Embed Size (px)
Citation preview
Structural Health Monitoring of Civil Infrastructure:
from Research to Engineering Practice
B. F. Spencer, Jr.Nathan M. and Anne M. Newmark Endowed Chair in Civil Engineering
Director, Newmark Structural Engineering LaboratoryDirector, MUST‐SIM: NEES@Illinois
Director, Smart Structures Technology Laboratory (SSTL)
Outline
◊Background and Motivation◊Enabling Wireless Smart Sensor Technology
□ High‐fidelity Hardware□ Service‐oriented Software Framework
◊ Full‐scale Implementations◊ Future Directions◊Conclusions
To detect and localize damage Deteriorating structures Lessons from catastrophic bridge
collapses Limits in visual inspection SHM reduce inspection cost, while
providing increased public safety Lifetime monitoring of future
construction projects
Corrosion
Fatigue (Bay Bridge in CA, 2009)
Sung‐su bridge collapsein Korea (1994)
I‐35 bridge collapse in MN (2007)
1997 2006
$ billion
20
40
21.5 23.030.1
40.4
System expansion (39.9% increase)Rehabilitating (75.2% increase)
Capital investment (US DOT 2008)
Why Monitor Infrastructure?
Why Monitor Infrastructure?
◊ To validate the structural designs and characterize performance (e.g., develop database)
◊ To assist with infrastructure maintenance
◊ To design appropriate retrofit measures
◊ Improve seismic risk assessment
◊ To monitor and control the construction process
◊ To characterize loads in situ
◊ Assess load carrying capacities
◊ To assist with emergency response efforts, including building evacuation and traffic control
Structural Health Monitoring
Limitation of traditional methods Dense arrays of sensor are required to effectively monitor structures Wired monitoring systems are expensive, with much of the cost derived
from cabling and installation Centralized data collection is not challenging for monitoring large civil
infrastructure
Cabling & instrumentation for 400 wired strain sensors
Bill Emerson Memorial Bridge SHM system: $1.3M for 86 sensors, ~$15k/sensor (Caicedo et al. 2002; Celebi et al. 2004)
Enabling Wireless Smart Sensor Technology
* low cost* ease of installation
* accurate data
7
Computers are Faster and Cheaper
19901980 2005
$400,000/MIPS (Cray-I)
$500/MIPS (i860)$1/MIPS
.
. .
8
Where Computing is Done
year
streaming informationto/from physical world
Number CrunchingData Storage
productivityinteractive
Networking Plays an Important Role
1946
The first computer
High performance Computing
PersonalComputers
Internet
IoT/WSN
Cyber Physical Systems
19651980
19952010
Vision of the Future: Internet of Things
◊ Vast networks of sensors, installed in the urban environment, can serve as the eyes and ears of a first line of defense against various vulnerabilities
◊ Different types of sensors can be envisioned, each reporting information that provides insight to the status of critical infrastructure in real‐time
◊ Wireless Smart Sensors will act as the fundamental building block for such sensor networks□ Low costs allow for dense deployment as needed
□ Modularity provides inherent flexibility for use in both permanent and temporary applications
Smart Sensors
◊ But... what is a smart sensor?
◊ First open source hardware and software platforms developed at Berkeley in the late 1990s as part of DARPA’s Smart Dust project (Mica family of smart sensors).
