See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/348617294

Rocking isolation of bridge pier using shape memory alloy

Article  in  Bridge Structures Assessment Design and Construction · January 2021

DOI: 10.3233/BRS-200174

CITATIONS READS

4 393

4 authors, including:

Rajesh Rele Stergios Aristoteles Mitoulis

University of Surrey University of Birmingham
3 PUBLICATIONS   24 CITATIONS    119 PUBLICATIONS   1,527 CITATIONS   

SEE PROFILE SEE PROFILE

Subhamoy Bhattacharya
University of Surrey
315 PUBLICATIONS   5,953 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Extending the applications of Integral Abutment Bridges View project

Increasing the collapse time for non-engineered buildings View project

All content following this page was uploaded by Subhamoy Bhattacharya on 22 January 2021.

The user has requested enhancement of the downloaded file.

Bridge Structures 16 (2020) 85–103 85
DOI:10.3233/BRS-200174
IOS Press

Rocking isolation of bridge pier using shape

memory alloy
Rajesh R. Relea,∗ , Ranjan Balmukunda , Stergios A. Mitoulisb and Subhamoy Bhattacharyab
a R. R Consulting Engineers, Mumbai, India

PY
b University of Surrey, Guildford, UK

CO
Abstract. The conventional design philosophy of bridges allows damage in the pier through yielding. A fuse-like action
is achieved if the bridge piers are designed to develop substantial inelastic deformations when subjected to earthquake
excitations. Such a design can avoid collapse of the bridge but not damage. The damage is the plastic hinge formation formed
at location of maximum moments and stresses that can lead to permanent lateral displacement which can impair traffic flow
and cause time consuming repairs. Rocking can act as a form of isolation by means of foundation uplifting which act as a
mechanical fuse, limiting the forces transferred to the base of the structure. In this context, this paper proposes a novel resilient
controlled rocking bridge pier foundation, which uses elastomeric pads incorporated beneath the footing of the bridge piers
OR
and external restrainer in the form of shape memory alloy bar (SMA). The rocking mechanism is achieved by restricting the
horizontal movement of footing by providing stoppers at all sides of footing. The pads are designed to remain elastic without
allowing their shearing. The pier, the footing and the elastomeric pads are assumed to be supported on firm rigid concrete
sub base resting on hard rock. By performing nonlinear dynamic time history analysis in the traffic direction of the bridge,
the proposed pier with the novel resilient foundation is compared against a fixed-based pier and classical rocking pier (CC).
The proposed pier rocking on elastomeric pads and external restrainer (CP+SMA) has good re-centering capability during
TH

earthquakes with negligible residual drift and footing uplift. In this new rocking isolation technique, the forces in the piers
are also reduced and thus leading to reduced construction cost with enhanced post-earthquake serviceability.

Keyword: Bridges, pier, rocking isolation, stoppers, elastomeric pad, resilient, shape memory alloy bar
AU

1. Introduction The seismic demands of bridge pier can be reduced

Bridges are important structures as they carry a
roadway or railway and connect places across a city
or banks of a river. Failure and severe damage to
these lifeline structures in an earthquake can be cat-
astrophic as the access to other damage sites is pre-
vented and thus causes further loss of life. In the
current seismic design philosophy of bridges, the ear-
thquake resisting element is the bridge pier which
deforms inelastically to dissipate energy and prevent
collapse of the bridge while the superstructure is all-
owed to respond elastically to the strong ground
motion.
∗ Corresponding author.
Rajesh R. Rele, R. R Consulting Engi-
neers, Mumbai, India. E-mail: rajesh rele@rediffmail.com.
by increasing the natural period of vibration but at
the same time the robustness of structure should be
maintained by not making it too slender but ductile
enough to resist the strong earthquake motion. To
achieve this, we need to isolate the base itself so that
entire structure rocks on its foundation by means of
structural rocking. In order to avoid any large uplift
of foundation additional restrainers made up of smart
materials could be installed to have complete resilient
foundation which shall have the least damage when
subjected to very strong seismic motion.
Housner [1] first introduced the concept of pure
rocking of structures when survival of elevated water
tanks was observed during Chilean earthquakes in
1960. Beck and Skinner [2] proposed new tech-
nique to enhance the earthquake resistance of tall

1573-2487/20/$35.00 © 2020 – IOS Press and the authors. All rights reserved
86 R.R. Rele et al. / Rocking isolation of bridge pier using shape memory alloy

reinforced concrete bridges by using elastomeric

pad and energy absorbing devices to prevent over-
turning of the structure.
Many researchers have proposed to place central
tendon in the bridge pier along with external dissipa-
tors in order to minimize the damage during seismic
events. Mander and Cheng [3] developed a concept
termed Damage Avoidance Design (DAD) in which
bridge piers were allowed to rock. Unbonded ten-
dons were utilised for stability and ductility purposes
at rocking interface to prevent damage during move-
ment. Palermo et al. [4] proposed a hybrid bridge

PY
pier wherein the self-centering capacity was achieved
using prestress tendon and energy dissipation was
achieved using mild steel rebars. Liu and Palermo
[5] proposed concept of dissipative controlled rock-
ing (DCR) which was a combination of free rocking,

CO
post-tensioning and dissipative devices in order to
have least damage in the pier. The use of external en-
ergy dissipators (ED) for self-centering precast con-
crete (SCPC) was proposed by Guo et al. [6] having
post-tensioned basalt fiber-reinforced polymer (BF
RP) tendons. Sakai and Mahin [7] proposed a de-
sign wherein the longitudinal post-tensioning strands
OR
replace some of the mild steel bars and the seismic
performance of such partially prestressed columns
is investigated through a series of quasi-static and
dynamic analyses. Trono and Panagiotou [8], Da-
wood et al. [9] and Zatar et al. [10] used prestress-
TH

ing strands to study the seismic performance of

rocking piers to have minimum damage along with
some energy dissipating elements. Espinoza et al. Fig. 1. Typical bridge failures [18].
[11] performed shake table and centrifuge experi-
ment to study the rocking of bridge pier subjected to pier yielding, deck unseating and pounding of deck
AU

multi-directional earthquake loading. The rocking is [16, 17, 18].

