Experimental and Numerical Study to Improve Damage and Loss - - PowerPoint PPT Presentation

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

CR CRC C 2nd

nd An

Annual Mee eeting: g: Feb ebruary 1 1 – 3, 3, 2017, 2017, C Chapel H Hill, NC NC

Experimental and Numerical Study to Improve Damage and Loss Estimation due to Overland Wave and Surge Hazards on Near-Coast Structures

John van de Lindt

Colorado State University

Daniel Cox

Oregon State University

Kevin Cueto, Diego Delgado, Trung Do, Ben Hunter, Tori Johnson, Hyoungsu Park, Willliam Short

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Pr Project Overview

Objective 1: Quantify surge/wave forces on near-coast structures and develop new predictive equations. Objective 2: Develop the conditional probabilities (fragilities) for exceeding key thresholds. Objective 3: Illustrate next-generation risk- informed design.

Introduction

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Exi xisting FEMA Damage Predictions

Introduction

Case 1:

• Freeboard=-2.7 m
• Damage: Interior water damage, minimal

structural damage

• Significant Wave Height: 0.99m

Case 2:

• Freeboard=-1.5 m
• Damage: 100%
• Significant Wave Height: 1.37 m

Predicted Damage: 72% Predicted Damage 60%

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Te Technical Approach

Task 1: Hydraulic model test program at OSU (Yr 1) and data analysis (Yr 1, 2). Task 2: Numerical model program at CSU. Verification (Yr 1) and fragility development (Yr 1, 2). Task 3: Develop performance based design examples to illustrate methodology for engineering practice (Yr 2).

Introduction

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Task 1: Hydraulic model test program

Research Accomplishments Hurricane Ike 2008 / Bolivar Peninsula, TX

Observed surge and wave action:

Offshore significant wave height: 5.8 m Storm surge: 4.3 m Inundation: 4.2-4.7 m Estimated overland waves: 1.9 m

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Geometric scale 1:10

Wave height: 0.10 < H < 0.50 m Inundation: 0.40 m Specimen dimensions: 1.02 x 1.02 x 0.61 m

Froude similitude 1:3.16

Wave period: 2.5 < T < 5.0 s

Simplifying Assumptions

• No substructure
• No sediments, scour
• No debris
• No currents

First comprehensive measurements

• f wave forces on elevated

residential structures

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Research Accomplishments

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Research Accomplishments

Gulf of Mexico Barrier island Bay

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Research Accomplishments

Gulf of Mexico Barrier island Bay

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Research Accomplishments

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Research Accomplishments

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Research Accomplishments

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Research Accomplishments

Exp. TMA REG TRAN HS (m) TP (s) h (m) Dur. (min) 𝐼 " (m) 𝑈 $ (s) h (m) Dur. (min) A (m) TR (s) h (m) ts (s) X1 0.10 3.72 2.15 40.0 0.10 4.10 2.15 4.00 0.51 36.4 2.00 10.0 X2 0.19 3.86 2.15 40.0 0.21 4.10 2.15 4.00 0.34 51.0 2.00 15.0 X3 0.29 4.10 2.15 40.0 0.29 4.10 2.15 4.00 0.28 87.2 2.00 20.0 X4 0.40 4.10 2.15 40.0 0.40 4.10 2.15 4.00 0.21 109 2.00 25.0 X5 0.50 3.86 2.15 40.0 0.50 4.10 2.15 4.00 0.18 117 2.00 30.0 X6 0.16 2.52 2.15 25.0 0.16 2.52 2.15 2.50 0.16 120 2.00 35.0 X7 0.21 2.98 2.15 30.0 0.23 2.98 2.15 3.00 0.14 154 2.00 40.0 X8 0.25 3.28 2.15 35.0 0.26 3.64 2.15 3.50 0.13 162 2.00 45.0 X9 0.34 4.68 2.15 45.0 0.35 4.68 2.15 4.50 X10 0.39 5.04 2.15 50.0 0.42 5.04 2.15 5.00

• Data are hosted on NSF DesignSafe-CI

for public access

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Research Accomplishments

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Research Accomplishments

a (m)

Air Gap cases REG TMA TRAN

a0

• 0.40
• 0.25

a1

• 0.30
• 0.15

a2

• 0.20
• 0.05

a3

• 0.10

0.05 a4

• 0.05

0.10 a5 0.00 0.15 a6 0.05 0.20 a7 0.10 0.25 a8 0.20 0.35 a9 0.28 0.43

SWL a

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Research Accomplishments

15

Effects of Air Gap on Horizontal, Vertical Forces

• Horizontal force increases as air

gap decreases.

• The maximum vertical force is

found when the air gap is zero.

