Anda di halaman 1dari 17

Engineering Structures 150 (2017) 98114

Contents lists available at ScienceDirect

Engineering Structures
journal homepage: www.elsevier.com/locate/engstruct

Simple design method of structure with metallic yielding dampers based


on elasticplastic response reduction curve
Hua Shen a,b, Ruifu Zhang a,, Dagen Weng a, Cong Gao a, Hao Luo a, Chao Pan a,c
a
Research Institute of Structural Engineering and Disaster Reduction, Tongji University, Shanghai 200092, China
b
College of Civil Engineering, Nantong Vocational University, Nantong 226007, China
c
College of Civil Engineering, Yantai University, Yantai 264005, China

a r t i c l e i n f o a b s t r a c t

Article history: Metallic yielding dampers (MYDs) can effectively improve the seismic performance of structures and are
Received 15 January 2017 used in many practical engineering applications. In this paper, a simple elasticplastic design method of a
Revised 14 July 2017 structure with MYDs is proposed based on the elasticplastic response reduction curve (EPRRC). In
Accepted 16 July 2017
widely used traditional design methods based on the elastic response reduction curve, the elasticplastic
behavior of the main structure is neglected. However, such methods may overestimate the vibration con-
trol effect of MYDs, especially their effect on acceleration reduction, which is demonstrated in this study.
Keywords:
The EPRRC, which can directly reflect the relationship between MYDs characteristic parameters and the
Metallic yielding damper
Elasticplastic design method
responses of the structure with MYDs in the elasticplastic range, is proposed to improve the reliability of
Response reduction curve a structure with MYDs. Then, the corresponding design procedures are presented. To illustrate and verify
Response spectrum the effectiveness of the proposed design method, a six-story reinforced concrete frame is analyzed by
Energy dissipation time history analysis. In conclusion, the proposed design method for the structure with MYDs can ade-
quately satisfy the seismic performance targets under different seismic levels.
2017 Elsevier Ltd. All rights reserved.

1. Introduction vibration control. It was confirmed that such devices can effec-
tively improve the structural seismic performance [1,516].
In traditional aseismic structures, seismic energy is absorbed by According to the type of yielding mechanism, MYDs can be catego-
the plastic deformation of the structural members such as beams, rized into flexural plate systems, torsional bar dampers, yielding
columns and braces, which may cause structural damage under ring dampers, extrusion devices, shear panel dampers, etc. [17].
earthquake events [1]. Such damage is usually difficult to check, In addition to added damping and stiffness devices and bucking-
expensive to repair, and may pose a serious threat to peoples lives restrained braces, many new MYDs have been developed in recent
and properties. To prevent such damage, the concept of control years, such as a dual-function metallic damper [18], a shear and
strategies has been introduced into the field of structural engineer- flexural yielding damper [19], and dual-pipe dampers [20].
ing. Among the commonly used structure control methods, passive MYDs have been found to have many advantages such as an
energy dissipation technology has been proven to be an effective easy manufacturing process, convenient installation, relatively
method and is widely used in building and bridge structures [2,3]. low cost, and extreme reliability; therefore, they have been
A metallic yielding damper (MYD) is a type of hysteretic dam- adopted in many engineering practices [2125]. Generally speak-
per made of metal that utilizes the plastic deformation of hys- ing, by incorporating MYDs into a structure, the stiffness and
teretic materials, such as mild steel, to dissipate the input damping of the structure can be significantly adjusted [26]. More
seismic energy. Kelly et al. [4] first introduced MYDs into a building seismic energy is absorbed because of the increased damping.
structure to dissipate seismic energy after conducting conceptual Hence, the seismic responses of a structure with MYDs can be
and experimental studies. In the ensuing years, considerable clearly reduced. However, according to the response spectrum the-
efforts were made to investigate the seismic performance of the ory, the added stiffness of the structure with MYDs may result in a
structure with MYDs and use these devices in the field of structural decrease in displacement and an increase in acceleration. Because
of the great influence of MYDs on the structural performance and
Corresponding author.
the coupling effect of the added damping and stiffness, the practi-
cal design of a structure with MYDs is more complex than that of a
E-mail address: zhangruifu@tongji.edu.cn (R. Zhang).

http://dx.doi.org/10.1016/j.engstruct.2017.07.047
0141-0296/ 2017 Elsevier Ltd. All rights reserved.
H. Shen et al. / Engineering Structures 150 (2017) 98114 99

structure with viscous dampers, where only the damping charac- 2. Response reduction curve
teristic is adjusted [17].
To adopt MYDs in practical engineering applications and meet 2.1. Response reduction indexes
performance targets, the development of appropriate design meth-
ods becomes very necessary. Tsai et al. [8] proposed a design The peak responses of an elastic single-degree-of-freedom
method that incorporates the capacity design concept and the opti- (SDOF) system can be predicted by a common elastic response
mal strength ratio of the main structure into MYDs, and Clark et al. spectrum. The mathematical relationship of the peak displacement
[27] suggested using the equivalent static lateral-force provision to Sd , pseudo-velocity Spv , and pseudo-acceleration Spa can be
design a structure with MYDs. The Japan Society of Seismic Isola- expressed as follows [45]:
tion proposed a design method based on the elastic response  2
reduction curve (ERRC) to design a structure with MYDs [28], T T
Sd T; f Spv T; f Spa T; f 1
which has been widely applied. However, these design methods 2p 2p
do not consider the inelastic behavior of the main structure. Under where T and f are the fundamental period and the damping ratio of
a moderate or severe seismic event, the main structure may expe- the SDOF system, respectively.
rience an obvious inelastic behavior, and these design methods The forcedisplacement relationship of an elastic main struc-
may not be suitable in such cases. To overcome these deficiencies, ture with an MYD is illustrated in Fig. 1, where the elastic stiffness
two well-known displacement-based design methodsa coeffi- of the main structure is K f , and the paralleled MYD is idealized as a
cient method based on FEMA 273 [29] and a capacity-spectrum bilinear model with elastic stiffness K a , yielding force F ay , and dis-
method based on ATC-40 [30]are used to design structures with
placement uay . The total curve is bilinear with the initial stiffness
MYDs. Because of the limitations involved in the application of
K 0 calculated by adding the elastic stiffness of the main structure
MYDs to existing structures and the complexity in the static non-
K f and that of the MYD K a . Once the MYD reaches its yielding dis-
linear analysis of structures with MYDs, the direct displacement-
placement, the corresponding total lateral force is F y , and the
based design method [3133] was developed. However, structural
increase in the lateral force is resisted only by the main structure
damage is caused by both the maximum displacement and the
until the maximum displacement umax . The equivalent linear
accumulated energy dissipation during seismic events. Thus, the
method is adopted herein to consider the inelastic behavior of
energy-based design method was developed for structures with
the structure with an MYD. Hence, the bilinear system can be sub-
MYDs [3436]. Considering the actual mechanical interaction
stituted with an equivalent linear model with stiffness K eq , as
between the damper and the brace, Lomiento et al. [37] proposed
shown in Fig. 1. Then, the seismic response of the structure with
a graphical design method. Some optimization algorithms and con-
the MYD can be estimated using the design response spectrum
trol theories are also used to design structures with MYDs, that is,
with the equivalent period T eq and the damping ratio feq . This esti-
to determine the number of MYDs along with their locations and
mation procedure is discussed further in this section.
characteristic parameters [3844]. Although some design methods
The equivalent period T eq can be determined as:
that consider the inelastic behavior of a structure have been pro-
s s
posed, they are still very complex for practical use. Therefore, it Kf la
is necessary to develop a design method for structures with MYDs T eq Tf Tf 2
K eq la K a =K f
that is simpler and more practical but also considers the structural
inelastic behavior. where la umax =uay is the maximum ductility coefficient, and T f is
In this study, while considering the elasticplastic behavior of the period of the main structure.
the main structure, the existing ERRC is expanded to the elastic Considering the inherent damping ratio f0 and the hysteretic
plastic response reduction curve (EPRRC) based on the equivalent damping ratio of the MYD fa , the total equivalent damping ratio
linear theory and the response spectrum theory. Within the elas- of the elastic structure with the MYD feq can be expressed as
ticplastic range, the EPRRC can directly reflect the relationship follows:
between the MYDs characteristic parameters and the responses
of a structure with MYDs. Then, the results of comparisons feq f0 fa 3
between the EPRRC and the ERRC are analyzed, and a simple design where f0 is usually set to 5% for an RC structure and 2% for a steel
method based on the EPRRC is proposed. Lastly, the effectiveness of structure. The term fa can be calculated as follows [30]:
the proposed design method is verified by applying it to a six-story
reinforced concrete (RC) frame and performing a time history ED
fa 4
analysis. 4pES

