Validation of a steel dual-core self-centering brace (DC-SCB) for seismic resistance: from brace member to one-story one-bay braced frame tests

Abstract

A steel dual-core self-centering brace (DC-SCB) is an innovative structural member that provides both energy dissipation and self-centering properties to structures, reducing maximum and residual drifts of structures in earthquakes. The axial deformation capacity of the DC-SCB is doubled by a parallel arrangement of two inner cores, one outer box and two sets of tensioning elements. This paper presents cyclic test results of a DC-SCB component and a full-scale one-story, one-bay steel frame with a DC-SCB. The DC-SCB that was near 8 m-long was tested to evaluate its cyclic behavior and durability. The DC-SCB performed well under a total of three increasing cyclic loading tests and 60 low-cycle fatigue loading tests without failure. The maximum axial load of the DC-SCB was near 1700 kN at an interstory drift of 2.5%. Moreover, a three-story dual-core self-centering braced frame (DC-SCBF) with a single-diagonal DC-SCB was designed and its first-story, one-bay DC-SCBF subassembly specimen was tested in multiple earthquake-type loadings. The one-story, one-bay subassembly frame specimen performed well up to an interstory drift of 2% with yielding at the column base and local buckling in the steel beam; no damage of the DC-SCB was found after all tests. The maximum residual drift of the DC-SCBF caused by beam local buckling was 0.5% in 2.0% drift cycles.

Keywords

dual-core self-centering brace (DC-SCB) / braced frame tests / residual deformation

Introduction

A steel braced frame that resists lateral loads primarily by developing high axial forces in diagonal members has smaller drifts than a moment-resisting frame that resists lateral loads by forming plastic hinges in beams. The design philosophy for the braced frame is to ensure that the major plastic deformation in braces keeps columns and beams undamaged in strong earthquakes. One typical system is a Concentrically Braced Frame (CBF) that dissipates seismic energy by brace buckling in compression and yielding in tension. However, brace buckling degrades both strength and stiffness in large drifts and in some cases concentrates damage in a limited number of stories. The disadvantage of the CBF can be overcome by using a Buckling-Restrained Brace (BRB) that yields in both tension and compression without overall buckling. Numerous works [ 1– 6] have demonstrated satisfactory performances of BRBs or Buckling-Restrained Braced Frames (BRBFs), but the BRB under large ground motions provides not only large energy dissipation but also large residual deformation to structural systems. Figure 1(a) shows a typical cross section of a proposed Sandwiched Buckling-Restrained Brace (SBRB) that sandwiches a core plate between a pair of restraining members using fully tensioned high-strength A490 bolts to expedite the assembly process [ 6, 7]. The advantage is the ability to disassemble the SBRB, which not only means that the core plate can be replaced independently of the restraining members, but also provides an opportunity for inspection of the core in the field. A 7.5 m-long SBRB specimen [ 8] that was tested at NCREE, Taiwan, China (Fig. 1(b)) showed acceptable hysteretic responses with maximum axial forces of 6145 and 6404 kN in tension and compression (Fig. 1(c)), respectively. The value of the compression strength adjustment factor was 1.04, much less than 1.3 specified by AISC seismic provisions [ 9]. The cumulative ductility of the SBRB specimen after six phase tests was 878, also exceeding the value of 200 specified by AISC seismic provisions. Figure 1(d) shows applications of SBRBs in a new Gansu Science museum, China and Kaohsiung City library, Taiwan, China; these two buildings can be classified as a dual system in the US because both moment-resisting connections and SBRBs are used in the frame structure for earthquake resistance.