• In some cases vertical force can

exceed horizontal force

• There are limited provision in

FEMA 55 or ASCE 7-16 to estimate these forces Horizontal Force Vertical Force

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Contributed to October 2016 feature article on CRC page Participation in SUMREX program Kevin Cueto; Diego Delgado7-week summer program at HWRL

Thank you, Josh!

Education and Outreach

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Task 2: Numerical model program at CSU. Verification (Yr 1) and fragility development (Yr 1, 2).

• Model verification using existing data
• Tsunami loads on wood-frame wall at full scale test by

Linton et al. (2013)

• Uplift forces on a large scale bridge superstructure by

Bradner et al. (2011)

• Elevated Structure Impact test at OSU (2016)

Research Accomplishments

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Calibrating model using wood-frame wall at full scale test

Research Accomplishments

1.85 m 2.36 m 3.66 m 3.58 m 2.44 m Flume botom Flume wall 0.33 m

Sketch of transverse wood wall in flume Numerical model of wood wall

Load cells (Supports) Displacement (Location 3) 0.33 m 1.85 m 3.58 m 2.44 m 0.065 m Location 3

v1= v (z,t) Wave maker Wood wall 7.3m 0.5m 2.5m 25.4 m Tsunami wave 1:12 2.36 m 3.66 m 3.58 m 2.44 m Location 1 Location 2 3.6 m

c)

Location 3 2.36 m 7.0 m 3.66 m 35.7 m sea water material Empty material bottom face: v3=0 flow-out face, p=0 side face, v2=0 flow-in face, v1=v(z,t) top face p=0

Numerical of wave flume and wood wall

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Comparing numerical results and tested data

Research Accomplishments

Compare deep water wave height at location 1 (a), running up wave height(b) and velocity (c) In front of the wall (location 2) Compare flux (d), total horizontal force (e) at location 2, and deflection at mid span of the wall (location 3)

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Model verification with uplift loads

Research Accomplishments

Compare result for total uplift on bridge between model and test Analysis of failure mechanics: total uplift exceed the self-weight of bridge superstructure causing failure

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Research Accomplishments

Elevated Structure Impact test at OSU (2016)

Numerical model of flume and elevated structure in wave flume Locations of pressure gauges Locations of wave gauges

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Video showing wave impact on elevated structure

Research Accomplishments

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Comparing wave heights at wage gausses

Research Accomplishments

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Research Accomplishments

Comparing results at pressure gauges

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Task 2: Numerical model program at CSU. Verification (Yr 1) and fragility development (Yr 1, 2).

• Fragility development
• 3 Building archetypes are selected from 6 residential wood

building archetypes of the hurricane wind project

• Set up numerical model and collect total uplift and shear as

well as force on components such as doors, windows, and walls

• Establish damage states based on damage of components

such as door, windows, and nails connection of wood walls.

• Generate fragility surfaces based on both significant wave

heights and flood levels

Research Accomplishments

Archetype 1 23 x 50 ft (7x15m), Rectangle, 1-story Archetype 2 38 x 52 ft (12x16m), 2-story Archetype 3 38 x 65 ft (12x20m), 2-story

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Building models

Research Accomplishments

• The buildings are modeled in ANSYS Fluent
• Piles rising from 0, 1m, 2m, and 3m, from the

ground

Archetype 1 With 3-m elevated pile from ground

• TMA spectrum for hurricane waves with 𝐼" = 1,

2 , and 3 m, which can cover up to 12m significant wave height in deep water, and wave peak period, 𝑈

$, from 8s to 14s

• Surge (SWL, food) levels ℎ = 1, 2, and 3m

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Example of wave impact on building archetype 1

Research Accomplishments

• Piles rising 1m

from the ground

• Significant wave

height = 1m

• Flood level = 1m

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Pressure measured location and distribution at one wave impact event

Research Accomplishments

PG1 PG2 PG3 PG4 PG5 PG6 PG7 PG3 PG5 PG7

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Table of combinations for each building archetype

Research Accomplishments

Pile Elevation (m) Surge level (Hs, m) Significant wave height (m)

Archetype 1 1

1 2 3

2

1 2 3

3

1 2 3

1 1

1 2 3

2

1 2 3

3

1 2 3

Pile Elevation (m) Surge level (Hs, m) Significant wave height (m)

Archetype 1 2 1

1 2 3

2

1 2 3

3

1 2 3

3 1

1 2 3

2

1 2 3

3

1 2 3

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Define of fragility curve for coastal structures

Research Accomplishments

A fragility, 𝐺

' , can be expressed as

𝐺

'(𝑦) = 𝑄[𝑅 > 𝑆|𝑦]

For elevated structures subjected to storm surge:

• Q

= loading from the model for every combination of (𝐼", 𝑈

$)and/or surge levels, 𝑇

• R= Resistant/ capacity
• Hazard intensities, x = Significant wave height

(𝐼", 𝑈

$), Surge levels, 𝑇.