Fig. 1. Forcedisplacement relationship of the elastic structure with the MYD.


100 H. Shen et al. / Engineering Structures 150 (2017) 98114

where ED and ES are the energy dissipated by damping in one cycle


T eq
of reciprocating movement and the maximum strain energy of the Rd Df 15
Tf
structure with the MYD, respectively. Then, using the maximum
ductility coefficient la , fa can be obtained as follows [46]: The expression for Ra is the same as that in Eq. (14).

2K a =K f la  1
fa 5 2.2. ERRC
pla K a =K f la
While under seismic excitation, considering that the maximum Substituting Eqs. (2) and (12) into Eq. (13) for the SPS, Rd can be
ductility coefficient varies from 1 to la because of the randomness rewritten as:
of the seismic response, the equivalent damping ratio feq can be s !
calculated as follows: Df la 1 la
Rd  16
Z la 2 la K a =K f 1 K a =K f la K a =K f
1 2 1 pla  1
feq f0 fa dl f0 ln 6
la 1 la pp lpa Substituting Eqs. (12) and (16) into Eq. (14) for the SPS, Ra can
be rewritten as:
1 s !
p 7 Df 1 la K a =K f
1 K a =K f Ra  1 17
2 1 K a =K f la
To compare the seismic responses of the structure with the
MYD with those of the structure without the MYD, two response Rd and Ra can be expressed as a function of K a =K f and la from Eqs.
reduction indexes Rd and Ra are defined as follows [46]: (16) and (17), respectively. Rd and Ra for the MPSs and the long-
period structures can also be derived in the same manner. The
Sd T eq ; feq T eq Spv T eq ; f0 expressions for Rd and Ra are based on the assumption that the main
Rd Df 8
Sd T f ; f0 T f Spv T f ; f0 structure is elastic and can predict the control effects on the seismic
responses of the main structure with an MYD. Hence, the Rd  Ra
Spa T eq ; feq T f Spv T eq ; f0 curve is usually called an ERRC.
Ra Df 9 In this study, to derive the ERRC, the design response spectrum
Spa T f ; f0 T eq Spv T f ; f0
provided by the Chinese code for the seismic design of buildings
s [47] is used and can be expressed as follows:
1 25f0 8
Df 10 10g2  0:45T 0:45amax g; 0 < T 6 0:1 s
1 25feq >
>
>
>
>
< g2 amax g; 0:1 s < T 6 T g
where Rd and Ra are the displacement reduction index and the Spa T g c 18
pseudo-acceleration (base shear) reduction index, respectively, > T g2 amax g;
>
> T g < T 6 5T g
>
>
:
and Df represents the average reduction of the seismic responses. g2 0:2c  g1 T  5T g amax g; 5T g < T 6 6 s
Considering the characteristics of pseudo-velocity and pseudo-
acceleration spectra within different fundamental periods, the for- 0:05  f
mulas of Rd and Ra can be derived according to the following c 0:9 19
0:3 6f
procedures:
0:05  f
(i) For short-period structures (SPSs), Spa is a constant, and Spv is g1 0:02 20
4 32f
a linear function. The equivalent period of the structure with
the MYD changes from the initial fundamental period T 0 to
0:05  f
T eq . The spectral velocity of the structure with the MYD g2 1 21
0:08 1:6f
can be evaluated according to the following equation:
Z T eq where T g is the characteristic period of the design response spec-
1
Spv T eq ; f0 Spv T; f0 dT 11 trum; amax is the maximum value of the seismic influence coeffi-
T eq  T 0
cient; c is the exponential index of the descending branch; g1 is
T0

the adjusting factor of the linear descending branch; g2 is the damp-


where Spv is the average pseudo-velocity from T 0 to T eq , and
ing modification factor; and g is the acceleration caused by gravity.
T 0 can be calculated as follows:
In practice, the fundamental periods of most building structures
s s
are between 0.1 s and 5T g ; therefore, the discussions here are
Kf 1
T0 Tf Tf 12 mainly limited to an SPS 0:1s < T 6 T g and an MPS
K0 1 K a =K f
T g < T 6 5T g . Based on the concept of an ERRC, the acceleration
Substituting Eq. (11) into Eqs. (8) and (9), Rd and Ra can be response spectrum is first transformed into a pseudo-velocity
obtained as follows: response spectrum according to Eq. (1), and then, Rd of the ERRC
is derived as follows:
T eq T 0 T eq
Rd Df 13
Tf 2T f (i) For SPSs 0:1 s < T 6 T g , Rd can be expressed as:

 2  
Tf Sd T eq ; feq g2eq T eq T eq T 0
Ra Rd 14 Rd 22
T eq Sd T f ; f0 g20 T f 2T f

(ii) For medium-period structures (MPSs) and long-period where g20 and g2eq are the damping modification factors of f0
structures, Spv is a constant. Therefore, Eq. (8) can be simpli- and feq , respectively.
fied as follows: (ii) For MPSs T g < T 6 5T g , Rd can be expressed as:
H. Shen et al. / Engineering Structures 150 (2017) 98114 101

