Managing Civil Infrastructure on the Basis of Structural ......1 Managing Civil Infrastructure on...

1

Managing Civil Infrastructure on the Basis of Structural Health Monitoring

Christian Boller, Jochen Kurz and Gerd DobmannFraunhofer Institute of Non-Destructive Testing Methodology (IZFP)

Universität des Saarlandes, Saarbrücken/Germany

© C. Boller, Fraunhofer IZFP

1

51st Brazilian Concrete CongressCuritiba/Brazil, October 6-10, 2009

Built forever!?


2

2

B52 – Built to fly 100 years?


3

Design Principles for Structures100%

50%0%Stress Without inspection of the component

(no corrosion effects considered)

Safe Life

SafeLife

Log cycles

StressInspectable

crack possible

Damage tolerance

Fracture With inspection of the component(no corrosion effects considered)


4

Log cycles

Short cracks Slow crack propagation

Fail safe

Endurance limit SafeLife

3

Estimation of damage accumulation

Time

ltage

A

B D

I. Load signal

Time

ain

A

B Dmax

II. Countingload cycles

IV. Fatigueevaluation

Vol

C

Stra

C

min


5

III. Rainflow matrix

)Strength Residual( fN =Δ

Res

idua

l stre

ngth

Critical crack length

Detectable crack length

Cra

ck le

ngth

ΔN

Δa


6

Crack lengthNumber of cycles or time

4

Steps to be considered for crack propagation calculation

1. Initial crack length a02 Y factor

For example:2. Y factor3. Load (stress)4. K value5. Crack propagation

equation (i.e. Forman)6. Crack extension ∆a

waY πsec=

aYKI πσ=


7

7. a = a0+∆a & back to step 1 with a0 = a ( ) KKR

KCdNda

c

m

F

F

Δ−−Δ

=1

Influence of Stringers on Fracture


8

Source: A. Grant: Fundamentals of Structural Integrity

5

Fatigue Design

Airframe Fatigue Design Principles

Safe Life Design Damage Tolerant Design

Fail SafeSlow Crack Propagation

How muchpotential left?


9

Crack StopperMultiple Load Path

Ageing bridge infrastructure in Germany

Autobahn

National

Ageing bridge infrastructure in Germany over the last centuryBridges and highways

Age of structure relative to the number of constructions [%] (Date: 31.12.2005, source BASt)

15.8BABB Str World

changed:

• More traffic

• Environment

• etc.

National Highways

5.4

2.5

5.9

11.912.3

8.4

7.17.5

11.8

9.7

2.3 2.52.6

4.8

7.3

9.6 9.6

13.5

9.6

8.88.4 8.5 8.4

B-Str


10

BUT

2008 Civil infrastructure remained

0.0 0.0 0.0 0.1 0.20.5

0.1 0.40.70.8

0.40.7 0.8

0.3 0.6 0.6

bis1899

1900 -1909

1910 -1919

1920 -1929

1930 -1934

1935 -1939

1940 -1944

1945 -1949

1950 -1954

1955 -1959

1960 -1964

1965 -1969

1970 -1974

1975 -1979

1980 -1984

1985 -1989

1990 -1994

1995 -1999

2000 -2004

2005 -2009

year of establishment

6

Civil infrastructure - An issue for monitoring• Bridges

• Pipelines and processing industry

• Power supply (e. g. supply lines, wind energy

Example: bridge area in 106 m2 in Germany sorted by type of construction:

plants)

• Dams

• Buildings of the early ages of reinforced concrete

• Tunnels

• Waste refilling

• Cultural heritage0,16

5,14

1,381,840,02SteelComposite


11

Date 31.12.2005, source BASt

g

• Others19,39

StoneConcretePrestressed ConcreteWood

Options and challenges for life cycle management

Options: Challenges:

• Sensors and miniaturisation• Robotics

• Enhanced computation power

• Internet

• Wireless data communication

• CAE simulation

P ti

• New materials• Higher loads

• Global warming

• Criminal threats and security policies

• Regulations

• Possibly others


12

• Prognostics

• Risk assessment

• Life cycle cost analysis

7

BetoScan: self navigating mobile robot

Batteries

BetoScan

Sensors

Robot with 2 computer systems


13

Air temperature and humidity

Concrete humidity (surface)

