experimental and numerical study on seismic response ... cost base isolator_s.k.deb.pdf ·...

Prof. S.K. Deb

Experimental and Numerical Study on Seismic Response Control of Unreinforced Masonry

Test Model using U-FREI

Department of Civil EngineeringIndian Institute of Technology Guwahati

IITG-IITK Workshop: 25-06-2016

Other Members of the Research Group:

Prof. Anjan Dutta, IITG Mr. Ngo Van Thuyet, Research

Scholar (R/S) Dr. Animesh Das, BHEL (former R/S) Mr. A. Hazarika, former PG student Mr. K. Reddy, former UG student

Industry Partner: M/S METCO Pvt. Ltd., Kolkata

Outline of the presentation

Basic Concepts of Base Isolation

Design and detailing issues

Brief Literature Review

FE Simulation of Isolator

Comparison of Experimental and FE Results

Shake Table Testing of unreinforced masonry test model

Prototype Implementation

Advantages of Base Isolation System

Reduced floor acceleration and inter-story drift

Less (or no) damage to structural members

Better protection of secondary structural systems

Prediction of response is more reliable

Philosophy of base isolation

Lengthening of time period

Enhancement of damping

Fixed Base

Base IsolatedPeriod

Significantly Increase the Period of the Structure and the Damping so that the Response is Significantly Reduced

The Concept of Base Isolation

Idealized force-displacement hysteretic behavior of isolation system

Laminated Rubber bearing: Design Concept

Conventional Laminated Rubber Seismic Isolation Bearings

Estimation of Displacement [ASCE / SEI: 7-05]

Construction Details

Literature Review

Experimental and Analytical Study on FREI

Kelly(2001), Tsai (2005), Moon(2008), Nezhad et al. (2008)

FE Analysis

Mordini (2008), Nezhad (2011), Kelly(2012), Osgooei (2014),

Spizzuoco (2014), Das et al. (2014)

Shake Table Testing of Base Isolated building

Nezhad (2009), Das et al. (2016)

The main objectives of this study are:

(i) To carry out numerical simulation of FREI

(ii) To develop lightweight, low cost FREI

(iv) To carry out experimental study on U-FREI

(v) To carry out shake table testing of base isolated TestModel

(vi) Prototype implementation

The simple single degree-of-freedom natural frequency(fn) of the base isolated model structure is given by

𝑓𝑓𝑛𝑛 =1

2𝜋𝜋𝐾𝐾𝐻𝐻𝑚𝑚𝑡𝑡

Horizontal stiffness of bearing

𝐾𝐾ℎ = 𝐾𝐾𝐻𝐻Number of bearings

= 60.217 ⁄kN m

Designs of Seismic Isolator

For (SREI), the horizontal stiffness is given by simple shear

formula

𝐾𝐾ℎ =𝐺𝐺𝐺𝐺𝑡𝑡𝑟𝑟

From above equation, 𝑡𝑡𝑟𝑟=95 mm. To achieve high stiffness

of the bearings, 19 layers of elastomer, each of 5 mm thick,

is selected. The elastomer layers are separated by 0.55 mm

thick fiber reinforcement.

𝐾𝐾𝑣𝑣 =𝐸𝐸𝐶𝐶𝐺𝐺𝑡𝑡𝑟𝑟

The vertical stiffness is given by,

The compression modulus EC modulus for square isolator isgiven by:

𝐸𝐸𝑐𝑐 = 96𝐺𝐺𝑆𝑆2

𝜋𝜋2 𝛼𝛼𝛼𝛼 2 ∑𝑛𝑛=1∞ 2𝑛𝑛−1/2 2

𝑡𝑡𝛼𝛼𝑛𝑛ℎ𝛾𝛾𝑛𝑛𝛼𝛼𝛾𝛾𝑛𝑛𝛼𝛼

− 𝑡𝑡𝛼𝛼𝑛𝑛ℎ𝛽𝛽𝑛𝑛𝛼𝛼𝛽𝛽𝑛𝑛𝛼𝛼

𝐸𝐸𝑐𝑐= 101212.6 kN/m2. 𝐾𝐾𝑣𝑣 = 10653.96 kN/m. Therefore,ratio of the compressive stiffness to shear stiffness is equal173.