   
Sd T eq ; feq g2eq T eq 2ceq T f c0 ceq trilinear model can be obtained, as shown in Fig. 3. A comparison
Rd 23
Sd T f ; f0 g20 T f Tg of this figure with Fig. 1 indicates that the linear main structure
is replaced by a bilinear model. In Fig. 3, the yielding force, yielding
where c20 and ceq are the exponential indexes of the descend- displacement, elastic stiffness, and ratio of post-yielding stiffness
ing branches of f0 and feq , respectively. are denoted by F fy , ufy , K f , and b, respectively. K 0f and K 0eq are the
According to Eqs. (14), (22), and (23), for f0 0:05, secant stiffnesses of the main structure and the total curve, respec-
1:5 6 la 6 15, 0 6 K a =K f 6 4, and T f =T g 1:5, the ERRCs of the tively. The other parameters have already been described in the
SPS and the MPS can be plotted in Fig. 2(a) and (b), respectively. previous section.
For a constant la , as K a =K f increases, both Ra and Rd of the SPS When minor inelastic deformations are sustained in the main
decrease (Fig. 2(a)), which implies that the control effect on struc- structure, the assumption of superposition can still be used
tural vibration is enhanced. When K a =K f exceeds a certain value approximately. To maximize the energy dissipation capacity of
and increases further, the improvement in the vibration control the MYDs and keep the main structure in the minor inelastic range,
effect may be limited. For a constant K a =K f , as la increases, Rd it is necessary to achieve sufficiently large displacement ductility
changes slightly whereas Ra decreases significantly. lmax lfy =lay , as recommended by Vargas et al. [33]. Based on
A comparison of Fig. 2(a) and (b) shows some similarities as their studies, the range lmax P 5 was proposed for most cases,
well as some differences between them. As K a =K f increases, Rd of i.e., la P 5lf , for the same umax .
the MPS decreases significantly, as in the case of the SPS. For a
smaller value of the constant la , Ra increases as K a =K f increases, 2.3.2. Equivalent period
whereas for a larger value of the constant la , Ra first decreases For the structure with the MYD, the MYD should be designed to
and then increases as K a =K f increases, as in the case of the MPS. yield before the main structure, as shown in Fig. 3. As the displace-
ment u increases, the equivalent period T 0eq of the inelastic main
2.3. EPRRC structure with the MYD may vary as follows:

Under frequent seismic events, the main structure can remain (i) When u 6 uay , both the main structure and the MYD are in
elastic, whereas MYDs can yield and dissipate seismic energy even the elastic range. Therefore, T 0eq is equal to T 0 and can be cal-
though their deformation may be small. However, under moderate culated according to Eq. (12).
or severe seismic events, both the main structure and the MYDs (ii) When uay < u < ufy , the main structure is in the elastic range,
may yield and enter the plastic range. In such cases, the stiffness
whereas the MYD is in the plastic range. Thus, T 0eq is equal to
of the structure with the MYDs decreases significantly and its
T eq and can be calculated according to Eq. (2).
dynamic performance changes accordingly. Based on the concept
(iii) When u P ufy , both the main structure and the MYD are in
of performance-based seismic design, a structure should satisfy
the performance targets under different intensity-level seismic the plastic range. With the plastic development of the main
events. Clearly, the seismic responses of the structures based on structure, the fundamental period increases as its stiffness
the elastic design method cannot be guaranteed to meet the per- declines. By using the equivalent linear method, K 0f and K 0eq
formance targets of the structures in the plastic range. Therefore, can be considered to be the equivalent stiffnesses of the
it is worthwhile to develop an EPRRC to consider the main struc- main structure and the structure with the MYD, respectively.
tures inelastic behavior and establish the theoretical basis for a Then, the equivalent period of the main structure T 0f and that
simple elasticplastic design method for structures with MYDs. of the structure with the MYD T 0eq can be calculated as
follows:
2.3.1. Forcedisplacement relationship s s
In this study, the main structure is considered to be inelastic Kf lf
T 0f T f T 24
and is represented by a bilinear model. With the superposition of K 0f 1 blf  b
f

the bilinear model of the main structure and that of the MYD, a

Fig. 2. ERRCs (f0 0:05).


102 H. Shen et al. / Engineering Structures 150 (2017) 98114

Fig. 3. Forcedisplacement relationship of the inelastic structure with the MYD.

s v Z lf
u 1
K 0f u 1 blf  b f0f f0f ldl
T 0eq T 0f T 0f t 25 lf
K 0eq 1 blf  b 1 Ka
a Kf
1
( )
2qa  m lf n
respectively, where lf umax =ufy and a ufy =uay are defined ln ln n 30
pm  nlf mlf  1 nn=m m
as the ductility coefficient of the main structure and the ratio
of the yielding displacement of the main structure to that of
q a  ab K a =K f 31
the MYD, respectively.

2.3.3. Equivalent damping ratio m ab 32


Under moderate or severe seismic events, both the main struc-
ture and the MYD may enter the plastic range and dissipate seismic n a K a =K f 33
energy. Therefore, by including three parts, namely, the inherent
damping ratio f0 , the nonlinear hysteretic damping ratio of the
2.3.4. Elasticplastic response reduction indexes
MYD f0a , and that of the main structure f0f , the total equivalent
As mentioned above, Rd and Ra are used as the evaluation
damping ratio of the structure with the MYD can be expressed as
indexes of the dampers control effect, although they are usually
follows:
calculated by neglecting the inelastic behavior of the main struc-
f0eq f0 f0a f0f 26 ture. Once the main structure enters the inelastic range, it will start
to dissipate seismic energy. In such a case, the main structure itself
According to Eq. (4), following the calculation detailed in can also be seen as a special damper, and this advantageous effect
Appendix A, the two terms f0a and f0f with the maximum ductility should also be considered in the performance evaluation of struc-
coefficients la and lf , respectively, can be expressed as follows: tures. Therefore, considering the inelastic behavior of the main
2K a =K f la  1 structure, the total equivalent damping ratio of the structure with-
f0a 27 out the MYD feq can be expressed as follows:
pla K a =K f a bla  a
feq f0 ff 34
2aq1  blf  1
f0f 28
plf K a =K f a ablf  1 where ff is the nonlinear hysteretic damping ratio of the main
structure and is different from the damping ratio f0f in Eq. (26)
where q is the damping modification factor, and its values are listed
because of the difference in the maximum strain energies of the
in Table 1.
two systems. According to Eq. (4) and following the calculation
Under seismic excitation, the corresponding nonlinear hys-
detailed in Appendix A, ff with the maximum ductility lf can be
teretic damping ratios can be calculated as follows considering that
the ductility coefficient of the MYD ranges from 1 to la and that of calculated as follows:
the main structure ranges from 1 to lf because of the randomness 2q1  blf  1
of the seismic response: ff 35
( )
plf 1 blf  b
Z la
1 2K a =K f q bla qb=b q b
f0a f0a ldl ln  lnq b Under seismic excitation, considering that the maximum ductil-
la 1 pqla la b ity coefficient varies from 1 to lf because of the randomness of the
29 seismic response, the nonlinear hysteretic damping ratio of the
main structure ff can be expressed as follows:
Z lf
1 2q 1 blf  b
ff ff ldl ln 36
lf 1 pblf lbf
Table 1
Damping modification factor q referenced from ATC-40 [30]. Based on the parameters of the main structure T 0f and ff , R0d and
Structural behavior type Feature of the hysteretic loop q R0a are defined as elasticplastic response reduction indexes and
A Full 1 the R0d  R0a curve is called the EPRRC in this study. Then, the elas-
B Moderate 2/3 ticplastic displacement reduction index R0d can be derived as
C Poor 1/3
follows:
H. Shen et al. / Engineering Structures 150 (2017) 98114 103

Fig. 4. EPRRCs and ERRCs (f0 0:05).