achieved by placing the neoprene pad below the foot-
ing of reinforced concrete column. Marriott et al. [12,
13] & Gabriele et al. [14] have investigated numer-
ically and experimentally the rocking behaviour of
bridge structures to avoid major damages and forma-
tion of plastic hinges at the critical zones.
A rocking isolation technique was proposed by
Agalianos et al. [15]. In this technique, the pier was
allowed to uplift by not connecting the pier to the
foundation. In order to avoid sliding of pier, a recess
was suggested to promote the rocking phenomenon.
The poor seismic performance in majority of bri-
dge failures is due to the old elastic-based design phi-
losophy which is based on yielding of structures and
a lack of attention to structural detailing of bridges
piers and superstructure. The typical bridge fail-
ures are shown in Fig. 1 which are mainly due to
To reduce the damage to minimum level the foun-
dation of the structure should not be restrained but
allow the foundation to uplift so that the moments
can be released and transferred to other sacrificial ele-
ments. The damage to the pier and the residual drifts
post-earthquake can be significantly reduced by uti-
lizing uplift in the structural system since the pier is
subjected to bending moment that is only required
to uplift. This reduces the seismic demands on the
bridge pier and thus with the concept of rocking phe-
nomena, bridges can survive severe ground motions.
The conventional seismic design philosophy does not
allow uplift and thus possibly fail to resist severe
seismic excitations.
In this context, this article proposes a controlled
rocking resilient foundation which uses elastomeric
pad incorporated beneath the footing of bridge piers
 R.R. Rele et al. / Rocking isolation of bridge pier using shape memory alloy 87

and shape memory alloy bars as external restrainers.
The seismic performance in longitudinal direction of
bridge in terms of forces and residual drift in the pier
of proposed resilient foundation is compared with the
fixed base pier and classical rocking pier (CC). The
proposed resilient foundation showed enhanced post-
earthquake serviceability since residual drift were
negligible and thus pier yielding could be avoided.
The external restrainers demonstrated a high level of
energy dissipation in the form of superelastic hystere-
sis and controlled the movements of pier subjected to
relatively high seismic excitation.
3. Mechanics of rocking foundation on pads
and external restrainers
The uplift of footing having pads with external
restrainers is shown in Fig. 3. The two pads as seen
in elevation has lost contact in footing and only the
other two are in contact. An expression for resisting
moment and overturning moment for the proposed
rocking isolation technique is explained in this sec-
tion. As the foundation is supported on discrete pads,
total reactions of pads are considered to simplify the
expression obtained.

PY
Moment of Resistance (MR ) at corner point of
footing is as follows:
2. Resilient bridge foundation rocking on  
elastomeric pads and shape memory alloy L
MR = Ws −  + (WP + WF ) · L/2
bars 2

2.1. Description of proposed pier model

(CP+SMA)

The proposed controlled rocking foundation as

CO − RP1· c − RP2 · b + Rtsma · d + Rcsma · e

=W·
L
2
− WS ·  − RP1· c − RP2 · b

+ Rtsma · d + Rcsma· e (1)

shown in Fig. 2(a) rests on firm concrete rock base
OR
with elastomeric pads placed first on a rough surface
and then the footing along with pier is erected. The Moment of Overturning (Mo) is as follows:
recess made in the base are for stoppers to restrict
horizontal sliding and thus allow only rocking of Mo = P.(H + H1 ) − P.H2
the foundation without any translational movements.
Tanθ = r /H = H1 /(L/2 − a) (2)
In practice, a cushion in the form of rubber pad is
TH

attached vertically alongside of the stopper in order

to avoid damage to footing due to pounding of the two
concrete surfaces. For future replacement of bearings
jack locations are shown in Fig. 2(b). The diameter of
flat jack [19] shall be 350 mm having load carrying
AU
By equilibrium of moments, the lateral force (P)
required to initiate uplift in case of proposed control
rocking is as follows:

capacity of 1100 kN can be used to lift the system by
6 to 10 mm so that pads can be replaced. The pads
in the center of footing are replaced first after the
stoppers and pads at periphery are removed.
2.2. Assumptions and limitations in the proposed
technique
1. The horizontal translation of footing and pad is
NOT allowed by means of stoppers.
2. Footing simply rests on the pad as there is no
connection between pad and footing.
3. Uplift of footing is allowed and hence the sys-
tem has rocking effect.
4. Strata on which the system rests are assumed to
be unyielding and have a firm base such as hard
rock or pile cap.
W. L2 − WS  − Rp2 b + Rtsma .d + Rcsma .e
P = .
H + H 1 − H2
(3)
On the same basis, the lateral force (PCLR ) required
to initiate uplift in case of classical rocking pier sys-
tem [Fig. 2(a)] is as follows:
W. L2 − WS  − RC a
PCLR = (4)
H + H 1 − H2
Thus, it can be seen from equations (3 and 4) that
the proposed rocking system is more stable due to
enhanced moment of resistance given by the exter-
nal restrainers and thus the uplift can be controlled.
Also, the plastic impacts can be avoided as in case
of classical rocking as the proposed pier rocks on the
elastomeric pads placed beneath the footing of pier.
88 R.R. Rele et al. / Rocking isolation of bridge pier using shape memory alloy

PY
CO
(a)
OR
TH
AU

(b) (c)

Fig. 2. Proposed Resilient Bridge Foundation (a) Elevation (b) Bottom Plan of footing (Section A-A) (c) Top plan of footing (Section B-B).

4. Numerical modelling of pier models
In this study a lumped mass approach is adopted
to introduce the proposed concept while the effects
of abutments have not been explicitly modelled and
for which a full bridge model would be appropri-
ate. The controlled rocking pier bridge foundation
is modelled in CSi Bridge 2017 [20] and the results
are compared with the conventional monolithic pier
[Fig. 4(a)] and classical rocking pier [Fig. 4(b)].
The classical rocking pier consists of precast foot-
ing which simply rests on concrete slab without any
connection between the footing and the slab while the
recess in concrete slab prevents the sliding of footing
 R.R. Rele et al. / Rocking isolation of bridge pier using shape memory alloy 89

PY
CO
OR
(a) (b)

Fig. 3. Mechanics of rocking isolation technique (a) CP+SMA pier model (b) CC pier model.
TH
AU

(a) (b) (c)