Cross section of isolator

Geometrical properties of square and circular isolator

Description Square Circular

Width (2a) or radius (R) of isolator = 100 mm 112 mmThickness of fiber layer (tf) = 0.55 mm 0.55 mmNumber of fiber layer = 18 18Thickness of single rubber layer (te) = 5 mm 5 mmNumber of rubber layer = 19 19Total height of isolator (h) = 104.9 mm 104.9 mm

Hardness IRHD = 60Shear modulus of elastomer (G) = 0.7 MPaElongation at break > 400%Young’s modulus of fiber reinforcement (Ef) = 4400 MPaPoisson’s ratio of rubber (ϑf) = 0.20

Material properties of elastomer

Buckling load of Square Isolator

sl No. Investigator Buckling Load (kN)1 Gent 43.992 Koh and Kelly 42.773 Kelly 40.50

FE Simulation of Isolator Behaviour

Finite Element Modeling

Element Type for Finite Element Model

Element type of fiber reinforcement

SOLID46, 8-Node Layered Structural Solid is used to model thereinforcement.

Element type of ElastomerSOLID185, 8-Node Structural Solid element isused to model the elastomer

Contact, Target Elements

This modeling is done using 3-D surface-to-surface contact elements CONTA173 andTARGE170.

Material Model

Ogden 3-terms model

Loading History

A vertical load of 12.0kN ±3.0kN

02468

10121416

0 10 20 30 40 50 60 70 80 90 100

Ver

tical

Loa

d (k

N)

Time (s)

-70

-50

-30

-10

10

30

50

70

0 3 6 9 12 15 18

Horiz

ontal

Disp

lacem

ent (

mm

)

Time (s)

Three cycles ofspecific displacementup to 60 mm isapplied on the top ofisolator with constantvertical load

FE Analysis of Square Isolator

X axis aligned with fibersoriented along 0°, while Yaxis aligned with fibers along90°. The orientation ofhorizontal loading 00 and 450

are along X-axis and 450 toX-axis respectively

The top and bottom surfaces can roll off the support surfacesand no tension stresses are produced in un-bonded FREI. Tensilestresses are not transferred to the contact surfaces of un-bondedapplication

Free body diagram in laterally deformed FREI with different boundary condition

Stress and Strain of Square Isolator

Square isolator with 00 loading direction

Distribution of normalized stress S33/Pn.

(a) 60mm horizontal displacement

(b) 40mm horizontal displacement (c) 20mm horizontal displacementPeak stress is 44% higher for bonded.

-6

-4

-2

0

2

0 0.2 0.4 0.6 0.8 1No

rmal

ized

stre

ss S

33/P

nNormalized width of isolator

Square Unbonded (0)Square Bonded (0)

-6

-4

-2

0

2

0 0.2 0.4 0.6 0.8 1

Nor

mal

ized

stre

ss S

33/P

n

Normalized width of isolator


-6

-4

-2

0

2

0 0.2 0.4 0.6 0.8 1

Nor

mal

ized

stre

ss S

33/P

n



Contour of normal stress S33 (kN/m2) in mid rubber layer of theisolator at horizontal displacement 60mm (00 loading direction &positive value indicate tension)

(a) Un-bonded isolator at 60mm displacement

(b) Bonded isolator at 60mm displacement

Contour of shear strain in therubber layer of isolator athorizontal displacement 60mm(00 loading direction) and strainalong mid height of elastomer

(a) Un-bonded

(b) Bonded

0.00

0.10

0.20

0.30

0.40

0.50

0 0.2 0.4 0.6 0.8 1

Shea

r str

ain


Square unbondedSquare Bonded

0.00

0.10

0.20

0.30

0.40

0.50

0 0.2 0.4 0.6 0.8 1

Shes

r str

ain



0.00

0.10

0.20

0.30

0.40

0.50

0 0.2 0.4 0.6 0.8 1

Shea

r str

ain



20mm horizontal displacement



Hysteresis of square isolator with 00 loading direction

-3

-2

-1

0

1

2

3

-80 -60 -40 -20 0 20 40 60 80

Shea

r For

ce (k

N)