Concrete humidity (volume)

Concrete coverage Geometry information

Reinforcement information Corrosion risk

Hytelog Hf MOIST RP, microwaveHf MOIST PP,

microwaveProfometer, eddy

currentAcsys A1220,

ultrasound Mala ProEx, radarCanin, potential

(resistivity)

BetoScan: multi-sensor data acquisition

BetoScan


14

8

OOnnSSite ite SCASCAnnenneRR

OSSCAR


15

www.vdivde-it.de/innonet


OSSCAR

Reinforcement

Data cube also called OLAP (Online Analytical Processing) cube data format


16

information Geometry information Concrete coverage

Mala ProEx, radar Acsys A1220, ultrasound Profometer, eddy current Generalized data concept

9


OSSCAR

Data acquisition and


17

Data acquisition and motion control

Data analysis: comparison, fusion, joint-inversion

Data fusion:

Outlook data analysis

Test specimen before placing of concrete

Fused image of NDT data measured at specimen on the left

Source: BAM Berlin

combination of multiple source data to achieve inferences

Joint inversion or multi-


18

objective optimization: finding model that explains several data sets at once Data cube Function container

10

Visualisation of inner parts of constructionsUltrasound measurements on a concrete deck


19

Source: Alexander Taffe, BAM Berlin

Diagnosis of prestressed tendons

Magnetic rotating head scanner

Rotating speed: 2HzLinear speed: 720m/hRotor diameter: 3.5m

electromagnet

magnetic sensors

sensor traces

Application: prestressed steel rupturedetection in concrete plates rotating head

scanner

measurement results of a concrete bridge


20

prestressing tendons

ruptures

……

Source: Klaus Szielasko and Albert Kloster, IZFP Saarbrücken

11

Crack and flaw detection in pipelines

Inspection of pipelines using pigs

LDefect contour Simplification

Crack and flaw detection by ultrasound pigs

Correlation of data from mearuemnts and geometry models of cracks for assessment


21

dt

L• Up to 300 km of pipelines have to be inspected in one run

• Signals from up to 1000 sensors have to analysed for each segment

Source: W. Bähr IZFP & O.A. Barbian NDT Systems & Services

Barkhausen noise and Eddy current MIcroscope (BEMI)

3D Manipulator Control PCMiniaturised inductive sensor(modified video recorder head)

Stationary Electromagnet(for Barkhausen Mode)

3D Manipulator Control PC (modified video recorder head)


22

Resolution: ~ 10 μm• Manipulator control• Eddy current hardware• Barkhausen noise hardware

Records Barkhausen noise

Source: M. Rabung, U. Rabe, S. Hirsekorn, I. Altpeter, Fraunhofer IZFP

12

HCM ~ HC HCµ ~ HC

Monitoring Technique

Flux Density B

Barkhausen Noise

CM C Cµ C

Higher Mode AnalysisMulti Frequency Eddy

Current Impedance

Superimposed Permeability

Magnetisation Force H


23

Source: K. Szielasko, R. Tschuncky, I. Altpeter, Fraunhofer IZFP

Quantitative NDT / Calibration

Data Base(NDT Recorded Data + Reference Data)

Computation

(Approximation Function, Similarities, …)

Residual Austenite

(Approximation Function, Similarities, …)Hardness

(Approximation Function, Similarities, …)

Residual Stresses

p(Regression Analysis, Pattern Recognition, …)

QNDT


24

QNDTHardness

QNDTResidualAustenite

QNDTStress


13

3MA Test System

Control Unit:- Notebook- Industrial PC- Server

3MA-Frontend- TCP/IP interface- Network link- ExtendableExtendable

Modular Hardware

- Components linked via aTCP/IP interface

- Expandable throughadditional components(i.e. magnetostriction, US)

- Easily upgradable withnew components


25

3MA-Sensor- Available in different sizes and shapes

new components


MikroMach Test System

Mikromagnetic Materials charakterisationMaterials charakterisation in its most compact form