104 H. Shen et al. / Engineering Structures 150 (2017) 98114

(i) For the SPS 0:1s < T 6 T g , R0d is expressed as: same as the parameters used to plot the ERRCs in Fig. 2. Under
the same conditions, the EPRRCs (solid lines) and the ERRCs (dot-
!
g2eq T 0eq T 0eq T 0
Sd T 0eq ; f0eq ted lines) are plotted together and compared, as shown in Fig. 4.
R0d 37 The figure indicates that the curve shape of the EPRRC is similar
0 
Sd T f ; ff g2f T 0f T 0f T f
to that of the ERRC for different SPS and MPS cases. Some differ-
where g2f and g2eq are the damping modification factors of ff ences are also found between the two sets of curves. For a conve-
nient evaluation of the differences, the displacement reduction
and f0eq , respectively.
ratio ud and the acceleration reduction ratio ua are respectively
defined as follows:
(ii) For the MPS T g < T 6 5T g , R0d is expressed as:
!2ceq !cf ceq R0d  Rd
g2eq T 0eq
Sd T 0eq ; f0eq T 0f ud  100% 40
R0d 38 R0d
Sd T 0f ; ff g2f T 0f Tg

where c2f and c2eq are the exponential indexes of the descend- R0a  Ra
ua  100% 41
ing branches of ff and f0eq , respectively. Ra
The elasticplastic acceleration reduction index R0a of both the For example, for K a =K f 1, the response reduction ratios of the
SPS and the MPS can be expressed as follows: SPS and those of the MPS are listed in Tables 2and 3, respectively.
!2 In the cases of the SPS and MPS, as la increases, ud increases and
T 0f goes from negative to positive, which means that for a large la , the
R0a R0d 39
T 0eq ERRC may overestimate the control effect of displacement. In con-
trast, the ERRC may underestimate the control effect of displace-
ment for a small value of la . ud decreases as lf increases, which
2.4. Comparisons between EPRRC and ERRC thereby implies that an improvement in the ductility of the main
structure enhances the control effect of displacement. As la and
By setting T f =T g 1:5, b 0:6, and q 2=3 (type B in Table 1) lf increase, ua increases in the cases of the SPS and MPS, which
and according to Eqs. (37)(39), a series of EPRRCs (lf 1:5, 2.0, means that the control effect on the structural acceleration accord-
and 3.0) of the SPS and the MPS can be obtained with the param- ing to the ERRC is overestimated and the extent of overestimation
eters f0 0:05, 1:5 6 la 6 15, and 0 6 K a =K f 6 4, which are the increases as la and lf increase, especially in the case of the MPS.

Table 2
Response reduction ratios of the SPS (K a =K f 1).

Reduction ratio lf la
1.50 2.00 3.00 4.00 8.00 15.00
ud 1.5 4% 3% 1% 1% 4% 6%
2.0 7% 5% 2% 0% 4% 6%
3.0 9% 7% 4% 1% 3% 6%
ua 1.5 2% 2% 3% 4% 6% 7%
2.0 3% 3% 4% 5% 7% 8%
3.0 4% 4% 5% 6% 7% 8%

Table 3
Response reduction ratios of the MPS (K a =K f 1).

Reduction ratio lf la
1.50 2.00 3.00 4.00 8.00 15.00
ud 1.5 1% 2% 5% 7% 10% 11%
2.0 3% 0% 4% 6% 10% 11%
3.0 5% 2% 3% 5% 9% 10%
ua 1.5 5% 7% 9% 10% 12% 12%
2.0 6% 8% 10% 11% 13% 13%
3.0 8% 10% 12% 13% 13% 13%

Table 4
Response reduction ratios of the SPS (la 2).

Reduction ratio lf K a =K f

0.30 0.50 1.00 2.00 3.00 4.00


ud 1.5 3% 3% 4% 3% 2% 1%
2.0 3% 3% 4% 3% 1% 0%
3.0 2% 3% 3% 2% 0% 2%
ua 1.5 3% 4% 6% 6% 6% 6%
2.0 4% 5% 7% 8% 8% 8%
3.0 4% 5% 7% 9% 10% 10%
H. Shen et al. / Engineering Structures 150 (2017) 98114 105

Table 5
Response reduction ratios of the MPS (la 2).

Reduction ratio lf K a =K f

0.30 0.50 1.00 2.00 3.00 4.00


ud 1.5 5% 7% 10% 11% 11% 10%
2.0 5% 7% 10% 10% 10% 9%
3.0 5% 7% 9% 10% 9% 8%
ua 1.5 6% 8% 12% 14% 15% 16%
2.0 6% 9% 13% 16% 18% 19%
3.0 6% 9% 13% 18% 20% 21%

On the other hand, consider the example where la 8. For this structural acceleration according to the ERRC is overestimated,
case, the response reduction ratios of the SPS and those of the MPS and the extent of overestimation increases as K a =K f and lf
are listed in Tables 4 and 5, respectively. In the case of the SPS and increase, especially in the case of the MPS.
the MPS, as K a =K f increases, ud initially increases and then The abovementioned analysis indicates that the ERRC may over-
decreases, which means that the control effect of the displacement estimate the vibration control effects of the MYD, especially their
according to the ERRC is overestimated, and the extent of overesti- effect on acceleration reduction, because the inelastic behavior of
mation initially increases and then decreases. As lf increases, ud the main structure is neglected. The overestimation can make the
decreases in the cases of the SPS and MPS, which implies that structure designed according to the ERRC unsafe. Therefore, to
the control effect of displacement can be improved by increasing ensure the safety of a structure with MYDs, it is necessary to con-
the ductility of the main structure. In addition, as K a =K f and lf sider the inelasticity of the main structure during a seismic event
increase, ua increases, which means that the control effect on the and develop a new design method based on the EPRRC.

Fig. 5. Design flowchart of the structure with the MYDs.