Radio
Data Storage
Embedded Computing
Local Power
Historic Decade of Smart Sensors
CrossbowMica2 (2004)
BTnode rev3 (2004)
EYES (2003)
WINS 1 (1999)
U3 (2002)
Prototype by Prof. Lynch (2002)
Intel Imote (2004)
Imote2 (2006)
SmartDustWeC (1999)
Full‐scale Implementations to Bridges (selected)Implementation Purpose Sensor
Platform# of
NodesSensor
(channels) Test duration In‐network Processing Results
Alamosa Canyon (Lynch et al. 2003)
Wireless system proof‐of‐concept
WiMMS(prototype) 7 Accel. (14) Short‐term Independent
FFTModal frequencies
Ben Franklin (Galbreath et al. 2004)
Demonstration of wireless system with remote programming
MicrostrainSG‐Link 10 Strain +
Temp. (10)Continuous Medium‐term None
Streaming real‐time strain time histories
Guemdang(Lynch et al. 2006)
Wireless sensor prototype and embedded computing validation
WiMMS(prototype) 14 Accel. (14) Short‐term
Independent FFT and peak picking
Modal frequencies, ODS
Golden Gate (Pakzadet al. 2008)
Test wireless sensor network requiring multi‐hop communication
MicaZ 64 Accel. (128) Temp. (64) Medium‐term None
Modal analysis after central data collection
Gi‐Lu(Weng et al. 2008)
Test for health monitoring of cable‐stayed bridge
WiMMS(prototype) 12 Accel. (12) Short‐term None
Modal analysis and cable tension force
Smart Sensor Monitoring of the Golden Gate Bridge (2008)
Smart Sensor Monitoring of the Golden Gate Bridge (2008)
Requires 12 hours to transmit 80 seconds of data back to the base station!
Full‐scale Implementations to Bridges (selected)Implementation Purpose Sensor
Platform# of
NodesSensor
(channels) Test duration In‐network Processing Results
Alamosa Canyon (Lynch et al. 2003)
Wireless system proof‐of‐concept
WiMMS(prototype) 7 Accel. (14) Short‐term Independent
FFTModal frequencies
Ben Franklin (Galbreath et al. 2004)
Demonstration of wireless system with remote programming
MicrostrainSG‐Link 10 Strain +
Temp. (10)Continuous Medium‐term None
Streaming real‐time strain time histories
Guemdang(Lynch et al. 2006)
Wireless sensor prototype and embedded computing validation
WiMMS(prototype) 14 Accel. (14) Short‐term
Independent FFT and peak picking
Modal frequencies, ODS
Golden Gate (Pakzadet al. 2006)
Test wireless sensor network requiring multi‐hop communication
MicaZ 64 Accel. (128) Temp. (64) Medium‐term None
Modal analysis after central data collection
Gi‐Lu(Weng et al. 2008)
Test for health monitoring of cable‐stayed bridge
WiMMS(prototype) 12 Accel. (12) Short‐term None
Modal analysis and cable tension force
Stork(Meyer et al. 2010)
Validation of long‐term field test using wireless sensor network
Tmote Sky 6 Accel. (6) Long‐term None Cable tension force
Ferriby Road(Hoult et al. 2010)
Evaluation of the potential of wireless sensor network
MicaZ 7
Crack (3)Inclin. (3)Temp (7)Humid.(6)
Long‐term NoneCrack growth and inclination of deck
New Carquinez(Lynch et al. 2010)
Wireless system proof‐of‐concept
Narada/Imote2 30 Accel. (14) Long‐term SSI Modal
Analysis Mode Shapes
What’s Been Lacking?
◊ Hardware□ Platform with computational capacity for high‐data rate
applications and distributed computing□ Sensor hardware that to produce high‐fidelity data that is
appropriate for SHM
◊ Software□ Middleware services to acquire high‐fidelity data□ Application software to implement distributed SHM□ Flexible software that supports network and application scalability
◊ Full‐scale implementation considerations□ Communication performance evaluation□ Autonomous network operation□ Power management□ Fault tolerance
High‐fidelity Hardware
Imote2 with Sensor Boards
MicaZ Telos iMote2Microprocessor Atmel ATmega128L TI MSP430 Intel XScalePXA271
Clock speed (MHz) 7.373 8 13‐412Active Power (mW) 24 10 44 @ 13 MHz
Non‐volatile memory (bytes) 128 K (Flash) + 512 K (EEPROM) 48K (Flash) 32 M (Flash)
Volatile memory (bytes) 4 K 1024K 256 K +32 M (SDRAM)
Dimensions (mm) / Weight (g) 58*32*7 / 18 65*31*6 / 23 36 *48*9 / 12
From MEMSIC (2010)
Sensor Board
External Antenna(Antenova 2.4GHz)
iMote2
Battery Boardw/ 3 AAA Batteries
Basic Sensor Board
LEDBasic I/O
Basic I/O
PMICPXA27
Advanced I/O
Advanced I/O
Mini USB Connector
CC2420Antenna
SMA Connector
48mm
36mm
Imote2 and Basic Sensor Board
Imote2: Top and Bottom
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5-1
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1
Time (sec)0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5
-1
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1
Time (sec)
Aliasing
◊ Consider a 48 Hz signal◊ What if the signal is sampled
at 50 Hz?◊ The resulting measured signal
has an apparent frequency of 2 Hz
Aliasing
◊ Cannot have significant energy above the Nyquist frequency (fs/2)
◊ Anti‐aliasing filters must be used in dynamic measurements to preserve signal integrity
fN fs
?