Sensor = Test Equipment

Link control PC via


26

Standardised USB link

Source: K. Szielasko, I. Altpeter, Fraunhofer IZFP

14

Quantitative Determination of Residual Stresses

∆α = 14.04°

∆α = 22.26°∆α =

0.95°

Stamp Edge

0

300

400

500

600M

Pa] (

3MA

) A001A003A002A004B001B002

A001 A003 ∆α = 14.04°

A002 A004

∆α = 22 26°

-300

-200

-100

0

100

200

0 1 2 3 4 5 6 7 8

Res

idua

l Str

ess σ

[M

B002X-Ray Ref. A001X-Ray Ref. A002X-Ray Ref. B001


27

04

8

22.26°

B001 B002

Position [cm]

Source: B. Wolter, K. Szielasko, I. Altpeter, Fraunhofer IZFP

17 CrNiMo 6 residual stress maeasurement in 400

600

800

Oberfläche

QNDT of Residual Stresses due to Machining

maeasurement in gear wheels

-400

-200

0

200

3MA

-Wer

t


28

-600-600 -400 -200 0 200 400 600 800

Eigenspannung [M Pa] (X-ray)


15

Results:

Calibaration Probes:

Residual Austenite Determination on Anchor Bolts using 3MA

44 Anchor bolts with a RA of 2,6% 43 Anchor bolts with a RA of 2,6%44 Anchor bolts with a RA of 4,6% 43 Anchor bolts with a RA of 11,3%15 Anchor bolts with a RA of 5,8%

Validation Probes:

Charge 1 with a RA of 3,4%


29

Charge 1 with a RA of 3,4% Charge 2 with a RA of 4,8% Charge 3 with a RA of 4,9% Charge 4 with a RA of 4,5% 25 anchor bolts each from different hardening charges


400

600

800

1000

1200

1400

1600

m [M

Pa] (

3MA

)

Ultimate Strength

Inspection of Sheet Steel using the 3MA System

600

800

1000

1200

1400

1600

.2 [M

Pa] (

3MA

)

0

200

400

0 200 400 600 800 1000 1200 1400 1600

Rm [MPa] (zerstörend)

R m

Yield Strength


30

0

200

400

0 200 400 600 800 1000 1200 1400 1600

Rp0.2 bzw. REH [MPa] (zerst.)

R p0.

Trolley with integrated 3MA Systemfor inspection of milled sheets during rolling

Unit for remote control of trolley with 3MA system


16

The Central Question

Without compromising safety, could we make our structures:– Better available?– Lighter weight?– More cost efficient?– More reliable?

by making sensors (and possibly also actuators) t b i t l t f th t t ?


31

to become an integral part of the structure?

What about …• Looking at advanced cheap sensors, which are

continuously becoming– Smaller,

Li ht– Lighter,– Cheaper?

• Making the sensors an integral part of the structural component?

• Combining the sensors through advanced microelectronics and possibly– Wireless technology with

Advanced microprocessors and


32

– Advanced microprocessors and – Advanced signal processing?

17

Structural Health Monitoring (SHM)

… is the integration of sensing and possibly also actuation devices to allow the loading and damaging conditions of a structure to be g grecorded, analysed, localised and predicted in a way that non-destructive testing becomes an integral part of the structure. SHM is not an additional NDT application, nor QNDT, a fatigue analysis, vibration analysis, signal processing, material science or any other single technological area but rather lateral integration


33

technological area but rather lateral integration of all those in terms of a structure’s life cycle management.

What We Need to Know

• Structural Behaviour and Performance• Loads• Design Principles• Maintenance• Systems for Structural Assessment• Emerging Technologies


34

e g g ec o og es

18

Enhancing Damage Tolerance through SHM

Stress

Gain in weight

Stress

Gain in weight


35

Life

withmonitoring

withoutmonitoring

Gain in life

Life

withmonitoring

withoutmonitoring

Gain in life

Source: H.-J. Schmidt, Airbus

Technologies Beyond State-of-the-Art in Aircraft NDT

WavelengthWavelength

Inte

nsity

Inte

nsity Incident Light Transmission

Index Modulation WavelengthWavelength

Inte

nsity

Inte

nsity Incident Light Transmission

Index Modulation

Smart Layer:Acousto-Ultrasonics

Optical Fibre Bragg Grating Sensor:• Strain• Pressure

Period Λ

Inte

nsity

λ Bragg

Wavelength

Reflection

Period Λ

Inte

nsity

λ Bragg

Wavelength

Reflection

• Pressure• Temperature

Acoustic Emission

Phased Array

Electro

MEMS


36

ComparativeVacuumMonitoring:• Crack

Electro MagneticLayer

19

What is a Sensor?