Fig. 4. Studied Configurations (a) Conventional monolithic fixed base pier (FIX) (b) Classical rocking pier (CC) (c) Rocking pier on pads
with SMA (CP+SMA).
90 R.R. Rele et al. / Rocking isolation of bridge pier using shape memory alloy
or the walking off phenomenon [21]. The parameters
adopted for comparison with the conventional pier
are residual drift, period and forces in piers during
a seismic event. It was assumed that the footing is
resting on rock base and soil modelling is not con-
sidered in the present study as the soil is considered
to be stiff enough, which is a typical requirement in
isolated bridges. A nonlinear stage construction case
was defined wherein SMA bars were installed after
the deformation of pads due to self-weight of pier,
footing and dead load of the superstructure. The non-
linear time history analysis was performed using the
stiffness at end of this nonlinear stage construction
case. The dead load mass of the superstructure having
area of deck as 9.85m² and superimposed dead load
load of 4.55 t/m which includes crash barrier load
and wearing coat for a span of 30 m was assigned as
CO
concentrated load of 875 t on the pier top which was
activated during the construction stage analysis per-
formed in the computer programme. For pier models
shown in Fig. 4(a), (b) & (c) the pier top was con-
sidered to be monolithically connected to the deck
(failures of deck unseating and pounding are avoided
using integral connection) and to achieve the pier
OR
fixity at top, the rotations about transverse axis is
restrained for all cases while the pier top is free to
move along all axes [22–24]. All the pier models are
assumed to rest on rock base.
TH
4.1. Material and geometry of pier models
A conventional monolithic wall type pier of size
0.9 m x3.8 m (Fig. 5) longitudinally and transversely
respectively having height of 10 m with spread foot-
AU
ing dimension of 3 m x 5 m x 1.5 m where the first
dimension is along the direction of traffic of the bridge
is considered in the analysis while the second dimen-
sion is in the transverse direction and third dimension
is the thickness of the footing. This geometry of
pier and footing are the most common dimensions in
bridge design of wall type pier [25] and hence were
adopted in this study of all the three pier models.
The material of pier and footing are C35 for concrete
Fig. 5. Cross-section of pier with steel ratio, ρl =As /Ac =1 %.
with reinforcing steel of 1 percent of the cross-section
area of pier conforming to BS 4449 [26]. The values
of inertia are on basis of the full section. For crack-
ing behavior, section designer was used by allowing
modification factor as per Eurocode 8 [41].
4.2. Nonlinear modelling of piers
To simulate post-yield behaviour of piers, a con-
centrated plastic hinge is assigned to the frame el-
ement of the pier. Deformation beyond the elastic
limit occurs only within the hinges modelled at the
PY
top and the bottom of pier for fixed and rocking pier.
Inelastic behaviour is obtained through integration of
the plastic strain and plastic curvature which occurs
within the pre-defined hinge length. To capture the
coupled axial and bending behaviour, hinge is assig-
ned to the piers at relevant locations with the input
hinge model being the moment-curvature graph
(Fig. 6) and the hinge length of 1.15 m is calculated as
per the equation given by Priestley [27]. The idealized
moment-curvature graph is obtained by balancing the
areas between the actual and the idealized M-ϕ, where
ϕ is the curvature, curves beyond the first reinforcing
bar yield point as per Caltrans seismic design cri-
teria [28]. The confinement of reinforced concrete
sections has been taken into account using the Man-
der [29] confined model to represent the stress-strain
behaviour of the concrete core. The models of the
proposed controlled rocking pier (CP+SMA), clas-
sical rocking pier (CC) and fixed pier is shown in
Figs. 7, 8, and 9 respectively. In these models the
pier is modelled with 10 frame elements while the
footing is modelled with 120 solid elements after
Fig. 6. Moment-curvature relationship for pier with 1 percent steel.
 R.R. Rele et al. / Rocking isolation of bridge pier using shape memory alloy
a check on mesh sensitivity was performed. In the
classical rocking pier model, to simulate the separa-
tion between the precast footing and concrete slab,
gap elements were used at each nodes of footing
[30]. These gap elements represent compression only
springs and implicitly allow varying rocking pivots,
a continuous support force and the possibility of pla-
nar impacts [31, 32]. The stiffness of gap element
chosen is 1E7 kN/m. This value is based on sensitiv-
ity analysis performed in which it was observed that
beyond a particular value of stiffness, change in bend-
ing moment of the pier and the period of the system
was negligible and the system responded in a rigid
manner. In order to simulate the stoppers, the nodes
are restraint by setting UX = UY=0 at middle height
of footing
CO5. Elastomeric pads supporting the pier
OR
TH
Fig. 7. Numerical model of controlled rocking bridge foundation
AU
with pads and external restrainers (concrete to pad with SMA bars
CP+SMA model).
Fig. 8. Numerical model of classical rocking bridge pier founda-
tion (concrete to concrete CC model).
91
PY
Fig. 9. Numerical model of fixed pier (FIX).
footing
A total of 12 pads with four at each ends of footing
and four at center of footing are modelled using fric-
tion isolators available in the CSi Bridge. The friction
isolators have coupled friction properties for shear
deformations and carries only compression. This par-
ticular non-linear link element was chosen since the
footing simply rests on bearings and thus compres-
sion is only allowed in the link element. Although,
the friction will not have much contribution in the
overall behavior, it is included in the modelling by
setting coefficient of friction as 0.8 [33]. The fric-
tion isolator model is based on hysteretic behaviour
proposed by Wen [34] and recommended for base
isolation by Nagarajaiah [35]. The size of bearing is
425 mm x425 mm with height of 170 mm which con-
sists of five rubber layers of 30 mm thickness and four
steel shims of 5 mm thickness was selected after the
design of pad was made as per Eurocode EN 1337-
Part 3 [36]. This particular size of bearing was chosen
to control any large vertical initial compression of pad
due to dead load of deck.
The vertical (kv ) and horizontal stiffness (kh )
of bearing is evaluated using following equations:
(Naeim and Kelly [37, 38]):
Ec· A
kv =  ; (5)
ti
G.A
kh =  ; (6)
ti
92 R.R. Rele et al. / Rocking isolation of bridge pier using shape memory alloy

Ec .B
Ec = ; (7)
Ec + B

Ec = 6.73 GS 2 (8)
Where, G is the shear modulus of the bearing having
value of 0.7 MPa, B is the bulk modulus as 2000 MPa
and S is the shape factor which for square pad is a/4t
where “a” is size of rubber pad and t is the thickness
of each pad. The initial vertical deformation of pad
with superstructure load is found to be 12.2 mm and
Fig. 10. Analytical model of SMA restrainer of 40 mm diameter
the stress in pad is found to be 4.66 MPa against
and unit length used in CSi Bridge.