Multifunctional Sensor Boards
External InputConnector
SHM‐A Board (bottom)
Basic Connector
Programmable Filterw/ ADC Quickfilter
QF4A512
SHM‐A Board (top)
AccelerometerST Microelectronics
LIS344ALH
OP AMPTI OPA 4344
Humidity & Temp. Sensor
SHT11
Light SensorTAOS 2561
Basic Connector
SHM‐A and SHM‐H Board
SHM‐H Board
Anemometer(RM Young,Model 8100)
Interfaced
Voltage
SHM‐DAQ Board
• Data Acquisition using commercial voltage‐output sensors
• Output Voltage: ‐5 to 5 V0 to 5 V
High‐sensitivity Accelerometer board (SHM‐H board) Pseudo‐static test: slender column with lumped mass on top
P ¼ ASTM A53 steel pipe(0.54OD, 0.36ID)
10.5’
SHM‐ASHM‐H
5kgmass
FE analysis (1~3th freq.):0.14Hz, 4.42Hz, 14.57Hz
0.1406 Hz 6Hz sampling rate was used Performed well in low‐frequency range
(< 0.2Hz)
Strain monitoring using SHM‐S board
Strain sensor board for Imote2 platform High‐throughput synchronized strain monitoringHigh precision auto‐balanced Wheatstone bridgeUp to 2500 times gain 0.3 µ‐strain resolution at 20Hz B/W Stacked use with SHM‐A or SHM‐DAQ boardBoth foil‐type strain gage and magnetic strain sensor can be used with SHM‐S board
SHM‐S board: top, stacked on SHM‐DAQ and stacked on SHM‐A
RemoteCommand SHMSAutoBalance
Tokyo Sokki magnet strain checker
High‐precision Strain sensor board (SHM‐S board)
SHM‐S vs. Ni‐DAQ (foil type)
0 1 2 3 4 5-40
-20
0
20
40
time(sec)
-st
rain
Detail (high-level strain)
195 196 197 198 199 200
-1
0
1
2
time(sec)
-st
rain
Detail (low-level strain)
0 10 20 30 40 50
10-5
100
frequency(Hz)
-st
rain
2 /Hz
PSD
0 50 100 150 200-50
0
50
time(sec)
-st
rain
Time history
Wired: Ni-DAQWireless: SHMS
Wired: Ni-DAQWireless: SHMS Wired: Ni-DAQ
Wireless: SHMS
Wired: Ni-DAQWireless: SHMS
a)
d)c)
b)
Pressure and D2A Sensor Board
TI‐DAC8565
SHM‐D2A Board
AMSYS 5812
• 16bits, Digital to Analog conversion of 4 channels
Stacked with SHM‐DAQ Board
SHM‐P Board • Measure wind/airpressure
Imote2 Sensor Enclosure
Acrylic JigSolar PanelExternal 5dBiAntenna
Uni‐directionalmagnet
RechargeableBattery
Antenna Cable
Bolt & Spacer
Future Hardware Enhancement
24‐bit DAQ (TI ADS1274, Delta‐sigma type ADC) Delta‐sigma noise shaping technology Low‐power consumption suitable for wireless sensor application Support full capabilities of low‐noise/high‐sensitivity sensors Make it easy to implement/design new external sensor boards
Service‐oriented Software Framework
Service‐Oriented Architecture (SOA)
◊ Previous smart sensor applications:□ Significant effort to create very specific applications□ Difficult to modify for other applications, even with extensive CS
knowledge
◊ Service‐Oriented Architecture simplifies SHM software development□ Applications are comprised of manageable, modular services that
exchange data in a common format□ The middleware framework connects the services by providing
communication and coordination
SDLV
Numerical ServicesApplication Services Foundation Services SHM Application
Time Synchronization
◊Each sensor has its own local clock which drifts over time
◊ Synchronization errors can be reduced to ~20s
◊Not the whole story…beacon
exchange beacon time&adjust clocks
Synchronized Sensing Service
◊ Uncertainty in start of acquisition □ Independent processors
Sensor 1
Sensor 2
t0,1
t0,2
Resampling middleware service achieves synchronized sensing
27.9 27.95 28-3
-2
-1
0
1
2
3
Time (sec)
Sam
ples
27.9 27.95 28-3
-2
-1
0
1
2
3
Time (sec)
Sam
ples
Power Management and Energy Harvesting
Base Station
DeepSleep
DeepSleep
DeepSleep
DeepSleep
WakeUp and Listen
Effective Power Management with Autonomous OperationSnoozeAlarm:ThresholdSentry:AutoMonitor:
Energy HarvestingSolar panel (or Micro wind turbine) + Rechargeable BatteryPMIC charger manipulates voltage and current for fast and stable charging.