It appears that the BRB or SBRB with large energy dissipation always has large residual deformation in loading cycles, increasing the possibility of building structures with large residual deformations in large earthquakes. Therefore, a brace with both the self-centering (SC) and energy dissipation properties has been proposed in the past few years; fiber-reinforced polymer (FRP) tendons or shape-memory alloy bars are used as tensioning elements to provide restoring forces to the brace [ 10, 11]. Chou et al. [ 12– 14] develops a new steel dual-core self-centering brace (DC-SCB), which utilizes three steel bracing members, two friction devices and two sets of tensioning elements. The three steel bracing members and the two sets of tensioning elements are arranged in parallel in the DC-SCB to double the axial elongation capacity of the self-centering energy-dissipating (SCED) brace [ 10] if the same tensioning elements are used in both braces.

Past works focused on the development of the DC-SCBs that can exhibit a flag-shaped hysteretic response with minimal residual deformations [ 8, 12– 14]. The seismic performance of a steel frame with DC-SCBs has never been studied experimentally, especially together with inelastic responses in the beam, column and brace in multiple cyclic tests. Therefore, the objective of the work is to validate the seismic behavior of a steel frame with the DC-SCB as an earthquake-resisting mechanism. The DC-SCB in this study uses high-strength steel tendons as tensioning elements to eliminate sudden failure of FRP tendons when the brace is overloaded beyond the design limit [ 12]. The brace also uses the second core as an intermediate member for half the tendon anchorages to reduce half the work during initial post-tensioning and to simplify the deformation mechanism of the brace compared to the previous DC-SCB [ 14].

This paper is meant to give more of a unified look at behaviors in both a novel brace member itself [ 14] and a one-story one-bay subassemblage frame with a DC-SCB [ 15]. Since the cross section and length of the DC-SCB are the same in these two test programs, the interaction of the beams and columns to the DC-SCB can be clearly identified in these two tests. A dual gusset configuration is designed based on the maximum axial capacity of the DC-SCB so that inelastic action is limited to the brace, beam and column base. The bulk of the test results for each individual test program can be found in other journal papers separately [ 14, 15]. Therefore, the first part of the paper presents the basic mechanics and cyclic responses of the cross-anchored DC-SCB. A 7950-mm long cross-anchored DC-SCB, which was designed with ASTM A572 Gr. 50 steel bracing members and ASTM A416 Gr. 270 steel tendons, was tested six times to evaluate its cyclic performance. A prototype three-story steel DC-SCBF was then designed and analyzed; a full-scale one-story, one-bay DC-SCBF specimen that represents the prototype first-story braced frame was tested under multiple loadings. The objectives of the test program were to 1) validate the system response of the DC-SCBF, 2) study force distributions in framing members as the damage progresses in the DC-SCB, beam or columns, and 3) investigate the repair and replacement characteristics of the braced frame, as the same frame, brace and post-tensioning (PT) elements will be reused in multiple tests.

Behavior of dc-scb

The basic concept of the cross-anchored DC-SCB, including a relative movement between the first core and the outer box, is similar to that of the original DC-SCB [ 12, 13] except for the movement of the second core and its end plates. Figure 2 shows the kinematics of the cross-anchored DC-SCB, in which four loading cases are illustrated in Fig. 2(c). The left end of the cross-anchored DC-SCB is an extension of the first core, and a tensile force P/2 is applied to the right end of the plates extending from the outer box (Fig. 2(a)). When the initial PT force and the force required to activate the friction device are surpassed, the outer box and the first core begin moving with respect to the second core. The relative displacements d between the outer box and the second core and between the first core and the second core result in an axial displacement of 2d in the brace (Fig. 2(b)), which doubles the elongation dof the outer and inner tendon sets. The tensile activation load of a cross-anchored DC-SCB at which the first core and outer box start moving is

(1) F_{dt}=P_{dt}+P_f={\displaystyle \frac{n{T}_{in}}{\mathrm{2}}}+P_f,

where P_dt represents the initial PT load in tendons, P_f is the frictional resistance of the energy dissipative device, n is the number of tendons, and T_in is the initial tension force in one tendon. The tensile axial deformation, D_dt, corresponding to the tensile activation load is