106 H. Shen et al. / Engineering Structures 150 (2017) 98114

3. Design method based on EPRRC where Sa and Sd are the spectral acceleration and displacement,
respectively; Q 1 and Q i are the base and ith story shear forces,
A flowchart of the proposed design method based on the EPRRC respectively; M is the equivalent mass; mi is the ith story mass;
is shown in Fig. 5, and the detailed design procedures are given as c1 is the modal participation factor in the first mode; ui;1 is the
follows: modal displacement of the ith story (the ith mass in the multiple-
Step 1: Obtain the capacity spectrum and the story stiffness of degrees-of-freedom system) in the first mode; un;1 is the modal dis-
the primary structure. A pushover analysis of the primary structure placement at the top of the structure in the first mode; and Dn is the
is carried out, and the relationship curves of the story shear force top story displacement.
Q i and the story drift di are obtained. Based on the Q i  di curves, The story stiffness K i can be expressed as follows:
the capacity spectrum can be obtained according to the following
equations: Q ei
Ki 46
Dei
Q1
Sa 42
M where Q ei and Dei are the ith story shear force and displacement
within the elastic range, respectively.
Dn
Sd 43 Step 2: Determine the stories where the MYDs should be
c1  ui;1 installed. The target structural performance point (TSPP) (Sdp , Sap )
P can be obtained according to the capacity spectrum and the design
mi  ui;1 2
M P 44 response spectrum [48], and then, the story drift distribution di can
mi  u2i;1 be determined using the TSPP and the Q i  di curve. The target
P story drift [d] is set according to the design requirement, and then,
mi  ui;1
c1 P 45 the ith story displacement reduction Rdi can be calculated as
mi  u2i1 follows:

Fig. 6. Plan and elevation of the RC frame.


H. Shen et al. / Engineering Structures 150 (2017) 98114 107

d
Rdi 47
di
when Rdi is less than 1.0. The stories with relatively large drifts,
which should be reduced, are selected to install the MYDs.
Step 3: Determine the parameters of the EPRRC. By using the
energy equivalent method, the equivalent bilinear representation
of the capacity spectrum associated with the TSPP is plotted, and
the yielding point (Sdy , Say ) can be determined. Then, the ductility
coefficient of the main structure lf and the ratio of post-yielding
stiffness b can be obtained according to the following equations:
Sdp Say
lf 48
Sdy Sap
Fig. 7. Trilinear hysteretic model for the plastic hinges.
Sap  Say Sdy
b 49
Sdp  Sdy Say
Table 7
The ith story yielding drift dyi of the Q i  di curves can also be Modal periods of ST0.
determined by using the energy equivalent method, and then the
ith story ductility coefficient lfi can be obtained as follows: Modals 1st modal 2nd modal 3rd modal
Period (s) 1.511 1.302 1.254
di
lfi 50
dyi
According to the design requirements and the performance of
practical products, the ith story yielding displacement of the MYDs
uai can be selected to meet the required condition, lai P 5lfi , and
then, the ductility coefficient of the MYDs of the ith story lai be cal-
culated as follows:
uai  hi d
lai 51
uai uai
where [uai ] and hi are the ith story target displacement of MYDs and
the ith story height, respectively.
Step 4: Determine the yielding force of the MYDs. For the given
maximum ductility coefficient lai of the EPRRC, it is possible to
obtain one or two optional values of the stiffness ratio K ai =K i ,
where K ai denotes the ith story elastic stiffness of the MYDs, and
it is easy to judge as to which one of these values is a reasonable
one from the EPRRC. Then, from the EPRRC, the ith story elastic Fig. 8. Capacity spectrum and its bilinear representation.
stiffness ratio K ai =K i is determined by using the values of lf , b,
Rdi , and lai . Finally, the ith story yielding force of the MYDs F ai is
calculated as follows:
Table 8
  Elastic stiffness distribution of ST0.
K ai
F ai K i uai K ai uai 52
Ki Story 1 2 3 4 5 6
Elastic stiffness K i (kN/mm) 141 170 173 171 164 140

4. Design example

4.1. Model information

A six-story RC frame is adopted to illustrate the proposed


design method. The plan and elevation of the RC frame are shown
in Fig. 6. The beam sections in the Y-direction and of the middle
span in the X-direction are 300  400 mm (width  height), and

Table 6
Model information.
P
Story Story height (m) Mass of story mi (t) Total mass of story mi (t)
6 3.6 594.5 594.5
5 3.6 515.7 1110.2
4 3.6 515.7 1625.9
3 3.6 515.7 2141.6
2 3.6 515.7 2657.3
1 5.5 558.7 3216.0
Fig. 9. Curves of the story shear force and the drift Q i  di .
108 H. Shen et al. / Engineering Structures 150 (2017) 98114

Table 9
Pushover analysis results of ST0 at the TSPP.

Story 1 2 3 4 5 6
Shear force (pushover) Q i (kN) 3140 2670 2304 1928 1422 743
Story drift (pushover) di (rad) 1/161 1/158 1/170 1/214 1/327 1/658
Target story drift [di ](rad) 1/200
Reduction ratio of displacement Rdi 0.81 0.79 0.85 1.07 1.64 3.29
Stories requiring dampers 1, 2, 3

Fig. 10. Plan and elevation of the MYD arrangement.

those remaining are 300  600 mm. The column sections of the X-direction are 2.5 kN/m2, and those remaining are 2.0 kN/m2.
first story are 450  450 mm, while the other columns are The roof dead and live loads are 6.5 kN/m2 and 2.0 kN/m2,
400  400 mm. The slabs of the floor and roof thicknesses respectively. Table 6 lists the mass of each story, considering the
are 120 mm. All the dead loads from the first to the fifth story entire dead load and 0.5 times the live load. The inherent damping
are 5.0 kN/m2. The live loads of the middle span in the ratio f0 of the RC frame is 0.05.
H. Shen et al. / Engineering Structures 150 (2017) 98114 109