PY
permissible stress of 10–25 MPa as per Eurocode EN
1337-Part 3.
The parameters to model shape memory alloy
5.1. Purpose of elastomeric pads bars are shown in Table1. These properties are ad-
apted from DesRoches and Delemont [39]. By using

CO
1. Pads isolate the base of footing and thus increase
the flexibility of the system.
2. Period of structure can be controlled by varying
the sizes of pad to suit the seismic demand.
3. Reduces the seismic demands and thus piers
are subjected to less force in regions of high
seismicity.
OR
4. Enhances recentering of pier and thus negligible
permanent drifts.
5. Plastic impacts between two concrete surfaces
are avoided as in case of conventional rocking.
TH
these stress-strain values, an analytical model in CSi
Bridge is created to have an idealized force-defor-
mation curve of 40 mm diameter SMA bar as shown
in Fig. 10 which shows the primary features of the
superelastic effects of shape memory alloy bars. The
force-deformation relationship is characterized by an
elastic region, a long horizontal plateau, followed by a
significant increase in stiffness which demonstrates a
high-level energy dissipation and a super elastic hys-
teresis. The SMA bars are modelled using a double
link element. It is possible to simulate superelastic
SMA behavior in CSI programs [20] using multiple

6. Modelling of shape memory alloy bars as
external restrainers
SMAs are unique materials with a paramount
potential for various application in bridges. The nov-
AU
link elements. The first element is multi linear elastic
element (MLE) as shown in Fig. 11(a) using Pivot
model [40] to define the hysteresis loop, and a multi
linear plastic element as shown in Fig. 11(b) (PLE) is
used to shift the hysteresis loop away from the origin.

elty of this material lies in its ability to undergo large Three shape memory alloy bars of diameter 40 mm
deformations and return to its un deformed shape composed nickel and titanium (Ni-Ti) alloy was cho-
through stress removal (super elasticity) or heating sen on each side of the footing (in the longitudinal
(shape memory effect). In particular, Ni-Ti alloys direction) and was connected to footing by stiff arms.
have distinct thermo mechanical properties including The number of SMA bar depends upon the uplift of
super elasticity, shape memory effect and hysteresis footing required and on the peak ground acceleration
damping. and thus is a trial and error process. The chosen num-
ber and diameter avoids excessive uplift of footing
Table 1 as in case of conventional rocking and hence were
Constitutive material properties for NiTi based SMA bar selected. The length of SMA is calculated consid-
Parameters Stress Strain ering the initial deformation of the pad. The length
Austenite to martensite 523 MPa 0.01164 of SMA bar to be installed after deformation of pad
starting stress (1) is 0.3 m considering initial deformation of pad under
Austenite to martensite 588 MPa 0.062395 dead load which is 12.2 mm. This particular length of
finishing stress (2)
Martensite to austenite 241 MPa 0.0498 bar was chosen to avoid buckling of bar and to avoid
starting stress (4) large uplift of the footing. With large length of bar,
Martensite to austenite 225MPa 0.005755 the stress-strain curve of SMA bar does not get cap-
finishing stress (3)
tured showing distortion in the curve and hence the
 R.R. Rele et al. / Rocking isolation of bridge pier using shape memory alloy 93

PY
Fig. 11. Modelling of SMA bar (a) Multi linear elastic link element (b) Multi linear plastic link element.

length of bar is limited to 0.3 m. The study is mainly

confined for excitation induced in the bridge in the
weak dimension of the pier. If the system were also

CO
to be studied in the transverse direction, SMA bars
would be in that direction too.

6.1. Purpose of external restrainers (SMA bars)

1. Reduces the pier horizontal displacements in

OR
case of bridges subjected to high vibrations.
2. Act as sacrificial element in case of very high
magnitude of earthquakes.
3. Controls the rocking behaviour and excessive
foundation uplifts when piers are subjected to
Fig. 12. Response spectra of accelerograms compatible to ground
excessively large ground accelerations.
TH

Type C-dependent Euro code 8 1 elastic spectra (PGA = 0.6 g).

7. Selection of earthquake excitations applied only in the longitudinal direction of the bridge
as the pier is weak in this direction.
The fixed pier (FIX), classical rocking pier (CC) As the seismic demand is function of period and
AU

and the proposed rocking pier (CP+SMA) were anal-
ysed for seven real accelerograms compatible to
ground Type C-dependent Eurocode 8–1 elastic spec-
tra. The peak ground accelerations (PGA) selected
was 0.60 g to represent relatively high seismic excita-
tions by scaling [41, 42] the real accelerograms to the
same PGA. The response spectra of the seven anal-
ysed accelerograms are shown in Fig. 12. The seven
acceleration time histories are shown in Fig. 13(a) to
13(g) for 0.6 g are then matched to ground Type C-
dependent Eurocode 8–1 elastic spectra using Seismo
Match programme [43]. These time histories are
imposed at the base of the model, which is free to
move along x-x axis which is the longitudinal direc-
tion of the bridge. The duration of all the time history
analyses is 55 secs to allow for the pier models to
balance after the ground motions for assessing their
post-earthquake condition. The seismic motion is
hence the stiffness of pier cannot be ignored in the
transverse direction where the pier would behave
more like a cantilever structure [46]. However, in the
present study the seismic response of the proposed
foundation is only limited to excitations applied in
the traffic direction.
In bridges, the dead loads are very high and gen-
erally the excitations in vertical directions are not
always critical depending on the excitation force and
thus only horizontal excitations in the weak direction
is only considered in this study.
8. Results and discussion
The effective fundamental period for fixed pier was
0.62 s. The CC pier had period of 0.73 s while the
proposed pier CP+SMA had a period of 0.87 s. Thus,
94 R.R. Rele et al. / Rocking isolation of bridge pier using shape memory alloy

PY
CO
OR
TH

Fig. 13. Acceleration time histories matched to ground Type C-dependent Euro code 8–1 elastic spectra for 0.6 g (a) Imperial (b) Hollister
(c) Kobe (d) Kocaeli (e) Kozani (f)Loma Prieta (g) Northridge.
AU

(a) (b) (c)

Fig. 14. (a) First mode of proposed controlled rocking pier (T = 0.87 s) (b) Second mode of proposed controlled rocking pier (T = 0.33 s) (c)
Third mode of proposed controlled rocking pier (T = 0.22 s).
 R.R. Rele et al. / Rocking isolation of bridge pier using shape memory alloy 95