PMIC
Imote2Solar Panels Micro Wind Turbine
ISHMP Services Toolsuite
Fault tolerant WSSNSkipping of unresponsive nodesData storage in non‐volatile memoryMonitoring of sensor powerExclusion of low‐power sensor nodesWatchdog timer and etc…
Enhanced operationAutonomous resuming of AutoMonitorThresholdSentry for multi‐hop communicationEmail notification of structural response and network anomalies
Reliable communication Time synchronization Numerical library (e.g. FFT, SVD, etc.)
Network data acquisition Decentralized data aggregation (DDA) Multi‐hop communicationHardware (Imote2)
OS (TinyOS)
Middleware services
SHM application
Correlation function estimation Eigensystem realization algorithm (ERA) Stochastic subspace identification (SSI) Frequency domain decomposition (FDD)
Test applications for Toolkit components Radio and antenna testing
Foundation Services
Network Services
Application Services
Tools and Utilities
ISHMP Service Toolsuite
http://shm.cs.uiuc.edu/
Illinois Structural Health Monitoring Project (ISHMP)
http://shm.cs.uiuc.edu
Full‐scale Implementation
2nd Jindo Bridge Haenam(Inland)
Daejeon (KAIST)
Seoul
Jindo
JindoBridges
Jeju Island
2nd Jindo Bridge
Type Cable-stayed bridge
Spans 70+344+70 = 484m
Girder Steel box (12.55m width)
Design velocity 70 km/hr
Designed by Yooshin cooperation (2000, Korea)
Constructed by Hyundai construction (2006, Korea)
(September 2008 – 2012)
US : B.F. Spencer, Jr. & G. Agha (UIUC)Korea : H.J. Jung & C.B. Yun (KAIST), H.K. Kim (SNU)
J.W. Seo (Hyundai Institute of Const. Tech.)Japan : Y. Fujino & T. Nagayama (U. of Tokyo)
US : B.F. Spencer, Jr. & G. Agha (UIUC)Korea : H.J. Jung & C.B. Yun (KAIST), H.K. Kim (SNU)
J.W. Seo (Hyundai Institute of Const. Tech.)Japan : Y. Fujino & T. Nagayama (U. of Tokyo)
US‐Korea‐Japan Collaborative Projecton SHM Test‐bed Using Wireless Smart Sensor Network
Deployment in 2009 (70 Nodes)Cable : 8Deck : 22Pylon : 3Total : 33
Cable : 7Deck : 26Pylon : 3Wind: 1
Total : 37
Nodes on pylon top(powered by solar cell)Deck Nodes
Reference NodesNodes on pylons Nodes on cables
Nodes on cables(powered by solar cell) Amemometer
Max. 420 channels of sensors(207 ch. Acceleration & 3 ch. Wind)
Cable node
Cable node
Pylon node
Base station Pylon node
Wind turbine
3D ultra‐sonic anemometer
Deck node