Horizontal Displacement (mm)

-5

-3.75

-2.5

-1.25

0

1.25

2.5

3.75

5

-80 -60 -40 -20 0 20 40 60 80

Shea

r For

ce (k

N)


Shear force vs horizontal displacement for un-bonded isolator at 0° loading

Un-bonded isolator at 60mm displacement (0°Loading)

Shear force vs horizontal displacement for bonded isolator at 0°

Bonded isolator at 60mm displacement (0°Loading)

un-bonded bonded Displac-ement(mm)

Effective Horizontal

Stiffness (𝑲𝑲𝒆𝒆𝒆𝒆𝒆𝒆𝒉𝒉 )

(kN/m)

Damping (β) (%)

Effective Horizontal

Stiffness (𝑲𝑲𝒆𝒆𝒆𝒆𝒆𝒆𝒉𝒉 )

(kN/m)

Damping (β) (%)

10 88.3 12.3 89.7 12.120 76.5 12.8 86.4 12.330 66.2 13.1 83.2 12.340 56.9 13.9 79.3 12.550 46.2 15.2 74.7 12.860 39.3 16.1 70.7 13.1

Computation Effective Stiffness and Damping

Kheff = (Fmax – Fmin)/(dmax – dmin) and β = Wd/(4πWs)

Hysteresis of square isolator with 450 loading direction

Shear force vs horizontal displacement for un-bonded isolator at 45° loading

Un-bonded isolator at 60mm displacement (45°Loading)

Shear force vs horizontal displacement for bonded isolator at 45° loading

Bonded isolator at 60mm displacement (45°Loading)

-4

-3

-2

-1

0

1

2

3

4

-80 -60 -40 -20 0 20 40 60 80

Shea

r For

ce (k

N)


-5

-3.75

-2.5

-1.25

0

1.25

2.5

3.75

5

-80 -60 -40 -20 0 20 40 60 80

Shea

r For

ce (k

N)


Lateral load vs displacement of the bonded & un-bonded square isolator

-5000

-3750

-2500

-1250

0

1250

2500

3750

5000

-60 -40 -20 0 20 40 60

Shea

r For

ce (N

)

Horizontal displacement (mm)

Unbonded FREIBonded FREI

Analytical (𝐾𝐾ℎ = 𝐺𝐺𝐺𝐺𝑡𝑡𝑟𝑟

) From Eq. (3.1) FE Analysis

73.68 73.65 70.7

Horizontal stiffness (N/mm) of bonded square FREI

Horizontal Stiffness of un-bonded is 70% less than bonded

Effect of Vertical Load on Shear Capacity

-3

-2

-1

0

1

2

3

-80 -60 -40 -20 0 20 40 60 80

Shea

r For

ce (k

N)


Analysis (100% Vertical load)Analysis (75% Vertical load)

100% and 75% of total vertical load

-3

-2

-1

0

1

2

3

-80 -60 -40 -20 0 20 40 60 80

Shea

r For

ce (k

N)


Analysis (100% Vertical load)Analysis (125% Vertical load)

100% and 125% of total vertical load

Vertical Load (%)

Maximum shear force (kN)

Minimum shear force (kN)

75 2.37 -2.37100 2.42 -2.41125 2.52 -2.50

Lateral Load Testing Arrangement and Instrumentation

Total weight arrangement consists of a two storied frame structureplaced on a steel plate, concrete slabs and beams.