(2) {\Delta }_{dt}={\displaystyle \frac{P_{ob,in}+P_f}{K_{ob}}},

where P_ob,in is the initial PT force in the outer box, and K_ob is the axial stiffness of the outer box. The elastic stiffness of the brace under tension is K_m,it ( = F_{dt}/{\Delta }_{dt} ). When the brace load reaches the tensile activation load, the post-elastic stiffness of the brace, K_m,pt, which is function of axial stiffnesses of the inner tendons, outer tendons and second core, is

(3) K_{m,pt}={\displaystyle \frac{\mathrm{1}}{{\displaystyle \frac{\mathrm{1}}{{\displaystyle \frac{n}{\mathrm{2}}}{K}_{ten}}}+{\displaystyle \frac{\mathrm{1}}{K_{\mathrm{2}c}}}+{\displaystyle \frac{\mathrm{1}}{{\displaystyle \frac{n}{\mathrm{2}}}{K}_{ten}}}}},

where

K_{ten}

is the axial stiffness of one tendon, K_2c is the axial stiffness of the second core, and n is the total number of tendons. The brace returns to its original position when the load is removed. The behavior of the brace under compression is similar to that under tension and the details can be found elsewhere [ 14].

Tests of a cross-anchored DC-SCB

The test program consisted of cyclic tests of a cross-anchored DC-SCB specimen. The specimen was placed in the test setup (Fig. 3(a)), which included one steel box column pin-supported to the laboratory floor and attached to two 1000 kN hydraulic actuators. The cross-anchored DC-SCB specimen was positioned at an incline of q = 26° with both ends welded to dual gusset plates, which were designed to remain elastic at the ultimate strength level [ 7]. The cross-anchored DC-SCB specimen had a first core of T250 × 280 × 8 mm, a second core of T210 × 240 × 10 mm, and an outer box of T340 × 340 × 8 mm. The specimen had 12 seven-wire ASTM A416 Grade 270 steel tendons, and only six tendons were anchored outside the outer end plates needed for the initial post-tensioning work (Fig. 3(b)). The red and blue ellipses represent locations of an inner tendon set and outer tendon set, respectively.

The minimum design story drift of the braced frame was 1% based on AISC Seismic Provisions [ 9]. The cross-anchored DC-SCB specimen was first tested to a drift of 0.36% before stressing bolts in the friction device to evaluate the initial PT load, P_dt, and then subjected to six phase tests. Two F10T 20-mm-diameter bolts were used to stress the friction device to provide energy dissipation in these six phase tests. The standard loading protocol consisted of two cycles per column drift of 0.09, 0.18, 0.36, 0.5, 1, 1.5 and 2% (Phase 1). The specimen was then subjected to 15 low-cycle fatigue tests at the column drift of 1.5% (Phase 2). The objective of the test was to evaluate the durability of the friction device and the tendon-anchorage system. The specimen was then reloaded in Phase 3 and 4 tests using the Phase 1 loading protocol but up to 2.5% to investigate the effects of steel tendon yielding on the brace behavior. The specimen was then subjected to additional 30 low-cycle fatigue tests at the column drift of 1.5% (Phase 5) and 15 low-cycle fatigue tests at the column drift of 2.5% (Phase 6).

The cross-anchored DC-SCB without friction devices revealed a bilinear elastic response; the change of stiffness occurred at a load equal to the initial PT force. Figure 4 shows the axial force versus axial displacement responses of the brace in Phase 1 and 2 tests. The specimen performed well under Phase 1 test up to the target drift of 2% (Fig. 4 (a)). Peak brace forces and tendon forces in the test were close to prediction. The maximum tendon strain obtained from strain gages was 0.66% at 2% drift, which was lower than the yield strain of 0.7% [ 14], so no yielding in tendons occurred. Figure 4(b) shows 15 cycle hysteretic responses of the DC-SCB in Phase 2 test, indicating a reliable energy dissipative device. No damage of the bracing members or tendon anchorages was found after all six phase tests.