TeamViewer(for remotely access to basestation)
Deployment in 2010
Deployment in 2010
104 80 25H
88 121 102 144 86 98 96 116 81 18 150 39 85 52
2 9 3 7 69 66 53 41 39 64 100 56 8 3WT23 6 20 152 33 35 53 83 54 69 114
69 53 88 37 133 34 87 120 104 24 51 81 85 113 51 32 24 73 23 89 48 2 34 64
52 132 6
72
170 145 35 21 105 86 23 3 41 70 7 150 151 56 135 119 4 40 20 133 65
118
24
T101S
70
32ST
20113
71151
19W
136T
145H
135H
133H
149W
146W
144H
135H
132H
129H
72 24
T101
: Jindo Deck network (single hop) : Haenam Deck network (single hop)
: Jindo Cable network (multi hop) : Haenam Cable network (single hop)
W : Anemometer (SHM‐DAQ)T : Temperature
WT : Wind Turbine powerH : High‐sensitivity (SHM‐H)
S : Strain sensor (SHM‐S1)ST : Strain + Temp correction (SHM‐S2)
Haenam side(North)
Jindo side (South)
East West East WestDeck networks
Cable networks
West
East
West
East
Haenam sidePylon
Jindo sidePylon
11869 53 88 37 133 34 87 120 104 24 51 81
695388371333487120104245181
Haenam side Jindo sideDeck Cable Deck Cable
Communication channel Ch 25 Ch 15 Ch 26 Ch 20
Network size 30 nodes 26 nodes 31 nodes 26 nodes
SHM‐H board (cluster head) 5 nodes ‐ 5 nodes ‐
SHM‐DAQ board (wind) 1 nodes ‐ 2 nodes ‐
Sentry node 3 nodes 2 nodes 3 nodes 2 nodes
Temperature node (exposed) 1 nodes ‐ 1 nodes 1 nodes
FunctionalitiesCommon RemoteSensing, AutoUtilsCommand, ChargerControl,
ThresholdSentry and SnoozeAlarm
Particular DDA CTE DDA CTE
Communication protocol Single‐hop Single‐hop Single‐hop Multi‐hop
In total, 669 sensing channels with 113 sensor nodes were deployed
In total, 669 sensing channels with 113 sensor nodes were deployed
Evaluation
Typhoon Kompasu (2010.09.01)
Aug. 29. 21:00Birth of Typhoon
Sep. 2, 3:00
Sep. 1, 21:00 (Closest to Jindo)
Sep. 2, 9:00
Sep. 2, 15:00
Sep. 2, 21:00
Sep. 3, 9:00
Sep. 3, 21:00
Korea
Japan
China
Jindo Bridges
Typhoon center
Jindo Bridge
Evaluation: Wind Monitoring
SHM‐DAQ + AnemometerWind Speed
Wind Direction
Measured Wind by KMA‐Jindo (Mt. Chun‐Chal) at Sep. 01, 2010.
15‐20m/s
0 200 400 600 800 10000
10
20
30
Wind
Spe
ed (m
/s)
0 200 400 600 800 1000100
150
200
250
Azim
uth
Dir.
(Deg
)
0 200 400 600 800 1000-40
-20
0
20
40
Time (sec)
Eleva
tion
(Deg
)
0 200 400 600 800 10000
10
20
30
Wind
Spe
ed (m
/s)
0 200 400 600 800 1000100
150
200
250
Azim
uth
Dir.