TSPP (Sdu , Sau ) is found to be (66.5 mm, 94.4 gal), and the yielding
point (Sdy , Say ) is found to be (36.0 mm, 62.0 gal) by using the
energy equivalent method. Then, the ductility coefficient of ST0
lf and the ratio of post-yielding stiffness b are calculated to be
1.21 and 0.62 according to Eqs. (48) and (49), respectively.
Based on the Q i  di curve obtained from the pushover anal-
ysis, the story elastic stiffness distribution of ST0 is determined
according to Eq. (46) and the stiffness values are listed in
Table 8. The story drift distribution di of ST0 corresponding to
the TSPP is shown in Fig. 9, and the associated parameter
values are listed in Table 9. Because the target story drift is
[d] = 1/200 (dashed line in Fig. 9), the displacement reduction
Rdi can be determined according to Eq. (47). The MYDs should
be installed in the stories, where Rdi exceeds 1. Therefore, from
Table 9, it can be inferred that the MYDs should be installed at
the 1st-3rd stories and the MYD arrangement is shown in
Fig. 10.
The first model period of ST0 T 1 1:511 s is between
T g 0:4 s and 5T g 2 s; therefore, ST0 can be considered as an
MPS. By setting lf 1:21, b 0:62, f0 0:05, and the period
Fig. 11. Story demand points in the EPRRC (T 1 =T g 3:78).
ratio T 1 =T g 3:78, the EPRRC is plotted, as shown in Fig. 11.
In this study, the yielding displacement of the MYD uai is set
to 2 mm. Then, the ductility coefficient of the MYD lai is
The RC frame is modeled as a three-dimensional frame model. obtained according to Eq. (51), and the obtained coefficient
For the computational efficiency, assuming that bending yielding satisfies the condition lai P 5lfi . When lai and Rdi are provided,
occurs at the ends of members, plastic hinges and fiber segments the elastic stiffness ratio K ai =K i can be determined according to
are placed at the end of the beams and columns, respectively. the EPRRC. Its values are listed in Table 10, and the correspond-
Fig. 7 shows a trilinear model adopted for the bending moment ing story demand points are shown in Fig. 11. Further, the elastic
curvature backbone relationships of the plastic hinges, where stiffness K ai and the design yielding force F ay are determined
superscripts p and n denote positive and negative directions, according to Eq. (52), and the parameters of the MYDs are listed
respectively; M y and M u denote the yielding bending moment in Table 10.
and the hardening bending moment, respectively; and wy , wu , and
wx represent the yielding curvature, hardening curvature, and ulti- 4.3. Verification by time history analysis
mate curvature of the beams sections, respectively. The fiber seg-
ments are chosen to simulate the mechanical behavior of the 4.3.1. Seismic input
columns. The biaxial bending and PMM interaction can be auto- To verify the control effect of the additional MYDs, 12 different
matically calculated. seismic waves are chosen to carry out the time history analysis:
The compression strength of the concrete is 30 MPa, and the AW1 and AW2, which are artificial seismic waves, and NW1
yielding strengths of the longitudinal steel bar and the stirrup NW10, which are natural seismic waves. These natural seismic
bar are 400 and 235 MPa, respectively. The characteristic period waves are selected from the strong motion database of the Pacific
of the ground motion T g and the peak ground acceleration (PGA) Earthquake Research Center [4951]. All the seismic waves follow
are set as 0.4 s and 0.2 g, respectively. The RC frame without the the Chinese design response spectrum with a characteristic period
MYDs and the RC frame with the MYDs are denoted as ST0 and T g of 0.4 s, and the normalized design spectrum in the Chinese code
ST1, respectively. A model analysis is carried out for ST0, and the is expressed as follows:
first three modal periods are obtained and listed in Table 7. 8
>
> 1 11:5T; 0 < T 6 0:1 s
>
>
4.2. Damper arrangement and design < 2:25; 0:1 s < T 6 0:4 s
Spa  0:9 53
>
> 2:25 0:4 ; 0:4 s < T 6 2 s
A pushover analysis is conducted for ST0, and the capacity spec- >
> T
: 0:9
trum obtained according to Eqs. (42)(45) is shown in Fig. 8. The 2:250:2  0:02T  2; 0:2 s < T 6 6 s

Table 10
Parameters of the MYDs.

Story 1 2 3
Story parameters of the MYDs Yielding displacement uai (mm) 2 2 2
Target displacement [ uai ] (mm) 27.5 18 18
Ductility coefficient lfi 2.29 1.79 1.77
Ductility coefficient lai 13.75 9 9
Elastic stiffness ratio K ai =K i 1.13 0.97 0.60
Elastic stiffness K ai (kN/mm) 159.33 164.90 103.80
Design yielding force F ai (kN) 318.66 329.80 207.60
Number of dampers 4 4 4
Parameters of the MYDs for each damper Yielding force of each damper F (kN) 80 90 60
Yielding displacement u (mm) 2 2 2
Elastic stiffness of each damper K (kN/mm) 40 45 30
110 H. Shen et al. / Engineering Structures 150 (2017) 98114

Table 11
Details of natural waves.

Name RSN Year Magnitude Epicentral distance (km) Event Station


NW1 9 1942 6.50 56.88 Borrego El Centro Array #9
NW2 15 1952 7.36 38.42 Kern County Taft Lincoln School
NW3 28 1966 6.19 17.64 Parkfield Cholame-Shandon Array #12
NW4 68 1971 6.61 22.77 San Fernando LA-Hollywood Stor FF
NW5 93 1971 6.61 39.45 San Fernando Whittier Narrows Dam
NW6 131 1976 5.91 41.37 Friuli_Italy-02 Codroipo
NW7 138 1978 7.35 24.07 Tabas_Iran Boshrooyeh
NW8 169 1979 6.53 22.03 Imperial Valley-06 Delta
NW9 186 1979 6.53 35.64 Imperial Valley-06 Niland Fire Station
NW10 215 1980 5.80 15.84 Livermore-01 San Ramon Fire Station

Fig. 12. Normalized response spectra.

Fig. 13. Seismic response of ST0 (PGA = 0.2 g).

The details of the five natural seismic waves are listed in 4.3.2. Verification of the damper arrangement
Table 11, and the corresponding response spectra and the code Fig. 13 shows the story displacements and the story drifts of ST0
spectrum are shown in Fig. 12. According to the Chinese code for obtained by conducting the time history analysis at PGA = 0.2 g.
seismic design of buildings [47], the PGAs of the seismic waves Then, the average story drifts of ST0 are obtained. It is found that
are set to 0.07 g, 0.2 g, and 0.4 g corresponding to the frequent, the average story drifts at the 1st-3rd stories do not meet the story
moderate, and severe seismic events, respectively, before carrying target drift [d] = 1/200; therefore, it is necessary to install MYDs at
out time history analysis. the 1st-3rd stories.
H. Shen et al. / Engineering Structures 150 (2017) 98114 111

Fig. 14. Average story drifts and shear forces (PGA = 0.07 g).

Fig. 15. Average story drifts and shear forces (PGA = 0.2 g).

Fig. 16. Average story drifts and shear forces (PGA = 0.4 g).
112 H. Shen et al. / Engineering Structures 150 (2017) 98114

Table 12
Reduction ratio of the seismic responses between ST0 and ST1.

Seismic response PGA (g) ST0 ST1 Reduction ratio j


Story drift (rad) 0.07 1/408 1/712 74%
0.2 1/169 1/218 29%
0.4 1/71 1/89 26%
Story shear force (kN) 0.07 1565.51 988.39 58%
0.2 2839.89 2543.94 12%
0.4 3597.85 3474.06 4%

Fig. 17. Hysteretic loops of the MYDs.