Design of a prototype three-story DC-SCBF

Figure 5 shows the elevation of the prototype building, which was assumed to be located on stiff soil in Los Angeles, California. The DC-SCBF system was designed to provide lateral load resistance in the east–west direction. Two one-bay DC-SCBFs were considered in design of the braced system; each DC-SCBF was composed of H-shaped steel columns, steel beams and single-diagonal DC-SCBs (Fig. 5(a)). The design of the DC-SCBF used a force reduction factor R of 8, an overstrength factor W₀ of 2.5 and a displacement amplification factor C_d of 5 as used in design of the BRBF based on ASCE standard [ 16]. The mapped MCE spectral response acceleration at a short period S_S and one second S₁ was 1.5 g and 0.6g, respectively. For the building located at site class D, the site coefficients F_a and F_v were 1.0 and 1.5, respectively, leading to design spectral response accelerations at a short period and one second of 1.0 g and 0.6 g, respectively. Table 1 lists the first-mode structural period T from the elastic dynamic analysis, the seismic response coefficient C_s and the seismic design base shear V_des that includes the redundancy factor r_y of 1.5 for the frame. A conservative value of r_y equal to 1.5 was used based on two earthquake-resisting frames in one loading direction and a specific ground floor area [ 15, 16]. The structural period is estimated based on:

(4) T_a=C_t{h}^x,

where C_t is 0.0731, x is 0.75, and h is the building height. It is found that the first-mode period of the DC-SCBF that is obtained from the three-story frame model using the PISA computer program [ 17] is close to the code period (from T_a to 1.4T_a). Table 2 lists the design force (F_req), activation force (F_dt), initial PT force (P_dt), friction force (P_f) and maximum force (F_ut) of the DC-SCB for the prototype three-story braced frame. All beams, columns and braces are made of ASTM A572 Grade 50 steel; high-strength tendons are made of ASTM A416 Grade 270 steel.

A pushover analysis was conducted on the three-story prototype frame and the one-story subassembly frame specimen. Figure 5(b) shows that the DC-SCB in each floor reaches the activation force at a roof drift of 0.2% (Step A), which is after the design lateral force, V_des, of 792 kN, and yielding of the first-story beam occurs at a roof drift of 1.1% (Step B). The column base and the second-floor beam yield at a roof drift of 1.3~1.5% (Steps C and D); the third-floor beam yields at a roof drift of 2% (Step F). The sequence of plastic hinge formation in the DC-SCBF is consistent with that as typically observed in the BRBF [ 7]. The analytical results indicate that the energy dissipation provided by the beam and column base might cause residual drifts of the DC-SCBF under large earthquakes. The pushover analysis of the one-story one-bay DC-SCBF subassembly specimen was also conducted as shown in Fig. 5(b). Since the ASCE load pattern [ 16] was applied to the prototype three-story frame with an increasing amplitude and only a lateral force was applied at the top of the one-story subassembly frame, the analytical results revealed less lateral force in the prototype frame than in the one-story subassembly frame. The progress of plastic hinges in the subassembly frame specimen is similar to that as seen in the prototype frame, indicating that the seismic performance of the one-story one-bay DC-SCBF subassembly obtained in the tests can represent that of the prototype three-story building in earthquakes.