Front view of the experimental set up for lateral loading test

Top cross sectional view of the experimental set up

Comparison of Numerical and Experimental Results

Result of Loading along 00 Orientation

-3

-2

-1

0

1

2

3

-12 -9 -6 -3 0 3 6 9 12

Shea

r Fo

rce

(kN

)


Test ResultAnalysis Result

-3

-2

-1

0

1

2

3

-25 -20 -15 -10 -5 0 5 10 15 20 25

Shea

r Fo

rce

(kN

)Horizontal Displacement (mm)


10mm displacement 20mm displacement

-3

-2

-1

0

1

2

3

-35 -28 -21 -14 -7 0 7 14 21 28 35

Shea

r For

ce (k

N)



-3

-2

-1

0

1

2

3

-45 -36 -27 -18 -9 0 9 18 27 36 45

Shea

r For

ce (k

N)



-3

-2

-1

0

1

2

3

-60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60

Shea

r For

ce (k

N)



-3

-2

-1

0

1

2

3

-70 -56 -42 -28 -14 0 14 28 42 56 70Sh

ear F

orce

(kN

)





Displaced Shape of Isolator (00 loading direction)

at 10mm maximum displacement






Concluding Remarks on Isolator Testing

Horizontal stiffness as obtained from numerical solutionshows close agreement with those obtained from experiment.

Horizontal stiffness of un-bonded isolator corresponding to450 loading is observed to be slightly higher than that for 00

loading.

Experimentally observed displaced shapes of the isolatorsare matching with analytically obtained shapes.

Shake Table Testing of Test ModelThe experiment carried out on a 1/5th scaled two storey unreinforcedmasonry building supported on four square un-bonded fiber reinforcedelastomeric isolator (U-FREI)

A scale factor of 1:5 is considered shake table size and its payloadcapacity

The laws of similitude are as follows

Parameters Scale Prototype 1/5-scale model

Length S 5 Mass S2 25 Displacement S 5 Time 𝐒𝐒 2.236 Acceleration S 5

Model building and its details

Parameter Prototype Building

Model Building

Length (m) 7.5 1.5Width (m) 5.5 1.1Height of storey (m) 3 0.6Thickness of slab (m) 0.15 0.08Base beam (m x m) 0.625x0.7

500.125x0.15

0Total weight of building (kg)

12140 971

Sample Ground Motions for Shake Table Test(1) Koyna (1967): Comp - Longitudinal, (2) Parkfield (1966): Comp - C02065, (3) El Centro (1940): Comp - 180, (4) Victoria (1980): Comp - CPE045 transverse earthquakes.

Earthquake Components Peak Ground Acceleration (g)

Frequency Range (rad/sec)

Koyna (1967): Comp - Longitudinal 0.63 0-12Parkfield (1966): Comp - C02065 0.48 0-40El Centro (1940): Comp - 180 0.32 0-65Victoria (1980): Comp - CPE045 0.62 0-160

Table: Characteristics of selected earthquake records

(a) Koyna (1967): Comp - Longitudinal (b) Parkfield (1966): Comp - C02065

(c) El Centro (1940): Comp - 180 (d) Victoria (1980): Comp - CPE045

Time scaled representation of the four selected earthquakesacceleration histories are shown in following figures

-0.8

-0.6

-0.4

-0.2

0

0.2

0.4

0.6

0 1 2 3 4 5

Acc

eler

atio

n (g

)

Time (sec)-0.6

-0.4

-0.2

0

0.2

0.4

0.6

0 5 10 15 20

Acc

eler

atio

n (g

)

Time (sec)

-0.4

-0.2

0

0.2

0.4

0 3 6 9 12 15

Acc

eler

atio

n (g

)

Time (sec)-0.8

-0.6

-0.4

-0.2

0

0.2

0.4

0.6

0 2 4 6 8 10 12A

ccel

erat

ion

(g)

Time (sec)

Displaced shape of isolator during shake table test for Parkfield input earthquake

(a) Parkfield (for 100% acceleration amplitude of earthquakes along X-axis).

(b) Parkfield (for 70% acceleration amplitude of earthquakes along 450 to X-axis.)