Amplifier DataA i iti

Switch Box

p Acquisition

Processor

Amplifier WaveformGenerator

Smart Suitcase™Smart Layer®

with 12 transducers


37

Smart Suitcasewith 12 transducers

Smart Sensing System with 1 input and output

Stress measurements of a suspension bridge

• Six strands on center-east and anchorage chamber measured

Fibre optic sensors for loads monitoringCable opened for tests

• SOFO Dynamic fiber optic sensing system used• Two 12-wheels trucks, ~30 tons each• Quasi-static test: 7 km/h; dynamic test: 72 km/h

Anchor block tested


38

Source: Smartec SA, Lugano Switzerland

20

Remote dynamic loads monitoring

SSI implemented in Labview VI operates in real-time and confirms operation of TMD during strong winds via eight-times increase in damping

25

30

10-2

100


39

Source: James Brownjohn, The University of Sheffield0 0.5 1 1.5 2 2.5 3 3.5 40

5

10

15

20

Frequency (Hz)

Ord

er

0 0.5 1 1.5 2 2.5 3 3.5 410-10

10-8

10-6

10-4

Structural health monitoring using wireless sensor networks

sensors- robust - low power consumption- high bandwidth - cheap

wireless data transmission( i l

self configuring netzwork

embedded networkingautomatic data

MEMS / MotesMEMS / Motes

Insitu-Processing

wireless RF data

transmission Daten

transmitter and receiver (internet)


40

Source: Christian Grosse MPA University of Stuttgart

(wireless communication)

netzwork extractionself calibrating

transmission of a few but meaningful data

extrahierten Daten

transmittermobile data collector

server

Analysis / AlarmAnalysis / Alarm

21

Acoustic emission sensors

Specimen Sensor Attachment

sensorsPhased arraytransducers

Smart Layer transducers


41

Acoustic emission sensors

3

4

1to22 (Path 1) 3to23 (Path 2) 2to24 (Path 3) 4to25 (Path 4) 5to26 (Path 5) 6to27 (Path 6) 7to28 (Path 7) 10to17 (Path 8)

Damage Index of the 1st wave packet in parallel wave path

Normalized Damage Index = Measured Damage Index / 1st Damage Index

Channel numberChannel numberThreshold

Data Processing Results

1

2

3

Dam

age

Inde

x

9to16 (Path 9) 12to19 (Path 10) 11to18 (Path 11) 14to21 (Path 12) 13to20 (Path 13)

132456710912111413

20 21 18 19 16 17 28 27 26 25 24 23 22


42

3000

040

000

5000

060

000

7000

080

000

9000

0

1000

00

1100

00

1200

00

1300

00

1400

00

1500

00

1600

00

1700

00

1800

00

0

No. of cycle

AU sensor - SMART Layer

Cracked area

22

Argumentation of Threshold Setting

Damage Index against the estimation of the crack length from the parallel paths at 183kcycle

1 0

1.5

2.0

2.5

3.0

3.5

4.0

Dam

age

Inde

x

Damage Index against Crack length


43

0.0

0.5

1.0

0 2 4 6 8 10 12 14 16crack length [in mm]

Constraints:

Transducer Locations

Crack Location

Crack Initiation at Countersunk Rivet Holes

Constraints:• Transducers only to be

placed inside the fuselage• Acoustic signals passing a

lap can reduce up to a factor of 4

• Area around riveted lap allows for a multitude of


44

acoustic wave reflections

OuterFuselage

InnerFuselage

23

Wave Modulation at Tapered Surfaces

Crack

Crack

Cra ck Initia tion


45

Am

plitu

de (a

rbitr

ary)

I: Incident Wae, R: Reflected Wave

UX,topUX,bottomUY,topUYb tt

(S0)I (S0)Rθ=90o

Am

plitu

de

I:Incident Wave, R: Reflected Wave, C: Converted Wave

UX,topUX,bottomUY,topUY bottom

(S0)R(S0)I (A0)CTRθ=45o

Mode Conversion at Tapered Surfaces

0 20 40 60 80 100 120Time (μsec)