(Deg
)
0 200 400 600 800 1000-40
-20
0
20
40
Time (sec)
Eleva
tion
(Deg
)
Jindo
Haenam(Inland)
41
Narrow Strait:Increased wind speed ‐Slight change in direction
Anemometer: Jindo‐side (at 21:14)
Anemometer: Haenam‐side (at 21:12)
Ch1: wind speedCh2: wind direction (horizontal)Ch3: wind direction (vertical)
Evaluation: Acceleration Responses
0 200 400 600 800 1000
-20
0
20
time(s)
acc(
mg)
1 (center span)
1 2 3 4 5 6 7 8 9 10 11 12 13 14
0 200 400 600 800 1000
-20
0
20
time(s)ac
c(m
g)
2
0 200 400 600 800 1000
-20
0
20
time(s)
acc(
mg)
3
0 200 400 600 800 1000
-20
0
20
time(s)
acc(
mg)
4
0 200 400 600 800 1000
-20
0
20
time(s)
acc(
mg)
5
0 200 400 600 800 1000
-20
0
20
time(s)
acc(
mg)
6
0 200 400 600 800 1000
-20
0
20
time(s)
acc(
mg)
7
0 200 400 600 800 1000
-20
0
20
time(s)
acc(
mg)
8
0 200 400 600 800 1000
-20
0
20
time(s)
acc(
mg)
9
0 200 400 600 800 1000
-20
0
20
time(s)
acc(
mg) 10
0 200 400 600 800 1000
-20
0
20
time(s)
acc(
mg)
11 (pylon)
0 200 400 600 800 1000
-20
0
20
time(s)
acc(
mg) 12
0 200 400 600 800 1000
-20
0
20
time(s)
acc(
mg)
13
0 200 400 600 800 1000
-20
0
20
time(s)
acc(
mg)
14 (girder end)
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 510-4
10-2
100
102
freq(hz)
z-axis PSD (Haenam side deck)0.446
0.6451.033
1.3561.555
1.653
1.871
2.2592.405 2.812
Evaluation: System ID
NExT & ERA method Identified frequency range: 0~3Hz 1000 seconds data at 25Hz sampling
ModeName
WiredSystem (2007)
FE analysis
WSSN (During Kompasu)
Haenam Jindo Avg.
DV1 0.440 0.442 0.446 0.446 0.446
DV2 0.659 0.647 0.645 0.647 0.646
DV3 1.050 1.001 1.033 1.033 1.033
DV4 1.367 1.247 1.356 1.342 1.349
DV5 1.587 1.349 1.555 1.549 1.552
DV6 1.660 1.460 1.653 1.635 1.644
DT1 ‐ 1.789 1.798 1.802 1.800
DV7 1.856 1.586 1.871 1.870 1.871
DV8 2.319 2.115 2.259 2.261 2.260
DV9 2.808 2.561 2.812 2.813 2.813
Identified natural frequencies (Hz)
DV1
DV2
DV3
DT1
DV9
Evaluation: Pylon Strain
Node 70 (sensor node)
Node 32 (sensor node + solar panel on it)
Solar panel for node 70
Active Dummy(Node 32)
(Node 70)Active
Enclosure inside
0 100 200 300 400-4
-2
0
2
4
time(sec)
mg
Acceleration time history
0 2 4 6 8 10 1210-5
100
frequency(Hz)
mg2 /ro
ot(H
z)
Acceleration PSD
0 100 200 300 400
-5
0
5
time(sec)
u-st
rain
Strain time history
0 2 4 6 8 10 1210-5
100
frequency(Hz)
ustra
in2 /ro
ot(H
z)
Strain PSD
x-axisy-axisz-axis
x-axisy-axisz-axis
2.4Hz (pylon bending)
2.4Hz (pylon bending)
Synchronized acceleration & strain sensing at 25Hz For pylon, strain measurement was more informative
09/21 09/22 09/23 09/24 09/25 09/26 09/27 09/28 09/29 09/30 10/01 10/02 10/03 10/04 10/05 10/06 10/07 10/08 10/09 10/10 10/11 10/12
-400
-200
0
Date (MM/DD)
Stra
in d
rift (
u-st
rain
)
Strain drift due to temperature effect
Node70: Quarter bridgeNode32: Half bridge with a dummy gage
Temperature compensation using the half bridge with a dummy gage worked
Evaluation: Girder Strain
Lead wire
Embedded Wheatstone bridge
Magnet strain checker
Sensor nodeat midspan
at pylon bearing position