4.3.3. Verification of the mitigation effect for a structure with MYDs. First, the existing ERRC method is intro-
ST1, which is designed according to the proposed method, is duced, which is then expanded to the EPRRC method by consider-
analyzed to investigate the mitigation effect by the time history ing the elasticplastic behavior of the main structure. Second, the
analysis. The 12 different seismic waves are used as the base exci- procedures of the proposed method are described, and the results
tation. The PGA is set as 0.07 g, 0.2 g, and 0.4 g, and the average of the EPRRC and ERRC are compared. Third, a six-story RC frame
story drifts and shear forces of ST1 and those of ST0 are shown model is analyzed to verify the effectiveness of the proposed
and compared in Figs. 1416. These figures indicate that MYDs design procedure. Based on the study results, the following main
can be used to effectively control the story drifts and shear forces, conclusions can be drawn:
and the story drifts of ST1 can adequately satisfy the performance
targets 1/550, 1/200, and 1/80 under frequent, moderate, and sev- (i) The existing ERRC method, in which the inelastic behavior of
ere seismic events, respectively. Hence, the effectiveness of the the main structure is neglected, is found to overestimate the
proposed method is verified. vibration control effects of a structure with MYDs, especially
For PGA values of 0.07 g, 0.2 g, and 0.4 g, the maximum story their effect on the acceleration reduction. Therefore, the
drifts and shear forces of ST0 and those of ST1 are listed in Table 12. inelastic behavior of the main structure should be consid-
For the convenient evaluation of the control effect, a comparison ered when evaluating the seismic performance of a structure
index is defined as follows: with MYDs.
(ii) When the inelastic behavior of the main structure is consid-
RST0  RST1
j  100% 54 ered, the proposed EPRRC can directly reflect the relation-
RST1 ship between the MYDs characteristic parameters and the
where RST0 and RST1 are the maximum seismic responses of ST0 and responses of the structure with the MYDs and, more impor-
ST1, respectively. tantly, can lead to a safer design of a structure with the
Table 12 indicates that both the maximum story drifts and the MYDs.
story forces of the main structure can be effectively reduced, espe- (iii) The proposed method can be used to effectively reduce both
cially in the case of PGA = 0.07 g, using the proposed method. the maximum story drifts and the story forces of the main
The hysteretic loops between the damping force and the actu- structure, and the seismic performance targets under differ-
ated displacement of three MYDs installed at the left side of the ent seismic levels can be adequately satisfied.
Y1 axis (Fig. 10) under the seismic wave NWX1 for PGA = 0.2 g
are shown in Fig. 17. The results indicate that the MYDs provide Acknowledgments
good energy dissipation during the seismic event.
This study was supported by the Shanghai Pujiang Program
5. Conclusion under Grant No. 17PJ1409200, the National Natural Science Foun-
dation of China under Grant No. 51308418, and by the Sichuan
In this study, based on the equivalent linear theory and the Department of Science and Technology under Grant No.
response spectrum theory, a simple design method is proposed 2016JZ0009.
H. Shen et al. / Engineering Structures 150 (2017) 98114 113

Appendix A. Calculation of the nonlinear hysteretic damping


ratios

When the main structure comes into the plastic range, the non-
linear hysteretic damping ratio of the MYD f0a and that of the main
structure f0f in the controlled structure and the nonlinear hysteretic
damping ratio of the main structure ff in the uncontrolled
structure are three important parameters to derive the EPRRC. In
this study, the detailed processes of the calculations for these
damping ratios are presented as follows according to Eq. (4):

(i) The nonlinear hysteretic damping ratio of the MYD f0a with
the maximum ductility coefficient la can be calculated as
follows, and the calculation diagram is shown as Fig. A1:

EDa 4uay K a umax  uay A1

1 Fig. A3. Calculation diagram of ff .


ESa umax uay K a ufy K f bK f umax  ufy  A2
2

EDa 2K a =K f la  1 (ii) The nonlinear hysteretic damping ratio of the main structure
f0a A3 f0f with the maximum ductility coefficient lf can be calcu-
4pESa pla K a =K f a bla  a
lated as follows, and the calculation diagram is shown as
where EDa is the energy dissipated by the MYD in one cycle of Fig. A2.
n
reciprocating movement, and ESa is the maximum strain
EDf 4 K f ufy bK f umax  ufy umax  K f u2fy  bK f umax  ufy 2
energy of the main structure with the MYD.
2ufy bK f umax  ufy 4K f 1  bufy umax  ufy A4

1
ESf umax uay K a ufy K f bK f umax  ufy  A5
2

EDf 2aq1  blf  1


f0f A6
4pESf plf K a =K f a ablf  1

where EDf is the energy dissipated by the main structure in


one cycle of reciprocating movement, and ESf is the maxi-
mum strain energy of the main structure with the MYD.
(iii) The nonlinear hysteretic damping ratio of the main structure
ff with the maximum ductility coefficient lf can be calcu-
lated as follows, and the calculation diagram is shown as
Fig. A3.

EDf EDf 4K f 1  bufy umax  ufy A7

1
Fig. A1. Calculation diagram of f0a . ESa umax ufy K f bK f umax  ufy  A8
2

EDf 2q1  blf  1


ff A9
4pESf plf 1 blf  b
where EDf is the energy dissipated by the main structure in
one cycle of reciprocating movement, and ESf is the maxi-
mum strain energy of the main structure without the MYD.

References

[1] Saeedi F, Shabakhty N, Mousavi SR. Seismic assessment of steel frames with
triangularplate added damping and stiffness devices. J Constr Steel Res
2016;125:1525. http://dx.doi.org/10.1016/j.jcsr.2016.06.011.
[2] Soong TT, Spencer Jr BF. Supplemental energy dissipation: stateoftheart
and stateofthepractice. Eng Struct 2002;24:24359. http://dx.doi.org/
10.1016/S0141-0296(01)00092-X.
[3] Pan C, Zhang R, Luo H, Li C, Shen H. Demand-based optimal design of oscillator
with parallel-layout viscous inerter damper. Struct Control Health Monit 2017:
e2051. http://dx.doi.org/10.1002/stc.2051.
[4] Kelly JM, Skinner RI, Heine AJ. Mechanics of energy absorption in special
devices for use in earthquake resistant structures. Bull N Z Soc Earthq Eng
Fig. A2. Calculation diagram of f0f . 1972;5:6287.
114 H. Shen et al. / Engineering Structures 150 (2017) 98114