For different time duration

𝑄

5 𝑗𝑜 𝑌 ℎ𝑝𝑣𝑠𝑡 = 1 − 1 − 𝑄 5 𝑗𝑜 𝑍 ℎ𝑝𝑣𝑠𝑡 A B

Damage State

Window/ door failure Wall failure Floor failure 0 (no damage) <1% No No 1 (Minor damage) >1% and <5% No No 2 (Moderate) >5% and <25% >5% and <25% >5% and <25% 3 (Severe damage) >25% and <50% >25% and <50% >25% and <50% 4 (Destruction) >50% >50% >50%

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Component Resistance Values Used to Model Residential Buildings (Hazus 2.1- Hurricane)

Research Accomplishments

Component Distribution Parameters Window on 1 story Weibull C = 54.49psf, k = 4.7 Window on 2 story Weibull C = 38.7psf, k = 4.8 Entry door Normal Mean=50psf, COV=0.2 Toe-nail Normal Mean=415lb, COV=0.25

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Research Accomplishments

Fragilities surfaces for building subjected to wave and surge Probability of failure base on two environmental variables given some damage states

• Significant wave height
• Surge/flood level

Example: when significant wave height = 2.0m, surge level = 2.5m => Probability of failure due to Damage State 1 = 19% when building was raised 1 m from the ground

Task 3: Develop performance based design examples to illustrate methodology for engineering practice (Yr 2). Application to Galveston, TX

• Damage to residential housing
• Crystal Beach, TX
• Hurricane Ike (2008)

Crystal Beach, TX

Applications

Applications

Current Flood Maps for Crystal Beach Online Mapping Tool: similar elevations

Applications

Study area: 394 houses. Use appropriate FIRM and distance from shore to estimate house elevation Date of Construction

Applications

Approximate elevation, LCM

Applications

Overlay building stock with hazard:

• Hs max, from ADCIRC simulations by Bret

Webb, U. South Alabama. via NIST project

• Houses are colored by the age classification.
• Fragility curve selection in Tomiczek et al.

Applications

Applications

Results: Likelihood of Failure (Collapse Limit State)

Applications

Results: Likelihood of Failure (Collapse Limit State)

Applications

Likelihood of Failure (Collapse Limit State) Increase elevation of all structures by 1 m Potential for decision support tool

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

End End Us User Eng Engag agement

We plan to work with the following people involved in the End-User Transition:

• HAZUS Program Manager at FEMA HQ
• FEMA Building Science Division
• Chad Berginnis, ASFPM Executive Director and CRC Advisory Board Member
• USACE Institute for Water Resources
• Coastal Engineer, FEMA Risk Analysis Branch, Atlanta GA

Revisions to FEMA 55 Coastal Construction Manual and ASCE 7-16

• Workshop July 19+20, 2017 at OSU to develop research roadmap and

implementation plan

• FEMA-55 CCM committee lead
• Chris Jones, Chair of ASCE 7 Flood Load Subcommittee
• Gary Chock, Chair of ASCE 7 Tsunami Subcommittee

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Pr Proposed Follow-on

• n Wor
• rk

Tracking modeling uncertainties

3 October 2016 43

• 1. Building Data
• 2. Energy/EPN
• 3. Water
• 4. Transportation
• 5. Communication

Fragility Functions Expected Damage

Kennedy et al., 2011

Hazard Data (ADCIRC/SWAN)

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Pr Proposed Follow-on

• n Wor
• rk
• Comparison of new physics-based fragilities to

empirical and Hazus fragilities

• Application to NJ coast (post-Sandy)

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Pr Proposed Follow-on

• n Wor
• rk

Mitigation measures for existing and new construction: Examples of mitigation measures to increase building performance for wind (a – d) from FEMA’s Coastal Construction Manual. (e) Example of testing elevated bridge at HWRL.

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Pr Proposed Follow-on

• n Wor
• rk

Debris Impact: (a) Flood-borne debris hazards (FEMA 2000), (b) log-structure impact at 1:1 scale and (c) shipping container-column impact at 1:25 scale.

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Pr Proposed Follow-on

• n Wor
• rk

Prototype-scale testing for breakaway walls: (a) New construction of elevated home with breakaway wall on first floor. (b) Example of prototype-scale testing at HWRL of wave loads

• n subassembly and (c) structural failure.

CRC 2nd Annual Meeting

• Feb. 1-3, 2017

The University of North Carolina at Chapel Hill

Thank you

Daniel Cox (dan.cox@oregonstate.edu), John van de Lindt, Kevin Cueto, Diego Delgado, Trung Do, Ben Hunter, Tori Johnson, Pedro Lomonaco, Hyoungsu Park, William Short