2/11/2013

1

B.M. Pacheco, J.Y. Hernandez Jr., E.A.J. Tingatinga, P.P.M. Castro,

F.J. Germar, U.P. Ignacio, M.C.L. Pascua, L.R.E. Tan,

I.B.O. Villalba, D.H.M. Aquino, R.E.U. Longalong,

R.N. Macuha, W.L. Mata, R.M. Suiza, M.A.H. Zarco

PALIWANAGAN SA UP DILIMAN OVCRD COLLOQUIUM 2013

January 21, 2013 National Institute of Physics

University of the Philippines Diliman 2

July 16, 1990 Philippine Earthquake (by Caranto)

March 11, 2011 Japan Earthquake (abc News international)

February 22, 2011 New Zealand Earthquake (San Antonio Express News)

3

Typhoon “UNDANG” (Agnes) Nov 3-6,1984 230 kph 895 deaths

Php 1.9 B damage

Typhoon “REMING” (Durian) Nov 28-Dec 1, 2006 281 kph (Gust 320 kph)

709 deaths 2,190 injured 753 missing Php 10.89 B damage

Typhoon “MILENYO” (Xangsane) Sep 25-29,2006 150 kph (Gust 160 kph)

45 deaths Php 1.251 B damage

4

2/11/2013

2

5

Will you, as private citizens, take steps to increase the capacity of your houses or buildings if you knew beforehand that they are at risk to a particular hazard? Will you reduce (if not minimize) the risk?

Are our government officials willing to invest resources in improving the condition of existing government infrastructure (buildings, bridges, elevated roads and railways, airports, etc.) if they knew beforehand that these are vulnerable to damage?

6

VULNERABILITY – the degree of loss to a given element at risk resulting from a given level of hazard (e.g. ground motion intensity, wind speed, depth of inundation). It is defined as the ratio of the cost of damage to the cost of structure and finishes, on a scale of 0 to 1 (or 0 to 100%)

VULNERABILITY CURVE – a plot of the vulnerability of a structure vs. the level of hazard.

7

Da

ma

ge

Ra

tio

Modified Mercalli Intensity (MMI) or Peak Ground Acceleration (PGA) 8

Da

ma

ge

Ra

tio

Wind Speed (kph)

2/11/2013

3

9

Da

ma

ge

Ra

tio

Flood Depth (m)

10

Wood Light-frame (W1) Bamboo Hut (W3)

Makeshift (N)

11

Concrete Hollow Block (CHB)

Unreinforced Adobe (URA) Walls

Unreinforced Masonry (URM) Walls

Concrete Hollow Blocks with Wood or Light Metal (MWS)

12

Reinforced Concrete Frames with Wood or Light Metal (CWS)

Reinforced Concrete Moment Frame (C1)

Concrete Shear Walls and Frames (C4)

2/11/2013

4

14

Steel Moment Frames (S1) Steel Frames with Cast-in-place Concrete Shear Walls (S4)

16

17

COMPUTATIONAL

EMPIRICAL

VULNERABILITY CURVE

HEURISTIC

18

2/11/2013

5

19 19

NSP Analysis

PERFORMANCE POINTS

PROBABILITY OF EXCEEDANCE

CURVE FITTING

FRAGILITY CURVES

BUILDING DATABASE

DAMAGE STATE EVALUATION

THRESHOLD VALUES

Capacity Spectrum Method

DEMAND SPECTRUM

Computational Method

for Earthquake

Vulnerability

Curves

VULNERABILITY CURVES

DIFFERENT ATTRIBUTES

20

21

18 m

Material Properties:

Unit weight = 23.5 kN/m3

Concrete compressive strength, fc’ = 21 MPa

Reinforcement yield stress, fy = 240 MPa

Modulus of elasticity, E = 25 GPa

Poisson’s Ratio,

ν = 0.20

22

(Source: ATC-40)

• Provides an estimate of the Capacity of the Building

Base Shear

Roof Displ.