Full-scale one-story one-bay DC-SCBF subassembly tests

The test program consisted of eight-phase tests of a one-story one-bay DC-SCBF that had two columns H414 × 405 × 18 × 28 (mm), a beam H500 × 200 × 10 × 16 (mm) and a DC-SCB with a first core of T250 × 280 × 8 mm, a second core of T210 × 240 × 10 mm and an outer box of T340 × 340 × 8 mm (Fig. 3). The beam flange and web were groove-welded to the H-shaped column using the ER70S-G electrode, which is similar to the E71T-8 or E70TG-K2 electrodes with a minimum specified Charpy V-Notch value of 27 J at -29°C (20 ft-lbs at -20F). The steel backing was left in place for the beam top and bottom flange groove welds, and no fillet welds were made between the steel backing and column. The beam-to-column connection details did not follow stringent requirements of the SMRF [ 9] because the drift demand was much smaller in the DC-SCBF than that in the SMRF from nonlinear time history analyses [ 15]. The column panel zone was designed to remain elastic in all tests. The DC-SCB had 12 seven-wire ASTM A416 Grade 270 steel tendons, of which six were anchored outside the outer end plates. The initial PT force in six steel tendons was 416 kN. Friction devices that used C2680 brass shim plates sandwiched by 20 mm-diameter F10T bolts were placed on one end of the brace to provide energy dissipation. At a target drift of 2%, the brace force was 1597 kN, as determined from the tendon strain of 0.76% in the specimen. The DC-SCB was connected to framing members using a dual-gusset-plate configuration, which was designed to remain elastic based on the past work [ 7]. The DC-SCB was fabricated by a local steel fabricator and post-tensioned at NCREE, Taiwan, China.

Eight test phases

The subassembly frame specimen that had a DC-SCB without friction devices was first tested to a drift of 0.5% to evaluate the initial PT force in the brace and then re-loaded to a drift of 2% to evaluate its cyclic behavior (Phases 1 and 2). The subassembly frame specimen was then subjected to six phase tests, in which two F10T 20 mm-diameter bolts were used to stress the friction device to dissipate seismic energy. In Phase 2 and 3 tests, the specimen was subjected to a standard loading protocol that consisted of two cycles at a column drift of 0.09, 0.18, 0.36, 0.5, 1, 1.5 and 2%. In Phase 4 to 7 tests, the specimen was subjected to displacement histories that were obtained from the first-story responses of the prototype three-story DC-SCBF subjected to 1992 Lander Earthquake (IND090 record) and 1999 Chi-Chi Earthquake (TCU039 record), respectively, in the design based earthquake (DBE) and the maximum considered earthquake (MCE) levels. The specimen was then subjected to 10 low-cycle fatigue tests at a column drift of 1.5% (Phase 8).

Test results

The DC-SCB without friction devices exhibited a bilinear elastic response, leading to a bilinear elastic response of the frame before a drift of 0.5%. The initial PT force of the brace that was obtained in Phase 1 and 2 tests was 440 kN. The DC-SCB reached the activation force at an interstory drift of 0.36%; the column base and beam yielded after an interstory drift of 1%. In Phase 2 test, the beam flange showed minor local buckling at an interstory drift of 1.5%. Local buckling was never observed on the other end of the beam with a gusset connection. To decrease the amplitude of beam local buckling in subsequent tests, the beam on the actuator side was strengthened by adding four internal flange stiffeners (IFSs) inside the beam flange. The IFS moment connection that was proposed to rehabilitate steel moment connections can exhibit minor beam local buckling after experiencing multiple seismic loads [ 18]. Four 12 mm-thick plates which were made of ASTM A572 Grade 50 steel were welded inside the beam flange of the DC-SCBF subassembly specimen for the Phase 3 to 8 tests.

The subassembly specimen in Phase 3 test was subjected to the same loading protocol as used in Phase 2 test, but two F10T bolts were stressed to provide the friction force (energy dissipation) of the brace. Figure 6(a) shows the actuator force versus the displacement responses of the frame in Phase 3 test. Beam yielding was observed at an interstory drift of 1.3%, and the amplitude of beam local buckling observed at an interstory drift of 2% was much smaller than that observed in Phase 2 test. Therefore, the IFS successfully moved beam local buckling away from the column face and reduced the amplitude of beam local buckling (Fig. 7).