At mid span At bearing location Using magnet strain checker with SHM‐S
0 50 100 150 200 250 300 350 400-8-6-4-202
time(sec)
u-st
rain
Strain time history (at pylon bearing position)
0 50 100 150 200 250 300 350 400-50
0
50
time(sec)
u-st
rain
Strain time history (at midspan)
double peaks
compression
tension
due to slip
Strain responses under traffic loadings Compressive strain at pylon bearing and tensile strain at midspan Double peaks (?) in the strain measurements at pylon bearing location
Evaluation: Strain Analysis Truck loading simulation using FE model
Box girder model Pylon model
3D Frame modelExtruded
appearance
5‐tonf truck 12 tonf when full loaded 60 km/hr (38 mile/hr)
using MIDAS/Civil
0 10 20 30 40-15
-10
-5
0
time(sec)
u-st
rain
Girder strain at pylon bearing location
0 10 20 30 40
0
20
40
time(sec)
u-st
rain
Girder strain at midspan
6.0 sec
3.1 sec
51m
100 m
(51m)/(3.1sec)*(3600sec/hr)/(1000m/km) = 59.2 km/hr
(north bound)Truck moving
Truck moving(north bound)
(100m)/(6.0sec)*(3600sec/hr)/(1000m/km) = 60 km/hr
Double peaks are caused by the double‐concaved influence line Can be utilized for estimation of vehicle’s speed, moving direction, and
weight (Bridge Weigh‐in‐motion, BWIM)
Evaluation: Battery Monitor
0:00 2 4 6 8 10 12 14 16 18 20 22 0:00 2 4 6 8 10 12 14 16 18 20 22 0:004
4.05
4.1
4.15
4.2
4.25
Time (HH:MM)
Bat
tery
Vol
tage
(V)
East-side nodes
0:00 2 4 6 8 10 12 14 16 18 20 22 0:00 2 4 6 8 10 12 14 16 18 20 22 0:00
0
100
200
Time (HH:MM)
Cha
rgin
g cu
rrent
(mA
)
East-side nodes
151 96116144 64 70135150132 39 2129 85 52
09/08/10 09/09/10 09/10/10 09/11/10 09/12/10
24
26
28
30
32
34
Date (MM/DD/YY)
Bat
tery
Vol
tage
(V)
Temperature nodes
151 64 39 85 3 35 83 86
Temperature monitoring Temp. inside enclosure
Temp. outside enclosure
Evaluation: Cable Tension and Temperature
c1 c2 c3 c4 c5 c6 c7 c8 c9 c11 c120
50
100
150
200
250
300
350
Cable ID
Tens
ion
(tonf
)
Design TensionPark et al. (2008)CableTensionEstimation
Cable tension estimation
[ Cable tension comparisons ][ Cable ID ]
Evaluation: Long‐term performance of Jindo‐side WSSN
Battery voltage (V)
Charging current (mA)
Number of responsive
node (out of 31)
Basestation problem
Average charging current > 100mA
Responsive rate : 70~85%
Deployment Achievements
669 sensors at 113 nodes High-sensitivity accelerometer board Solar panel with rechargeable battery Wind-induced energy harvesting
• Hardware
Multi-hop reliable data transfer and printing Fault tolerant operation Remote software updates Better user interface
• Software
Modal analysis with system synchronization Vibration-based tension estimation Multi-scale structural health monitoring
(Modal information + Cable tension force) Wind-vibration correlation analysis
International collaboration for test-bed Web-based data repository and sharing
• Analysis & Monitoring • Collaboration
World’s largest deployment to date of wireless sensors for civil infrastructures monitoring.
World’s first long-term deployment.