Table 2
Time history analysis for 7 different earthquakes for FIX pier model (0.6 g)
EQ PIER DISP (mm) B.M (kN-m) AF(kN) SF(kN) Footing uplift Residual Drift (%)
UX UZ TOP BOT MAX MIN (mm) (mm)
HOLLISTER 78 0 29792 29341.2 9412.9 9412.9 4772.6 0 0.82
IMPERIAL 92 0 27538 27538 9412.9 9412.9 5723.2 0 0.39
KOACELI 65 0 28028 27459.6 9412.9 9412.9 4302.2 0 0.097
KOZANI 62 0 27283.2 26646.2 9412.9 9412.9 3978.8 0 0.23
KOBE 123 0 38857 38053.4 9412.9 9412.9 5625.2 0 0.105
LOMA 72 0 31801 31193.4 9412.9 9412.9 4517.8 0 0.108
NORTHRIDGE 64 0 29576.4 29086.4 9412.9 9412.9 4890.2 0 0.302
Average 79.4 30410.8 29902.6 9412.9 9412.9 4830 0 0.3

by isolating the base, the natural period of system
elongates which is advantages for seismic design of
the structure.
The first three modes of only proposed controlled
rocking pier is shown in Fig. 14(a) to 14(c). The
PY
8.1. Comparison of bending moments and shear
forces in pier
The average of the maximum values of responses
obtained from Table 2, 3 and 4 is summarized in

CO
modal participating mass ratio is significant for the
first mode which is 0.93 and for other two modes
are negligible thus the first mode is mainly important
where the structure oscillates in the direction of traffic
on the bridge.
Tables 2, 3 and 4 shows the values of responses that
are calculated based on the seven acceleration time
OR
histories for PGA of 0.6 g for all three different piers
considered in this study.
Table 5. It can be observed from Table 5 that by
using rocking as form of isolation, the forces in the
pier are reduced. A preliminary check to the numbers
obtained from the computer programme is shown in
Appendix A.
The bending moment at top of CP+SMA pier and
CC pier is reduced by 45% and 30% respectively
when compared to fixed pier for PGA of 0.6 g. A
significant reduction in bending moment is observed

Table 3
Time history analysis for 7 different earthquakes for CC pier model (0.6 g)
TH

EQ PIER DISP (mm) B.M (kN-m) AF (kN) SF (kN) Footing uplift Residual Drift (%)
UX UZ TOP BOT MAX MIN (mm) (mm)
HOLLISTER 95 9.1 18551.4 15582 –14925 –6076 3410.4 20 0.038
IMPERIAL 145 17.1 25548.6 15180.2 –22452 –2440.2 4145.4 36 0.0094
KOACELI 68 3.7 19355 13622 –14788 –6125 3371.2 9.6 0.017
AU

KOZANI 67 3.5 19639.2 13592.6 –17630 –5027.4 3361.4 9.1 0.163

KOBE 168 18.6 22990.8 15885.8 –20129 –3802.4 3733.8 39 0.111
LOMA 152 14.2 20668.2 14357 –18532 –3782.8 3586.8 30.9 0.072
NORTHRIDGE 100 6.5 22961.4 14357 –20139 –2753.8 3675 15 0.015
Average 113.6 10.4 21387.8 14653.8 –18371 –4286.8 3612 22.8 0.061

Table 4
Time history analysis for 7 different earthquakes for CP+SMA pier model (0.6 g)
EQ PIER DISP (mm) B.M (kN-m) AF(kN) SF(kN) Footing uplift Residual Drift (%)
UX UZ TOP BOT MAX MIN (mm) (mm)
HOLLISTER 105 –12.1 14327.6 9917.6 10525.2 7977.2 2205 0 0.0359
IMPERIAL 159 –12.2 20119.4 13024.2 10280.2 8369.2 2842 6 0.093
KOACELI 95 –12.2 14376.6 9104.2 10397.8 8369.2 2009 0 0.045
KOBE 181 –12.1 17110.8 14278.6 10231.2 8702.4 2822.4 9.2 0.195
KOZANI 104 –11.54 14602 9976.4 10564.4 8045.8 2293.2 0 0.012
LOMA 186 –12.2 19198.2 14562.8 10152.8 8633.8 2979.2 8.35 0.0194
NORTHRIDGE 147 –12.2 15738.8 12808.6 10192 8506.4 2489.2 2.87 0.0042
Average 139.6 –12.1 16496.2 11953.2 10334.8 8372 2520 3.8 0.058
B.M = Bending moment in pier; AF = Axial force in pier; SF = Shear force in pier Ux = Displacement in traffic direction; Uz = Displacement
in vertical direction.
96 R.R. Rele et al. / Rocking isolation of bridge pier using shape memory alloy

Table 5
Average of the maximum values of the seismic loading for PGA 0.6 g
PGA 0.6g
Fixed pier Classical rocking Controlled rocking
Pier (CC) pier (CP+SMA)
Horizontal movement at pier 79 114 140
top (mm) pier top (mm)
Residual drift % 0.30 0.061 0.058
Footing uplift (mm) 0 22.8 3.8
Axial forces (kN) max /min –9413/–9413 –18371/–4286 –10334.8/–8372
Shear force (kN) 4830 3612 2520
B.M pier bottom (kN-m) 29902 14654 11953
B.M pier top (kN-m) 30410 21388 16496

PY
CO
OR
TH
AU

Fig. 15. Time history comparison of bending moment (B.M) of the proposed pier for Imperial earthquake of PGA 0.6 g (a) with Fix pier
(top B.M) (b) with Fix pier (bottom B.M) (c) with CCpier (top B.M) (d) with CC pier (bottom B.M).

at pier bottom as the foundation is allowed to uplift
and thus moments are released at the base of pier.
The bending moment at the bottom in the classical
rocking pier and the proposed controlled rocking pier
are reduced to more than half the value of bend-
ing moment in the conventional fixed pier. Rocking
motion also reduces the shear force in the pier. The
reduction in shear force for CP+SMA pier model is
47% while for CC pier is 25 % when compared to the
FIX pier model.
The time history comparison of bending moment
and shear force for Imperial earthquake is shown in
Figs. 15 and 16 respectively.
8.2. Comparison of axial forces in pier
It is observed the FIX and CP+SMA piers had neg-
ligible axial load fluctuations while the same is not
observed in case of classical rocking (CC) pier. The
axial force (A.F) in the pier for the CC model is 2.0
 R.R. Rele et al. / Rocking isolation of bridge pier using shape memory alloy 97

PY
Fig. 16. Time history comparison of shear force of the proposed pier (CP+SMA) for Imperial earthquake of PGA 0.6 g (a) with Fix Pier (b)
with CC Pier.

