Wireless sensor network and its application

to structural health monitoringTomonori NagayamaAssistant professor University of Tokyo

2010/08/06

2

A historic decade of wireless smart sensor research

Berkeley MoteMica2 (2004)

BTnode rev3 (2004)EYES (2003)WINS 1 (1999)

U3 (2002)

Prototype by Prof. Lynch (2002)

Intel Imote (2004)

Imote2 (2006)

In 2000, US Defense Advanced Research Projects Agency (DARPA)’s Networked Embedded Systems Technology (NEST) program funded projects aiming to build dependable, real-time, distributed, embedded applications comprising 100-100,000 simple computing nodes.

Since then a variety of WS platforms have been developed

3

Application examples:

• Soil moisture & temperature• Ambient temperature & humidity

eKo, MEMSIC Corporation

-Agricultural application-

Application examples:

4

Energy monitoring solution

Ecowizard, Crossbow Japan

5

Application examples:

MEMS inclinometer

Volumetric water content meter

Sensor units

Data link by Cell Phone Network (KDDI)

Data link by wireless chip MU-2(< 600m for each node)

Data can be accessed through internet

Warning to adjacent residents

Low-cost, easy instllation

Database for regional disaster mitigation programs

時間

傾斜

土壌水分

法尻部の土壌水分

No wiring work Low cost, easy instlation

(1) Sensor units with inclinometer and watercontent sensor.

(3) Data is collected at the gateway unit.

(4) Data is transferred via cell phone network to Internet. (every 10 mins.)

(5) Central server receives dataand issue warning if abnormal data is found.

(2) Data is relayed by mult-hopping among sensor units.

Assoc. Prof. T. Uchimura, University of Tokyo

-Landslide monitoring-

6

The prototype system is installed on steep slopes in Kobe City, Japan.

Senjo-Dani Slope

Ogawa-Dani Slope(which failed due to 1995 Hyogoken-Numbu earthquake)

-Landslide monitoring-

Installation takes less than 30 min per unit

Ubiquitous Structural Monitoring System

Application examples: -seismic response monitoring-

8

Application examples:

◇US‐Korea‐Japan joint research project (UIUC/KAIST/Univ. of Tokyo) ◇70‐111 multimetric sensor nodes ◇8/2008 to 1/2011

-Bridge vibration monitoring-

What can we obtain from dense structural vibration measurements?How can WSN contribute to bridge engineers?

Various purposes of structural vibration monitoring

• To monitor and control the construction process• To validate the structural designs and characterize

performance (e.g., develop database)• To characterize loads in situ• To assist with building/bridge maintenance• To detect and localize damage before it reaches a

critical level, thus increasing the safety to the public• To reduce the costs and down-time associated with

repair of damage• To assist with emergency response efforts, including

building evacuation and traffic control

(Prof. B. F. Spencer, Jr. UIUC)

ASK-1 ASK-2 ASK-5 ASK-3 ASK-4 ACK-1

ACK-2

ACK-3

ACK-4

ACK-5

AK-3,4

ACK-6

ACK-7

ACK-8

ACK-9

ACK-10

AK-1,2

4P

ACK-11

ACK-12

5P accelerometers (19pts)

How much can we know about the bridge How much can we know about the bridge from measurement by dense arrays of sensors?from measurement by dense arrays of sensors?

Measurement: more than 3 weeks

Sensors: 40 pts (dense array)Sampling: 100Hz

ASK-1 ASK-2 ASK-5 ASK-3 ASK-4 ACK-1ACK-2

ACK-3ACK-4

ACK-5AK-3,4

ACK-6ACK-7

ACK-8ACK-9

ACK-10AK-1,2

4PACK-11

ACK-12

5P0 1 2-0.8

-0.4

0

0.4

0.8

min

galTime history

Spatial view

Structural vibration monitoring: case study

Obtained: up to 23rd modeObtained: up to 23rd mode

Identification resultsModes

ζr fr(Hz)

1st 0.0235 0.1302nd 0.0122 0.1523rd 0.0066 0.221…

10th 0.048 1.3217th 0.031 3.2019th 0.055 3.86

Damping & freq.

-300 0 300-0.4

0

0.4

-300 0 300-0.4

0

0.419-th mode

Distance from tower(m)

Mod

al a

mp.

1st mode

Change in stiffness (KN/m)

Wind velocity (m/s)0 5 10

-80

-40

0

0 5 100

50

100Change in damping (KNs/m)

Wind velocity (m/s)

Result 1: Aerodynamic damping & stiffness

Identified

Ref. state.

Ref. state

Wind tunnel

Aerodynamic

Result 2 Bearing damping & stiffness

Coulomb friction

Coulomb frictionRMS Acceleration(cm/s )

Change in damping (MNs/m)

RMS Acceleration(cm/s )

Change in stiffness (MN/m)

0 1 2

0

4

8

0 1 2-12

-8

-4

0

Bearing