Research Summary

Dr. Elizabeth K. Ervin Assistant Professor

Department of Civil Engineering

Background

Ph.D., Carnegie Mellon University, Mechanical Engineering. M.S., Vanderbilt University, Civil Engineering; B.S., Tennessee

Technological University, Civil Engineering. Joined as Assistant Professor, Department of Civil Engineering,

in August 2006. September 2001 to March 2006 Bechtel Bettis, Inc. Bettis Atomic Power Laboratory, a Department of Energy/U.S.

Navy Contractor, Pittsburgh, Pennsylvania. Reactor Technology Activity, Reactor Engineering Division,

Structural Methods Analysis and Design, Shock and Vibration; Acoustic Design & Development, Noise, Vibration, & Shock

Also, Summer Research Program, Air Force Research Laboratories

Air Vehicles Directorate Structures Division, Structural Design & Development Branch (ARFL/VASD), WPAFB.

Funding from NSF, DOD, ORNL, NRC.

“Extreme” Effects

Contact

Shock

Blast

Seismic

Other “abnormal” events that cause load amplification

Aging

Fatigue

Fracture

Other “unexpected” outcomes from normal or abnormal loading

Triggering Causes of Infrastructure Failure

External Events, Natural Hazards, Manmade Events

Seismic

Fire Overload

Weidlinger

Modes of Failure Time Scale

Ductile

Brittle

Progressive

Weidlinger

“Health”

How can a massive structure be instrumented to show damage using only measurements?

What parameter will best indicate damage?

What is the uncertainty in this indicator?

How can this indicator show cumulative damage?

When does a structure become too damaged and need reinforcement or demolition?

Will the implemented method be cost beneficial?

Rytter’s Health Hierarchy:

Detection

Location

Severity

Prognosis

Two main approaches:

Inverse problem

Pattern Recognition

Data acquisition

Pre-processing

Feature extraction

Classification

Decision

0

5000

10000

15000

2 3 4 5 6 7 8 9 10

Am

pli

tud

e (g

/V)

Frequency (Hz)

Undamaged

Damage State 1

Damage State 2

34.60%

27.75%

3.78 Hz 3.78 Hz

5.78 Hz

3.78 Hz 3.78 Hz

5.78 Hz

8 Hz

Incrementally Damaged Tower: Experimental Series 1

Increasing Damage by Removing Bracing

Results

• Decreasing frequency

• Increasing displacement

Consistent with decreasing stiffness

1 2

Incrementally Damaged Building: Experimental Series 2

Increasing Damage by Removing Bracing

10 Damage Cases, 4 Reinforced Cases; Both modal tap tests and shake tests.

Schematic of Building Experiment

Xcitex ProAnalyst Post-Processing Edge Detection

for Displacement

Tri-axial

Accelerometer

Calibration

Table

Motion

0 10 20 30 40 50 60 70 80 90 100

Obtained Modal Peaks from STAR Modal using tap test data

Experimental Series 2 Results

Frequency (Hz)

Fre

qu

ency

Res

po

nse

Am

pli

tud

e

Incr

easi

ng

Dam

age

by R

emo

vin

g B

raci

ng

Splitting

Frequencies

Decreasing

Frequencies

Spatial Scale of Failure

Nano-scale

Meter-scale

SDOF Model of Cracked Beam

Parameter α ‘represents’ depth

of crack

Cracked Beam FRF Response (Worden)

Numerical Model

Fixed-Free Beam

Kt1 = k1= ∞

Kt2 = k2= 0

gap

Experimental Study

Shake

Table

High

Speed

Camera

Motion

Impact Mechanics and

Model Verification

Machinery Example

Other Areas of Interest

Various Modes of Excitation, including Shock Loading

Blast with Fluid Effects

Impact Response of Beam Structures

Contact Mechanics - Other than hybrid linear mapping

Active and Passive Vibration Control

Coupled Motions: Axial, Torsional, Transverse Vibration

Acoustics and flow-induced vibration

Composite material modeling

Experimental verification

Model reduction methods

Nuclear Engineering education

Outreach

More Information

http://home.olemiss.edu/~eke/

Homepage, Labpage, Nukepage