CO
OR
TH

Fig. 17. Time history comparison of axial force of the proposed pier (CP+SMA) for Imperial earthquake of PGA 0.6 g (a) with Fix Pier (b)
with CC Pier

times larger than the maximum axial force in the pro- for PGA of 0.6 g. The horizontal displacement in the
posed CP+SMA pier for PGA of 0.6 g as seen from CP+SMA pier was only 23% more than the CC pier
AU

Table 5. The axial fluctuation for the CC pier is due the
collision of the foundation on the concrete sub-base
and hence such system should be modified to avoid
such large impact forces. In the CP+SMA pier model,
such large axial fluctuations that may create tension
in piers are not observed as the external restrainers
holds down the footing when subjected severe seismic
excitations like 0.6 g. The time history comparison
of axial force for Imperial earthquake is shown in
Fig. 17.
8.3. Comparison of pier displacements and
residual drifts
The horizontal displacement at pier top of CC pier
and CP+SMA pier, which are related to deck dis-
placements, was 1.44 and 1.77 times the fixed model
as the external restrainers (SMA bars) controlled the
possible excessive pier horizontal displacements. The
time history comparison of pier drifts for Imperial
earthquake is shown in Fig. 18. The residual drift of
0.30 percent was observed in FIX pier model while
the CC and CP+SMA piers had almost full recenter-
ing capacity and did not experience any permanent
drift for PGA of 0.6g as seen in Table 5. Thus, rocking
isolation have enhanced post-earthquake serviceabil-
ity since after an earthquake the structure almost
comes back to its original position without any major
permanent displacements.
8.4. Comparison of footing uplift
The classical rocking pier model exhibited a mean
uplift of 23 mm at edges of precast footing and this
98 R.R. Rele et al. / Rocking isolation of bridge pier using shape memory alloy

PY
Fig. 18. Time history comparison of drift (%) of the proposed pier (CP+SMA) for Imperial earthquake of PGA 0.6 g (a) with Fix Pier (b)
with CC Pier.

CP+SMA pier had all pads in contact [Fig. 20] with

CO
negligible mean uplift of 3.8 mm as shown in Table 5
for PGA of 0.6 g. The Fig. 21 shows variation of
footing uplift for CC and CP+SMA pier models.

8.5. Comparison of stresses in elastomeric pads

OR
It can be seen from Table 6 which shows average
of maximum vertical stress in pad for seismic loading
with PGA of 0.6 g that outer pads are subjected to 45
% more stresses than inner ones for CP+SMA pier
model. This is due the fact during uplift the loads gets
TH

transferred to the outer ones and hence the increase in

the stress is justified. However, the SMA bars restricts
Fig. 19. Footing uplift of rocking pier for Imperial Earthquake the uplift and thus stress in the pad is in permissible
0.6 g CC pier model, t11.59 s. limit of maximum 25 MPa as mentioned in Eurocode
EN 1337-Part 3.
AU

Table 6
Average of maximum vertical stress in pad for the seismic
loading of PGA 0.6 g
PAD LINK ID MAX STRESS IN
PAD FOR CP+SMA
MODEL (MPa)
2 (outer pad) 11.25
43 (outer pad) 11.40
46 (outer pad) 11.25
47 (outer pad) 11.28
44 (outer pad) 11.43
Fig. 20. Footing uplift of rocking pier for Imperial Earthquake 5 (outer pad) 11.28
0.6 g, CP+SMA pier model, t = 11.72 s. 16 (inner pad) 7.86
20 (inner pad) 7.92
22 (inner pad) 7.86
would lead to separation of precast footing and the 23 (inner pad) 7.85
stiff concrete slab below it as 83percent area of foot- 21 (inner pad 7.91
17 (inner pad) 7.85
ing had lost contact as shown in Fig. 19. The proposed
 R.R. Rele et al. / Rocking isolation of bridge pier using shape memory alloy 99

PY
Fig. 21. Time history of footing uplift for Imperial Earthquake 0.6 g (a) CC pier (b) Proposed pier (CP+SMA).

by moment rotation curves. Generally, loops passing

CO
always through the axis of origin is good indica-
tion of self-centering capacity of the system [33]. It
has been observed from the moment rotation curves
that CC pier model and the proposed pier has good
recenetring capacity while the fixed pier did not show
such a behaviour in which some residual rotation is
OR
observed. The proposed CP+SMA pier and the CC
pier showed low energy dissipation compared to the
fixed pier and this is an acknowledge effect in rocking
isolation of piers [44, 45].

Fig. 22. Plan showing link ID of pad for both CP model and
TH

CP+SMA model. 8.7. Behaviour of SMA bars

8.6. Comparison of moment rotation curves Figure 24 shows the superlastic behaviour of SMA
bars in which the process of yielding and succes-
The energy dissipation capacity is being investi- sive increase in stiffness allow SMA bars to dissipate
AU

gated by plotting moment rotation curves as shown in large amount of energy with superelastic hysteresis
Fig. 23. This can be quantified based on area enclosed while reducing the footing uplift when compared to

(a) (b)