Plastic Hinges

Where yielding

Occurs (shear

and bending moment)

2/11/2013

6

23

Sp

ec

tra

l A

cc

ele

rati

on

(g

)

Period 24

PERFORMANCE

POINT

25

Spectral Displacement

Sp

ectr

al A

ccel

erat

ion

(g

)

Sd-Moderate = 0.035m

Sd-Extensive = 0.073m

26

PGA Number of Models Exceeding a Damage State Probability of Exceeding a Damage State

Slight Moderate Extensive Complete Slight Moderate Extensive Complete

0.048 0 0 0 0 0 0 0 0

0.096 4 0 0 0 0.267 0 0 0

0.128 4 3 0 0 0.267 0.2 0 0

0.16 4 4 0 0 0.267 0.267 0 0

0.24 4 4 0 0 0.267 0.267 0 0

0.32 6 6 5 2 0.4 0.4 0.333 0.133

0.4 10 10 10 6 0.667 0.667 0.667 0.4

0.48 15 15 15 13 1 1 1 0.867

0.64 15 15 15 15 1 1 1 1

2/11/2013

7

27 28

Damage States Damage Ranges (%) Damage Indices (%)

None 0-2 1

Slight 2-10 6

Moderate 10-50 30

Extensive 50-100 75

Complete 100 100

Damage index at different damage states

Using the fitted fragility curves, the vulnerability curves can be derived

Where,

Pv,GMI = damage probability at a certain GMI.

PF,I = Probability of a structure being in a certain damage state i (from Fragility Curves).

Di = Damage index for a certain damage state i from HAZUS.

iFiGMIv PDP ,, *

29

Da

ma

ge

Ra

tio

Modified Mercalli Intensity (MMI) 30

Curve-fitting of Lognormal Cumulative Probability Distribution

Computational Fluid Dynamics (CFD) Analysis

Pressure Distribution Due to Wind

Damage State Evaluation

Probability of Exceedance

Building Database

Computational Fragility Curves

HAZUS MH Damage States

Repeated for all building models in the database until 350 kph

Vulnerability Curve

Suction Pressure Threshold Values from

Experiments (Resistance)

Varying attributes

Varying wind speeds And 3 directions (0, 45, & 90 deg)

30

2/11/2013

8

31

0o wind direction - perspective view 0o wind direction - side view

90o wind direction - perspective view 90o wind direction - side view 32

Perspective View

Windward side

Roof

Leeward side

33 33

0

10

20

30

40

50

60

70

80

90

100

0 50 100 150 200 250 300 350

Dam

age

Rat

io (

%)

Wind Speed, kph

Datapoints

Fit

34

2/11/2013

9

35 36

37 38

Depth 0 m 0.5 m 1 m 2 m 3m 4 m 6 m 10 m

Damage

Ratio

(%)

------ ------ ------ ------ ------ ------ ------ ------

Depth 0 m 0.5 m 1 m 2 m 3m 4 m 6 m 10 m

Damage

Ratio

(%)

------ ------ ------ ------ ------ ------ ------ ------

2/11/2013

10

39 40

41 42

1. Collection of building damage observations due to recent/historical earthquakes from published papers, field reports, etc.

(Obtained the report from PHIVOLCS and GA last November 12, 2012)

2. Classification and extraction of useful data for the building types. For CHB or MWS, key phrases like “concrete hollow blocks”,

“residential house” (without description if it is reinforced concrete), and “masonry” were used as identifiers.

2/11/2013

11

43

For URA or URM, key phrases like “church” were used.

Validation of building type using available photos and descriptions (in the internet).

3. Assignment, based on the descriptions of building damage for several earthquakes, of damage ratio for each recorded earthquake intensity.

4. Development of the vulnerability curve for each type

44

45

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

4 5 6 7 8 9 10

Da

ma

ge

Ra

tio

MMI

CHB/MWS Vulnerability Curve

Data CHB_VC

Median 6.88

Beta 0.21 46

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

4 5 6 7 8 9 10

Da

ma

ge

Ra

tio

MMI

URM Vulnerability Curve

Data URM_VC

Median 7.12

Beta 0.17

2/11/2013

12

47 47

Typhoon Simulation by PAGASA

Building Type Identification

Building Damage Quantification

Post-storm data

Vulnerability Curve

Curve-fitting of Lognormal Cumulative Probability Distribution

48

Typhoon Pedring Maximum gust = 61mps = 219.6kph 55% Damage

Typhoon Melenyo Maximum gust = 65 mps= 234kph 65% Damage

49

Wind Vulnerability Curves for Buildings

0

10

20

30

40

50

60

70

80

90

100

0 50 100 150 200 250 300 350

Wind Speed (kph)

% D

amag

e W3

W1 & W2

CHB & CWS

C1 & C2

Dam

ag

e R

ati

o (

%)

50

Vulnerability curves were developed for each building type using the 3 methods (computational, heuristic, and empirical) for earthquake, severe wind, and flood hazards.

The vulnerability curves can now be used to estimate the risk to our existing building stock.

2/11/2013

13

51

The authors would like to acknowledge and thank PAGASA, PHIVOLCS, CSCAND, Geosciences Australia (GA), and Australian Aid (AusAid) for their collaboration and financial support of this project.

52