Peak Responses of Base Isolated Model

Fig: Acceleration response at shake table level and base level subjected to four earthquakes (full intensity) applied along X-axis

(a) Koyna (b) Parkfield

(c) El Centro (d) Victoria

-0.8

-0.4

0

0.4

0.8

0 1 2 3 4 5

Acc

eler

atio

n (g

)

Time (sec)

Shake TableBase of bldg

-0.8

-0.4

0

0.4

0.8

0 5 10 15 20

Acc

eler

atio

n (g

)

Time (sec)


-0.4

-0.2

0

0.2

0.4

0 3 6 9 12 15

Acc

eler

atio

n (g

)

Time (sec)


-0.8

-0.4

0

0.4

0.8

0 3 6 9 12

Acc

eler

atio

n (g

)

Time (sec)



Fig: Comparison of acceleration responses at base level, first floor and roof level subjected to four earthquakes (full intensity) applied along X-axis



-0.12

-0.06

0

0.06

0.12

0 1 2 3 4 5

Acc

eler

atio

n (g

)

Time (sec)

Base levelFirst floorRoof level

-0.3

-0.15

0

0.15

0.3

0 5 10 15 20

Acc

eler

atio

n (g

)

Time (sec)


-0.2

-0.1

0

0.1

0.2

0 3 6 9 12 15

Acc

eler

atio

n (g

)

Time (sec)


-0.2

-0.1

0

0.1

0.2

0 3 6 9 12A

ccel

erat

ion

(g)

Time (sec)



Fig: Displacement at base level and first floor level subjected to four earthquakes (full intensity) applied along X-axis



-10

-5

0

5

10

0 1 2 3 4 5

Disp

lacem

ent (

mm

)

Time (sec)

Base levelFirst floor

-60

-30

0

30

60

0 5 10 15 20

Disp

lace

men

t (m

m)

Time (sec)


-22.5

-15

-7.5

0

7.5

15

22.5

0 3 6 9 12 15

Disp

lace

men

t (m

m)

Time (sec)


-30

-20

-10

0

10

20

30

0 3 6 9 12

Dis

plac

emen

t (m

m)

Time (sec)


Peak acceleration and displacement at different levels of model subjected to four earthquakes (full intensity) along X-axis

EarthquakePeak Accelerations (g) Peak Displacement

(mm) At Shake

Table At Base At First Floor

At Roof Level

At Base level

At First Floor

Koyna 0.632 0.0873 0.0700 0.0867 6.326 7.240

Parkfield 0.476 0.2145 0.2081 0.2463 36.199 39.920

El Centro 0.319 0.1524 0.1601 0.1686 17.789 19.409

Victoria 0.615 0.1230 0.1310 0.1459 19.452 21.251

The following conclusions are drawn from the scaled model study:

U-FREIs are observed to be very effective in reducing seismic responses of model structure.

The effectiveness in seismic isolation increases with increased displacement, where U-FREI maintains a stable rollover configuration within the estimated displacement limit.

The U-FREIs are observed to be effective irrespective of loading directions.

Inertia forces, shear forces and bending moment of the test model supported on U-FREI are substantially lesser than the fixed base structure.

The designed U-FREIs used in this study are applicable to the scaled model building only. The isolators for this model building are slender because of lesser vertical load. This constraint would not be encountered in the design of prototype U-FREIs, and hence higher aspect ratio for prototype bearings is achievable.

Experimental study demonstrated potential of U-FREI in reducing the seismic vulnerability of un-reinforced masonry buildings . Introduction of U-FREI at the interface of superstructure and substructure of an un-reinforced masonry building would be simple and hassle free.

Base Isolated Masonry Building at TawangArunachal Pradesh (under-construction)

Isolation System: Un-bonded Fiber Reinforced Isolator (U-FREI) R&D and design: IIT Guwahati and Manufacturee: M/s METCO Pvt. Ltd. Kolkata

Testing of prototype FREIs

Validation of Numerical Model

IITG CAMPUS: SPRING 2016

Thank you for your kind attention

experimental and numerical study on seismic response ... cost base isolator_s.k.deb.pdf ·...

Documents