Fig. 23. Moment rotation comparison for Imperial EQ (0.6 g) (a) Fixed based pier with proposed pier (b) Classical rocking pier (CC) with
proposed pier
100 R.R. Rele et al. / Rocking isolation of bridge pier using shape memory alloy
Fig. 24. Stress-strain curve for SMA bar in proposed pier for
Imperial Earthquake 0.6 g.
the CC model. Also, it can be seen that restrainers
do not have symmetrical response for the Imperial
earthquake acclerogram (PGA 0.6 g) where the right
restrainers (SMA bars) resist more stress (673 MPa)
than the left ones (620 MPa) because the earthquake
pulses are not symmetrical. Note that the bars only
OR
act in tension since they have been designed for the
same as no response in compression zone is observed.
8.8. Comparison of ductility demands for the
pier models
TH
The comparison of average ductility demand (␮c)
of the pier from nonlinear dynamic time history
analysis is shown in Table 7 only for comparative
assessment for the piers considered in this study. The
AU
lower value of ductility demand is indication of flex-
ible foundation [28]. The proposed pier (CP+SMA
model) has least ductility demand as compared to
other pier models. For the FIX pier model the ductil-
ity demand is 86 % more than the CP+SMA model.
The ductility demand of CC model has reduced to
19% when compared to FIX model due to some flex-
ibility achieved on account of concrete to concrete
rocking.
Table 7
Average Ductility Demand of various pier models from
non-linear time history analysis
PARAMETER FIX CC CP+SMA
Peak Drift
 peak (%) 0.79 1.4
Yield Drift, y(%) 0.10 0.23
Ductility Demand, ␮d 7.90 6.08
9. Conclusions
This paper proposes a controlled rocking resilient
pier foundation which uses elastomeric pads at the
base of footing and shape memory alloy bars as exter-
nal restrainers. The rocking mode in pier is promoted
by restricting the sliding movement of footing by
means of stoppers around the footing. The proposed
rocking CP+SMA pier has been compared with the
fixed pier model and classical rocking (CC) pier on
basis of displacements, drifts and forces in piers. The
three pier models (CP+SMA, CC and FIX) are sub-
PY
jected to seven horizontal seismic excitations in the
longitudinal direction of bridge. Based on the analysis
performed the following conclusions were drawn:
1. The proposed rocking pier on pad with exter-
nal restrainers (CP+SMA model) is subjected
CO to reduced uplifts as compared to the conven-
tional rocking. It has re centering capability
since residual drifts are almost negligible and
thus seems to be viable solution for controlled
rocking isolation.
2. Classical rocking pier (CC model) showed neg-
ligible residual drift but is subjected to large
uplift of footing which is greater than 33% and
is subjected to large axial fluctuations. Thus, the
classical rocking would not be ideal solution
where strong ground motions are expected.
3. The external restrainers which are the SMA bars
showed super elastic hysteresis controlling the
uplift of footing in the proposed resilient bridge
pier foundation.
4. The proposed rocking isolation has reduced the
bending moment and shear forces in the pier by
nearly 50% when compared to the conventional
fixed base pier system. Thus, seismic demands
have reduced in the proposed system.
5. Based on advanced non-linear analysis, it can
be concluded that the proposed controlled rock-
ing resilient bridge pier foundation (CP+SMA
model) is promising and advantageous espe-
cially for areas prone to a relatively high seismic
zones as it has enhanced post-earthquake ser-
viceability. Thus, lifeline structures like bridges
would remain functional even after severe seis-
mic excitations.
Acknowledgments
1.29
0.31 The initial guidance on the concept of rocking
4.16
structures on spread foundations was provided by
 R.R. Rele et al. / Rocking isolation of bridge pier using shape memory alloy
George Mylonakis (University of Bristol, UK) is
highly valued and greatly appreciated by the authors.
References
[1] Housner GW. The behaviour of inverted pendulum struc-
tures during earthquakes. Bulletin of the Seismological
Society of America. 1963;35(2):403-17.
[2] Beck JL, Skinner RI, The Seismic Response of a Reinforced
Concrete Bridge Pier Designed to Step. International Jour-
nal of Earthquake Engineering and Structural Dynamics.
1974;2:343-58.
[3] Mander JB, Cheng C-T. Seismic resistance of bridge
piers based on damage avoidance design. Technical Report
NCEER-97-0014, Buffalo, NY. 1997.
[4] Palermo A, Pampanin S, Calvi GM, Use of ‘controlled
rocking’ in the seismic design of bridges. Proc., 13th
World Conf. on Earthquake Engineering, Vancouver 2004,
CO
Canada, Paper 4006.
[5] Liu R, Palermo A. Low damage design and seismic iso-
lation: What’s the difference? New Zealand society for
earthquake engineering conference, 2015.
[6] Guo T, Cao Z, Xu Z, Lu S. Cyclic load tests on self-centering
concrete pier with external dissipators and enhanced dura-
bility. Journal of Structural Engineering. 2015. 10.1061/
(ASCE) ST.1943-541X.0001357, 1-15.
OR
[7] Sakai J, Mahin SA. Mitigation of residual displacements
of circular reinforced concrete bridge columns. 13th World
Conference on Earthquake Engineering. 2004, Canada; pp
1622.
[8] Trono W, Jen G, Panagiotou M, Matthew S, Claudia P.O.
Seismic Response of a damage-resistant reentering post
tensioned-HYFRC bridge column. Journal of Bridge Engi-
TH
neering 2014;20(7):1-13.
[9] Dawood H, Mohamed G, Hewes J. Behavior of segmen-
tal precast posttensioned bridge piers under lateral loads.
Journal of Bridge Engineering 2012;17(5):735-46
[10] Zatar W, Mutsuyoshi H. Reduced residual displacements
of partially prestressed concrete bridge piers. 12th World
AU
Conference on Earthquake Engineering, 2000, Lisboa.
[11] Espinoza A, Mahin S, Jeremic B, Kutter B, Ugalde J. Rock-
ing of bridge piers subjected to multi-directional earthquake
loading. Fifth National Seismic Conference on Bridges &
Highways 2006, New York, USA.
[12] Marriott D, Pampanin S, Palermo A. Quasi-static
and pesudo-dynamic testing of unbonded post-tensioned
rocking bridge piers with external replaceable dissipa-
ters. Earthquake Engineering and Structural Dynamics
2009;38:331-54.
[13] Marriott D, Pampanin S, Palermo A. Biaxial testing of
unbonded post-tensioned rocking bridge piers with external
replaceable dissipaters. Earthquake Engineering and Struc-
tural Dynamics. 2011;40:1723-41.
[14] Gabriele G, Jose IR, Milena M, Vervelidis A. Seismic