[5] Skinner RI, Kelly JM, Heine AJ. Hysteretic dampers for earthquakeresistant [27] Clark P, Aiken I, Kasai K, Ko E, Kimura I. Design procedures for buildings
structures. Earthq Eng Struct Dyn 1974;3:28796. incorporating hysteretic damping devices. In: Proc. 68th annual convention,
[6] Skinner RI, Tyler RG, Heine AJ, Robinson WH. Hysteretic dampers for the Structural Engineers Association of California, Santa Barbara, California.
protection of structures from earthquakes. Bull N Z Natl Soc Earthq Eng [28] Manual JSSI. Design and construction manual for passively controlled
1980;3:2236. buildings. Tokyo, Japan: Japan Society of Seismic Isolation; 2007 [in Japanese].
[7] Whittaker A, Bertero V, Alonso J, Thompson C. Earthquake simulator testing of [29] FEMA 273. NEHRP guidelines for the seismic rehabilitation of
steel plate added damping and stiffness elements Report No. UCB/EERC-89/ buildings. Washington DC: Building Seismic Safety Council; 1997.
02. Berkeley, CA: Earthquake Engineering Research Center, University of [30] ATC-40. Seismic evaluation and retrofit of concrete building. Redwood City,
California; 1989. CA: Applied Technology Council; 1996.
[8] Tsai KC, Chen HW, Hong CP, Su YF. Design of steel triangular plate energy [31] Lin YY, Tsai MH, Hwang JS, Chang KC. Direct displacementbased design for
absorbers for seismicresistant construction. Earthq Spectra 1993;9:50528. building with passive energy dissipation systems. Eng Struct 2003;25:2537.
http://dx.doi.org/10.1193/1.1585727. http://dx.doi.org/10.1016/S0141-0296(02)00099-8.
[9] Dargush GF, Soong TT. Behavior of metallic plate dampers in seismic passive [32] Kim J, Seo Y. Seismic design of lowrise steel frames with buckling-restrained
energy dissipation systems. Earthq Spectra 1995;11:54568. braces. Eng Struct 2004;26:54351. http://dx.doi.org/10.1016/j.
[10] Matteis GD, Landolfo R, Mazzolani FM. Seismic response of MR steel frames engstruct.2003.11.005.
with lowyield steel shear panels. Eng Struct 2003;25:15568. http://dx.doi. [33] Vargas R, Bruneau M. Seismic design of multistory buildings with metallic
org/10.1016/S0141-0296(02)00124-4. structural fuses. In: Proc. 8th US national conference on earthquake
[11] Curadelli RO, Riera JD. Reliability based assessment of the engineering, San Francisco, California, USA.
effectiveness of metallic dampers in buildings under seismic [34] Choi H, Kim J. Energybased seismic design of buckling-restrained braced
excitations. Eng Struct 2004;26:19318. http://dx.doi.org/10.1016/j. frames using hysteretic energy spectrum. Eng Struct 2006;28:30411. http://
engstruct.2004.07.004. dx.doi.org/10.1016/j.engstruct.2005.08.008.
[12] Vargas R, Bruneau M. Effect of supplemental viscous damping on the seismic [35] Benavent-Climent A. An energybased method for seismic retrofit of existing
response of structural systems with metallic dampers. J Struct Eng frames using hysteretic dampers. Soil Dyn Earthq Eng 2011;31:138596.
2007;133:143444. http://dx.doi.org/10.1016/j.soildyn.2011.05.015.
[13] Kim D, Dargush GF, Basaran C. A cyclic twosurface thermoplastic damage [36] Habibi A, Chan RWK, Albermani F. Energybased design method for seismic
model with application to metallic plate dampers. Eng Struct 2013;52:60820. retrofitting with passive energy dissipation systems. Eng Struct
http://dx.doi.org/10.1016/j.engstruct.2013.02.030. 2013;46:7786. http://dx.doi.org/10.1016/j.engstruct.2012.07.011.
[14] Deng K, Pan P, Sun J, Liu J, Xue Y. Shape optimization design of steel shear [37] Lomiento G, Bonessio N, Braga F. Design criteria for added dampers and
panel dampers. J Constr Steel Res 2014;99:18793. http://dx.doi.org/10.1016/ supporting braces. Seism Isol Prot Syst 2010;1:5573.
j.jcsr.2014.03.001. [38] Gluck N, Reinhorn AM, Gluck J, Levy R. Design of supplemental dampers for
[15] Lee CH, Lho SH, Kim DH, Oh J, Ju YK. Hourglassshaped strip damper subjected control of structures. J Struct Eng 1996;122:13949.
to monotonic and cyclic loadings. Eng Struct 2016;119:12234. http://dx.doi. [39] Moreschi LM, Singh MP. Design of yielding metallic and friction dampers for
org/10.1016/j.engstruct.2016.04.019. optimal seismic performance. Earthq Eng Struct Dyn 2003;32:1291311.
[16] Basu D, Reddy PRM. A new metallic damper for seismic resilience: analytical [40] Dargush GF, Sant RS. Evolutionary aseismic design and retrofit of structures
feasibility study. Structures 2016;7:16583. http://dx.doi.org/10.1016/j. with passive energy dissipation. Earthq Eng Struct Dyn 2005;34:160126.
istruc.2016.06.011. [41] Apostolakis G, Dargush GF. Optimal seismic design of momentresisting steel
[17] Zhang C, Zhou Y, Weng DG, Lu DH, Wu CX. A methodology for design of frames with hysteretic passive devices. Earthq Eng Struct Dyn
metallic dampers in retrofit of earthquakedamaged frame. Struct Eng Mech 2009;39:35576.
2015;56:56988. [42] Parulekar YM, Reddy GR, Vaze KK, Ghosh AK, Kushwaha HS, Ramesh Babu R.
[18] Li HN, Li G. Experimental study of structure with dual function metallic Seismic response analysis of RCC structure with yielding dampers using
dampers. Eng Struct 2007;29:191728. http://dx.doi.org/10.1016/j. linearization techniques. Nucl Eng Des 2009;239:305461. http://dx.doi.org/
engstruct.2006.10.007. 10.1016/j.nucengdes.2009.08.016.
[19] Sahoo DR, Singhal T, Taraithia SS, Saini A. Cyclic behavior of shear- [43] Jensen HA, Sepulveda JG. On the reliabilitybased design of structures
and-flexural yielding metallic dampers. J Constr Steel Res 2015;114: including passive energy dissipation systems. Struct Safety 2012;34:390400.
24757. [44] Ribakov Y, Gluck J. Optimal design of ADAS damped MDOF structures. Earthq
[20] Mahjoubi S, Maleki S. Seismic performance evaluation and design of steel Spectra 2012;15:31730.
structures equipped with dualpipe dampers. J Constr Steel Res [45] Clougn RW, Penzien J. Dynamics of structures. 2nd ed. New York: McGraw-
2016;122:2539. http://dx.doi.org/10.1016/j.jcsr.2016.01.023. Hill; 1993.
[21] Martinez-Romero E. Experiences on the use of supplementary energy [46] Kasai K, Fu Y, Watanabe A. Passive control systems for seismic damage
dissipators on building structures. Earthq Spectra 1993;9:581625. mitigation. J Struct Eng 1998;124:50112.
[22] Perry CL, Fierro EA, Sedarat H, Scholl RE. Seismic upgrade in San Francisco [47] GB 50011-2010. Code for seismic design of buildings. Beijing: the Ministry of
using energy dissipation devices. Earthq Spectra 1993;9:55979. Housing and Urban-Rural Construction of the Peoples Republic of China;
[23] Kitagawa Y, Midorikawa M. Seismic isolation and passive responsecontrol 2010.
buildings in Japan. Smart Mater Struct 1998;7:5817. [48] Hao L, Zhang R. Structural safety redundancybased design method for
[24] Symans MD, Charney FA, Whittaker AS, Constantinou MC, Kircher CA, Johnson structure with viscous dampers. Struct Eng Mech 2016;59:82140.
MW, et al. Energy dissipation systems for seismic applications: current [49] PEER. PEER Ground Motion Database. Pacific Earthquake Engineering Research
practice and recent developments. J Struct Eng 2008;134:321. http://dx.doi. Center; 2016.
org/10.1061/(ASCE)0733-9445(2008)134:1(3). [50] Pan C, Zhang R, Luo H, Shen H. Baseline correction of vibration acceleration
[25] Pan P, Ye LP, Wei S, Cao HY. Engineering practice of seismic isolation and signals with inconsistent initial velocity and displacement. Adv Mech Eng
energy dissipation structures in China. Sci China Technol Sci 2016;8:111. http://dx.doi.org/10.1177/1687814016675534.
2012;55:303646. [51] Pan C, Zhang R, Luo H, Shen H. Target-based algorithm for baseline correction
[26] Whittaker AS, Bertero VV, Thompson CL, Alonso LJ. Seismic testing of steel of inconsistent vibration signals. J Vib Control 2017. http://dx.doi.org/10.1177/
plate energy dissipation devices. Earthq Spectra 1991;7:563604. 1077546316689014.