Next Steps
Wind Load and Wind Effects on Xihoumen Bridge
Hui Li, B.F. Spencer, Yan Yu, Jinping Ou
Add 32 wireless pressure modules(P992 connected to SHM‐DAQ)Range: ± 250‐500PaAccuracy: ≤ 0.5% of full span
SHM‐DAQ board
KAVLICO P992 Low‐level pressure sensor
Wind Load and Wind Effects on Xihoumen Bridge
册 子 岛（ 北 ） 金 塘 岛（ 南 ）
U A 3 / U A 4
北 塔 南 塔U A 4（主跨跨中右幅）
U A 3（主跨跨中左幅）
向册子行驶方向为左幅、向金塘行驶方向为右幅
册 子 岛（ 北 ） 金 塘 岛（ 南 ）AC10
AC11/AC12 AC14/AC15 AC17/AC18
AC16AC13AC4
AC5/AC6
AC11/AC11/AC12 AC16/AC17/AC18
AC13/AC14/AC15
AC4/AC5/AC6
册 子 岛（ 北 ） 金 塘 岛（ 南 ）
U A 3 / U A 4
北 塔 南 塔U A 4（主跨跨中右幅）
U A 3（主跨跨中左幅）
向册子行驶方向为左幅、向金塘行驶方向为右幅
册 子 岛（ 北 ） 金 塘 岛（ 南 ）AC10
AC11/AC12 AC14/AC15 AC17/AC18
AC16AC13AC4
AC5/AC6
AC11/AC11/AC12 AC16/AC17/AC18
AC13/AC14/AC15
AC4/AC5/AC6
Add 12 bidirectional wireless accelerometers in ten sections
Add 10 bidirectional wireless accelerometers
Vibrations of cable
CAC1/CAC2
册子岛（北）
CAC3/CAC4
CAC5/CAC6
CAC9/CAC
CAC11/CACCAC7/CAC8
Wind Load and Wind Effects on Xihoumen Bridge
Campaign Monitoring for Management of Railroad Assets
The Future
Cyberinfrastructure for Smart StructuresBridges
Dams
Tunnels
The Other Infrastructure
• Structure Monitoring•Hurricane & Earthquake Monitoring
• Traffic Monitoring • Emergency Response
• Structure Monitoring•Hurricane & Earthquake Monitoring
• Load Rating• Emergency Response
• Pressure Monitoring• Leakage Detection•Deflection Monitoring• Emergency Response
•Deflection Monitoring• Leakage Detection• Fire Detection• Traffic Monitoring• Emergency Response
WSN
WSN
WSN
WSN
Cloud Services
• WSN applications for structural monitoring• Repository of structure data and monitoring algorithms
• Repository of models
CitySee: City‐wide Urban Sensing
◊ Objective: 20 KM2 urban area in Wuxi□ 4000+ sensor nodes with temperature/humidity and light sensors□ 500+ nodes with CO2 sensor
Final Goal: 10,000 Nodes, 100 KM2
(Courtesy of Prof. Yunhao Liu of Tsinghua U.)
Asia-Pacific Summer School on Smart Structure Technology
To enhance students' understanding of the cross-disciplinary technological developments on the emerging subjects of smart structure technologiesTo develop the cross-cultural human-network and understanding for the future cooperation in their professional career development.3 weeks program for 6 years among US/Korea//China/Japan/India supported byNSF, NRF, JSPS, & NSFC.Coordinators
Goals of APSS Program
Korea : C-B. Yun (KAIST) US : B.F. Spencer (UIUC) Japan : Y. Fujino (U. of Tokyo) China : L. Sun (Tongji U.)
1st APSS at KAIST, Korea, 2008SISTeC‐KAIST, July 28 ‐August 16, 2008
50 graduate students attended. http://sstl.cee.uiuc.edu/apss
2nd APSS at U. of Illinois, USA, 2009July 3 – July 25, 2009
45 graduate students attended.
3rd APSS at U. of Tokyo , Japan, 2010July 15 – August 4, 2010
40 graduate students attended.
4th APSS at Tongji U., China, 2011July 28 – August 10, 2011
48 graduate students attended.
5th APSS at IIT, India, 2012July 23‐August 11, 2012 48 graduate students attended. 6th APSS at KAIST, Korea, 2013
Concluding Remarks
Concluding Remarks
◊ This paper discusses recent advances and field validationof a state‐of‐the‐art WSSN framework developed atthe University of Illinois at Urbana‐Champaign
◊ The open source ISHMP Services Toolsuite a wide variety of services and fault‐tolerant features
◊ Full‐scale WSSNs are realized for the purpose of SHM andthe evaluation as well as the data analysis show high practicality of wireless SHM systems for civil infrastructure
◊ Tremendous potential and a level of maturity of WSSN for SHM has been demonstrated through full‐scale deployment on the Jindo Bridge in Korea
◊ Education of the next generation of engineers in smart structures technology is of high importance
Acknowledgement
◊ NSF Grant CMS 06‐00433 (Dr. S.C. Liu, Program Manager)
◊ Global Research Network Program (NRF‐2008‐220‐D00117)from the National Research Foundation in Korea
◊ Smart Infra‐Structure Technology Center (SISTeC) at KAIST in Korea
◊ Ministry of Land, Transport and Maritime Affairs in Korea
◊ Hyundai Construction Co. Ltd.
Thank you for your attention!