Behavior of posttensioned self-centering precast concrete
dual shell columns.Journal of Structural Engineering,
2012;04(141):1-11.
[15] Agalianos A, Psychari A, Vassiliou MF, Stojadinovic
B, Anastasopoulos, I. Comparative assessment of two
rocking isolation techniques for a motorway overpass
bridge. Frontiers in Built Environment. 2017. DOI:10.3389/
fbuilt.2017.00047.
101
[16] Åkesson B. Understanding Bridge Collapses. Taylor & Fran-
cis, London, UK. 2008.
[17] Kappos AJ, Saiidi MS, Aydinoglu MN, Isakovic T.
Seismic design and assessment of bridges, Springer Sci-
ence+Business Media, Dordrecht, Netherlands. 2012.
[18] Cardone D, Perrone G, Sofia S. Seismic response of simply
supported deck bridges with auxiliary super elastic devices.
Journal of Procedia Engineering. 2011;14:2315-22.
[19] Freyssinet Flat Jacks Brochure, France, 2016.
[20] Computers and Structures, Inc. CSI analysis and reference
manual for CSI BRIDGE, Computers and Structures, Inc.
Berkeley 2015, CA.
[21] Makris N, Vassiliou MF. The Dynamics of the rocking
frame. Seismic Assessment, behavior and Retrofit of Her-
itage Building and Mouments, Computational Methods
PY
in Applied Sciences. 37. DOI 10.1007/978-3-319-16130-
3 2,2015.
[22] Mitoulis SA, Rodriguez JR. Seismic Performance of Novel
Resilient Hinges for Columns and Application on Irregu-
lar Bridges. ASCE Journal of Bridge Engineering. 2017;
22(2).
[23] Rodriguez JR, Mitoulis SA. Novel connection for acceler-
ated bridge construction with dissipation and recentering
capabilities. 1st International Conference on Natural Haz-
ards & Infrastructure, 28-30 June, 2016, Greece.
[24] Mitoulis S. Novel connection for accelerated bridge con-
struction with dissipation and recentering capabilities, paper
ID: 28, 1st International Conference on Resilience, 22-23
September 2016, Torino, Italy.
[25] Raina VK, Concrete Bridge Practice-Analysis, Design and
Economics, Tata Mc Graw Hill Publication. 1999.
[26] BSI (British Standards Institution). Steel for the reinforce-
ment of concrete—weld able reinforcing steel—bar, coil
and decoiled product—specification. BS 4449:2005, Lon-
don.
[27] Priestley M, Seible F, Calvi G. Seismic design and retrofit
of bridges, 1996, Wiley, New York.
[28] Caltrans Bridge design specifications manual. California
Department of Transportation, 1990.
[29] Mander J, Priestley M, Park R. Theoretical stress-strain
model for confined concrete. Journal of Structural Engineer-
ing, 1988. 10.1061/(ASCE)0733-9445(1988)114:8(1804),
1804-1826.
[30] Ma TQ, Butterworth JW. Simplified Expressions For Mod-
elling Rigid Rocking Structures On Two-Spring Founation.
Bulletin of New Zealand society for earthquake engineering.
2012;45(1):31-39
[31] Yim SCS, Chopra AK. SimplifiedEarthquake Analy-
sis of Multistory Structures with Foundation Uplift.
Journal of Structural Engineering New York, N.Y.,
1985;111(12):2708-31.
[32] Anderson DL. Effect of foundation rocking on the seismic
response of shear walls. Canadian Journal of Civil Engi-
neering. 2003;30(2):360-5.
[33] FIP INDUSTRIALE, Elastomeric Bearing manufacturer,
Brochure B-03, 2005, Italy.
[34] Wen YK. Method for Random Vibration of Hysteretic Sys-
tems. Journal of the Engineering Mechanics Division. 1976,
ASCE, Vol.102.N0.EM2.
[35] Nagarajaiah S, Reinhorn AM, Constantinou MC. 3D-
Basis:Nonlinear Dynamic Analysis of Three-Dimensiona;
Base Isolated Structures:Part IITechincal Repeort
NCEER91-0005, National Center for Earthquake Engi-
neering Research 1991, State University of New York at
Buffalo, Buffalo, N.Y.
102 R.R. Rele et al. / Rocking isolation of bridge pier using shape memory alloy
[36] BSI (British Standards Institution). Structural Bearings. Part
3—Elastomeric Bearings. 2005, EN 1337-3, London.
[37] Naeim F, Kelly JM. Design of seismic isolated structures:
From Theory to Practice, 1999, Wiley, and New York.
[38] Kelly JM, Konstantinidis DA. Mechanics of rubber bearing
for seismic and vibration isolation, 2011, Wiley, New York.
[39] Desroches R, Delemont M. Seismic retrofit of simply sup-
ported bridges using shape memory alloys. Engineering
Structures. 2002;24:325-32.
[40] Dowell RK, Seible FS, Wilson EL. Pivot hysteretic model
for reinforced concrete members, ACI Structural Journal.
Vol 95, pp. 607-617.
[41] BSI (British Standards Institution). Eurocode 8: design of
structures for earthquake resistance–Part 2: bridges. BS EN
1998-2 (2005), London.
[42] Kalkan E, Chopra AK. Practical Guidelines to Select and
Scale Earthquake Records for Nonlinear Response History
CO
OR
TH
AU
Analysis of Structures, United States Geological Survey,
2010.
[43] SeismoMatch (Computer Software), Seismosoft, Pavia,
Italy.
[44] Rodgers G, Mander J, Chase J, Dhakal R. Beyond ductil-
ity: Parametric testing of a jointed rocking beam-column
connection designed for damage avoidance. Journal of
Structural Engineering. 2015. 10.1061/(ASCE)ST.1943-
541X.0001318, C4015006.
[45] Cheng C. Energy dissipation in rocking bridge piers under
free vibration tests, Earthquake Engineering and Structural
Dynamics. 2007;36(4):503-18.
[46] Priestley MJN, Calvi GM, Kowalsky MJ. Direct displace-
ment based design of structures 2007, IUSS Press, Pavia,
Italy.
PY
 R.R. Rele et al. / Rocking isolation of bridge pier using shape memory alloy 103

Fig. A1. Fix base bending moment validation.

Appendix A 29902.6 kN.m at bottom of pier and 30410.8 kN.m a

PY
top of pier.
A.1. Validation of bending moments for fix The small difference of bending moment (6–8%) is
base due to the non-linear time history analysis performed
in the computer programme as against simple check

CO
Consider idealized diagram as shown in Fig. A.1 given for bending moment in pier following the linear
for a fix base with integral (monolithic) superstruc- laws of structural mechanics for a pier with both ends
ture. being fixed.
The axial force (average values of seven-time his- With this as the basis of explanation, the rocking
tories) for fix model from Table 2 is 9412.9 kN. pier should have the bending moment less than the
The horizontal force due to seismic = P=Axial load fixed base as the pier is allowed to uplift and thus
x PGA. bending moments are significantly dropped as seen
OR
The bending moment at base/ top of pier = P. H / in Tables 3 and 4.
2 = 9412.9 x 0.6 x 10/2 = 28238.7 kN.m.
This value is compared to the average of the values
from seven-time history analysis in Table 2 which is
TH
AU

View publication stats