA, Husam O, Al-Zand AW. Numerical study of the cyclic behavior
of steel plate shear wall systems (SPSWs) with differently shaped openings. Steel
according to differences in infill plate geometrical properties and the Compos Struct 2018;26(3):361–73.
presence or absence of boundary stiffeners. The results represent that [2] Alavi E, Nateghi F. Experimental study on diagonally stiffened steel plate shear
the beam-connected SPSWs and without stiffeners exhibit the least walls with central perforation. J Constr Steel Res 2013;89:9–20.
[3] Paslar N, Farzampour A, Hatami F. Investigation of the infill plate boundary
desirable performance among all models. The detailed results of this condition effects on the overall performance of the steel plate shear walls with
study is summarized as follows: circular openings. Structures 2020;27:824–36.
[4] Farzampour A, Laman JA, Mofid M. Behavior prediction of corrugated steel plate
shear walls with openings. J Constr Steel Res 2015;114:258–68.
• Decreasing the infill plate width could reduce the ultimate strength [5] Sahoo DR, Sidhu BS, Kumar A. Behavior of unstiffened steel plate shear wall with
and stiffness by 8% and 30%, respectively. simple beam-to-column connections and flexible boundary elements. Int J Steel
• Reducing the infill plate width by 50% would decrease the ultimate Struct 2015;15(1):75–87.
[6] Alinia M, Dastfan M. Behaviour of thin steel plate shear walls regarding frame
strength, energy absorption, and stiffness by 18.16%, 30.12%, and members. J Constr Steel Res 2006;62(7):730–8.
45.74%, respectively. [7] Cao Q, Huang J. Experimental study and numerical simulation of corrugated steel
• SPSWs with beam-connected infill plates and without stiffeners plate shear walls subjected to cyclic loads. Thin-Walled Struct 2018;127:306–17.
[8] Farzampour A, Mansouri I, Hu JW. Seismic behavior investigation of the
exhibit the least desirable performance among all models.
corrugated steel shear walls considering variations of corrugation geometrical
• Boundary stiffeners can improve the ultimate strength, energy ab­ characteristics. Int J Steel Struct 2018;18(4):1297–305.
sorption, and stiffness by 15%, 20%, and 24% on average. [9] Farzampour A. Structural behavior prediction of the Butterfly-shaped and straight
• SPSWs with boundary stiffeners can be used as a substitute for either shear fuses. Structures 2021;33:3964–72.
[10] Ozcelik Y, Clayton PM. Strip model for steel plate shear walls with beam-connected
classic SPSW or SPSW with beam-connected infill plates to exhibit web plates. Eng Struct 2017;136:369–79.
desirable performance and transfer the stress concentration points [11] Ozcelik Y, Clayton PM. Behavior of columns of steel plate shear walls with beam-
from joint areas. connected web plates. Eng Struct 2018;172:820–32.
[12] Paslar N, Farzampour A. Effects of infill plate’s interconnection and boundary
• For the computational models without boundary stiffeners, the element stiffness on steel plate shear walls’ seismic performance. Materials 2022;
strength of the infill plate is not utilized completely due to the 15(16):5487.
incomplete formation of the tension field action which boundary [13] Zhang X, Zhang A-L, Liu X. Seismic performance of discontinuous cover-plate
connection for prefabricated steel plate shear wall. J Constr Steel Res 2019;160:
stiffeners could be uses to alleviate this issue by a considerable 374–86.
margin. [14] Yu J-G, Liu L-M, Li B, Hao J-P, Gao X, Feng X-T. Comparative study of steel plate
shear walls with different types of unbonded stiffeners. J Constr Steel Res 2019;
159:384–96.
Further research is required to examine the effects of different types [15] Farzampour A, Eatherton MR. Parametric computational study on butterfly-shaped
of stiffeners on SPSWs, since this work only analyzes the impact of hysteretic dampers. Front Struct Civ Eng 2019;13:1214–26.
boundary stiffeners on SPSWs with partially connected infill plates. [16] Farzampour A. Innovative structural fuse systems for various prototype
applications. Materials 2022;15(3):805.
Experimental research could be considred to study the behavior of
[17] Cui J-C, Xu J-D, Xu Z-R, Huo T. Cyclic behavior study of high load-bearing capacity
SPSWs with partially interconnected steel plates and stiffeners. steel plate shear wall. J Constr Steel Res 2020;172:106178.
[18] Zhang Y, Zhan X. Study on seismic behavior of steel frame-steel shear wall with
CRediT authorship contribution statement assembled two-side connections. Math Probl Eng 2019;2019.
[19] Hajimirsadeghi M, Mirtaheri M, Zandi A, Hariri-Ardebili M. Experimental cyclic
test and failure modes of a full scale enhanced modular steel plate shear wall. Eng
Nima Paslar: Conceptualization, Investigation, Methodology, Soft­ Fail Anal 2019;95:283–8.
ware, Visualization, Writing – original draft. Alireza Farzampour: [20] Guo L, Rong Q, Ma X, Zhang S. Behavior of steel plate shear wall connected to
frame beams only. Int J Steel Struct 2011;11(4):467–79.
Conceptualization, Data curation, Investigation, Methodology, Formal [21] Vatansever C, Yardimci N. Experimental investigation of thin steel plate shear
analysis, Project administration, Resources, Software, Supervision, walls with different infill-to-boundary frame connections. Steel Compos Struct
2011;11(3):251–71.

762
N. Paslar et al. Structures 53 (2023) 749–763

[22] Schumacher A, Grondin GY, Kulak GL. Connection of infill panels in steel plate [27] Paslar N, Farzampour A, Hatami F. Infill plate interconnection effects on the
shear walls. Can J Civ Eng 1999;26(5):549–63. structural behavior of steel plate shear walls. Thin-Walled Struct 2020;149:
[23] Farahbakhshtooli A, Bhowmick AK. Seismic collapse assessment of stiffened steel 106621.
plate shear walls using FEMA P695 methodology. Eng Struct 2019;200:109714. [28] Vian D, Bruneau M, Tsai K-C, Lin Y-C. Special perforated steel plate shear walls
[24] Farzampour A. Evaluating shear links for use in seismic structural fuses [Doctoral with reduced beam section anchor beams. I: experimental investigation. J Struct
dissertation]. Virginia Tech; 2019. Eng 2009;135(3):211–20.
[25] Clayton PM, Berman JW, Lowes LN. Seismic performance of self-centering steel [29] ATC-24, Guidelines for Seismic Testing of Components of Steel Structures. Report
plate shear walls with beam-only-connected web plates. J Constr Steel Res 2015; 24, Applied Technology Council, Redwood City, CA, (1992).
106:198–208. [30] Farzampour A, Mansouri I, Lee C-H, Sim H-B, Hu JW. Analysis and design
[26] Ozcelik Y, Clayton PM. Seismic design and performance of SPSWs with beam- recommendations for corrugated steel plate shear walls with a reduced beam
connected web plates. J Constr Steel Res 2018;142:55–67. section. Thin-Walled Struct 2018;132:658–66.

763