The subassembly frame specimen was then loaded with two DBE and MCE motions (Fig. 6(b)); no damage was found in frame members, DC-SCB or tendons after experiencing two DBE and MCE motions (from Phase 4 to 7 tests). The DC-SCBF subassembly was then subjected to a fatigue loading with 10 cycles at a drift of 1.5% (Phase 8). The objective of the test was to evaluate the post-earthquake performance of the DC-SCBF system, energy dissipation device and tendon-anchorage system after experiencing multiple large seismic loadings. The one-story one-bay subassembly frame performed well in Phase 8 test (Fig. 6(c)). The initial PT force remained the same as observed in the previous test.

Figure 8(a) shows ratios of lateral forces in the moment-resisting frame (MRF), which is composed of two columns and a beam, and a DC-SCB in Phase 2 and 3 tests; more than half of the lateral force was resisted by the DC-SCB even in large drifts. The lateral responses of the DC-SCB in these two phase tests were different due to the use of an energy dissipating device in Phase 3 test. The lateral force in the DC-SCB was obtained by subtracting lateral loads in two columns, which were measured from rosettes, from the actuator forces. Figure 8(b) shows that the initial PT force in the DC-SCB decreased slightly after the Phase 2 test, but remained the same throughout rest of test phases. However, the gradual abrasion of brass shim plates causes smaller peak forces and energy dissipation of the brace in subsequent tests.

Conclusions

Traditional seismic-resisting systems in earthquakes dissipate energy in beams or braces, possibly leading to structural damage or residual drifts that are difficult and expensive to repair. Although SBRB or other BRBs has been widely used in building structures for seismic resistance (Fig. 1), a brace that can provide both energy dissipation and re-centering properties to structural systems in large earthquakes may be an alternative in seismic design. Therefore, a steel dual-core self-centering brace (DC-SCB) has been proposed to achieve the goal: dissipating seismic energy but not increasing residual drifts of frame structures. The DC-SCB can be installed in steel frames by using a dual-gusset configuration scheme to enhance the out-of-plan stability of gussets [ 7, 19].

To investigate the seismic performance of a steel frame with DC-SCBs, a prototype three-story DC-SCBF was designed based on ASCE standard [ 16] and available research works. A DC-SCB subassemblage was first tested to evaluate its mechanics and cyclic behavior; a full-scale one-story one-bay DC-SCBF subassembly specimen was then tested eight times to evaluate its system performance, damage progresses in the DC-SCB, beam or column, and post-earthquake behavior of the system. Subassemblage and braced frame tests confirmed that the DC-SCB performs well as the mechanics. Although the maximum axial force of the DC-SCB was around 1700 kN in the subassemblage and frame tests in this study, the maximum axial force of the DC-SCB that was ever tested was around 6000 kN [ 8], indicating a reliable force transfer mechanism in multiple seismic loadings. The initial PT force in the DC-SCB decreased 10% after Phase 2 test; no reduction of the initial PT force was observed from Phase 3 to 8 tests. Friction force in the DC-SCB decreased 14% after all eight tests due to gradual abrasion of the asperity of brass shim plates; this reduction might be acceptable because the one-story, one-bay DC-SCBF subassembly specimen had experienced several increasing cyclic loading tests and multiple DBE and MCE loading tests. Except for energy dissipation, no differences could be found from the hysteretic responses of the DC-SCB in Phase 3 and 8 tests. Moreover, yielding or local buckling in the beam increase residual drifts of the DC-SCBF subassemblage tests, especially at bottom floors. If the beam-to-column connection and column base are pinned, only for the shear transfer, residual drifts of the frame and repair of the columns and beams can be further minimized [ 20].

References

[1] Uang C M, Nakashima M. Steel buckling-restrained braced frames. In Earthquake Engineering from Engineering Seismology to Performance-based Engineering (Chapter 16). Bozorgnia Y, Bertero VV. (eds), CRC Press LLC: Boca Raton, FL; 16–1–16–37, 2003

[2] Fahnestock L A, Sause R, Ricles J. Analytical and experimental studies on buckling restraint braced composite frames. In: Proceedings of the International Workshop on Steel and Concrete Composite Construction. Rep. No. NCREE-03–026, Taiwan University, Taiwan, China, 2003

[3] Roeder C W, Lehman D E, Christopulos A. 2006. Seismic performance of special concentrically braced frames with buckling restrained braces. In: Proceedings of the 8th U.S. National Conference on Earthquake Engineering. Earthquake Engineering Research Institute, Oakland California, Paper No. 1503

[4] Tsai K C, Hsiao B C, Wang K J, Weng Y T, Lin M L, Lin K C, Chen C H, Lai J W, Lin S L. Pseudo-dynamic tests of a full scale CFT/BRB frame-Part I: Specimen design, experiment and analysis. Earthquake Engineering & Structural Dynamics, 2008, 37(7): 1081–1098

[5] Chen C, Gong H, Chou C C. Seismic behavior and application of buckling-restrained braces in China and Mainland Taiwan. In: the 14th World Conference on Seismic Isolation, Energy Dissipation and Active Vibration Control of Structures, San Diego, USA, 2015

[6] Chou C C, Chen S Y. Subassemblage tests and finite element analyses of sandwiched buckling-restrained braces. Engineering Structures, 2010, 32(8): 2108–2121

[7] Chou C C, Liu J H, Pham D H. Steel buckling-restrained braced frames with single and dual corner gusset connections: seismic tests and analyses. Earthquake Engineering & Structural Dynamics, 2012, 41(7): 1137–1156

[8] Chou C C, Chung P T, Cheng Y T. Experimental evaluation of large-scale dual-core self-centering braces and sandwiched buckling-restrained braces. Engineering Structures, 2016, 116: 12–25

[9] AISC. Seismic Provisions for Structural Steel Buildings. American Institute of Steel Construction, Chicago, IL, 2010

[10] Christopoulos C, Tremblay R, Kim H J, Lacerte M. Self-centering energy dissipative bracing system for the seismic resistance of structures: development and validation. Journal of Structural Engineering, 2008, 134(1): 96–107

[11] Miller D J, Fahnestock L A, Eatherton M R. Development and experimental validation of a nickel-titanium shape memory self-centering buckling-restrained brace. Engineering Structures, 2012, 40: 288–298

[12] Chou C C, Chen Y C. Development of steel dual-core self-centering braces: quasi-static cyclic tests and finite element analyses. Earthquake Spectra, 2015, 31(1): 247–272

[13] Chou C C, Chen Y C, Pham D H, Truong V M. Steel braced frames with dual-core SCBs and sandwiched BRBs: mechanics, modeling and seismic demands. Engineering Structures, 2014, 72: 26–40

[14] Chou C C, Chung P T. Development of cross-anchored dual-core self-centering braces for seismic resistance. Journal of Constructional Steel Research, 2014, 101: 19–32

[15] Chou C C, Wu T H, Beato A R O, Chung P T, Chen Y C. Seismic design and tests of a full-scale one-story one-bay steel frame with a dual-core self-centering brace. Engineering Structures, 2016, 111: 435–450

[16] Standard A S C E. Minimum Design Loads for Building and Other Structures. ASCE, 2010

[17] Tsai K-C, Lin B-Z. Development of an object-oriented nonlinear static and dynamic 3D structural analysis program. Center for Earthquake Engineering Research, Taiwan University, China, 2003

[18] Chou C C, Lo S W, Liou G S. Internal flange stiffened moment connections with low-damage capability under seismic loading. Journal of Constructional Steel Research, 2013, 87: 38–47

[19] Chou C C, Liou G S, Yu J C. Compressive behavior of dual-gusset-plate connections for buckling-restrained braced frames. Journal of Constructional Steel Research, 2012, 76: 54–67

[20] Chou C C, Chen J H. Analytical model validation and influence of column bases for seismic responses of steel post-tensioned self-centering MRF systems. Engineering Structures, 2011, 33(9): 2628–2643

RIGHTS & PERMISSIONS

Higher Education Press and Springer-Verlag Berlin Heidelberg