UY,bottom

0 20 40 60 80 100 120-0 4

-0.2

0

0.2

0.4

0.6

Am

plitu

de,

ε xx

S0(Incidend)

S0(Reflected)

0 20 40 60 80 100 120

Am

plitu

de,

ε xx

S0(Incident)

TR A0(Converted)S0

(Reflected)

0 20 40 60 80 100 120Time (μsec)

Y,bottom


46

0 20 40 60 80 100 1200.4Time (μsec)

0 20 40 60 80 100 120Time (μsec)

Time (μsec)

frequ

ency

(MH

z)

0 20 40 60 80 100 1200

0.4

1 IncidentS0

ReflectedS0

Transient C/RConverted

A0

Time (μsec)

frequ

ency

(MH

z)

0 20 40 60 80 100 1200

0.4

1 IncidentS0

ReflectedS0

24

X

Y (mm)

015

0 15 30 45 60

X

Y (mm)

015

0 15 30 45 60

Effect of Holes on Scattered Wave Distribution

X

Y (mm)

015

0 15 30 45 60

X

Y (mm)

015

0 15 30 45 60

(mm

)5

30(m

m)

530

0.9

1

1.1

0.9

1

1.1


47

(mm

)30

45(m

m)

3045

0.5

0.6

0.7

0.8

0.5

0.6

0.7

0.8

Meshing for Cracks around Holes

Without Crack With Crack


48

Without Crack With Crack

25

400

450

0.9

1

++++++

+++++++

+ 400

450

0.6

0.7

0.8

400

450

0.6

0.7

0.8

400

450

0 6

0.7

0.8

0.9

Effect of Cracks on Scattered Wave Differentials

X (mm)

Y (m

m)

0 50 100250

300

350

0.3

0.4

0.5

0.6

0.7

0.8

++ + + + +

++ + + + ++

+

X (mm)

Y (m

m)

0 50 100250

300

350

-0.2

-0.1

0

0.1

0.2

0.3

0.4

0.5

X (mm)

Y (m

m)

0 50 100250

300

350

-0.1

0

0.1

0.2

0.3

0.4

0.5

X (mm)

Y (m

m)

0 50 100250

300

350

-0.1

0

0.1

0.2

0.3

0.4

0.5

0.6

Reference 10 mm Crack 15 mm Crack 20 mm Crack


49

Crack length 2.5 x rivet diameter needs to be monitored no further away than 12.5 x rivet diameter

Crack 10.91

0.9

1

Multiple Path Monitoring for Confidence Enhancement

1

3

24

5A

6

8

79

10

B

20 10 20 15 15 20

PZT Transducer

Through Hole

30 3020 10 20 15 15 20

PZT Transducer

Through Hole

30 30

C ac

92

1 64

7

0 2

0.3

0.4

0.5

0.6

0.7

0.8

Nor

mal

ised

Am

plitu

de

9 46 27 10 2

0.3

0.4

0.5

0.6

0.7

0.8

Nor

mal

ised

Am

plitu

de

9 46 27 1


50

0 5 10 150.1

0.2

Crack 1 Length

7 1

0 5 10 150.1

0.2

Crack 1 Length

7 1

26

500

Loading Condition Influence on Damage Monitoring

InitialFatigue + Under static Load Fatigue + Unload

Ampl

itude

(mV)

100

200

300

400 Fatigue + Under static Load Fatigue + Unload

1

3

24

5A

6

8

79

10

B

300 30020 10 20 15 15 20

PZT Transducer

Through Hole

30 3020 10 20 15 15 20

PZT Transducer

Through Hole

30 30

Cracks

A 2

1

3

4

5


51

2 3 4 50Path Number

A 12 3 4 50Path Number

A

high frequency

• Acoustic Emission (passive)

A t Ult i ( ti )

Monitoring of Windmill Rotor Blades

• Acousto Ultrasonic (active)

Sensor/actor elements

• piezo fibre patches (HF) as sensors

(and low range actors)

36 HF piezo fibre transducer


52

( g )

• pzt stacks for lamb wave generation

(large range actuators)

6 HF pzt stacksSource: B. Frankenstein et al., Fraunhofer IZFP

27

• structure independent application for SHM

SHM Systems

• online reconfiguration

• online and offline analysis, documentation of structural damage

• realization of structure dependent data base

• development and test of online SHM algorithms


53

Source: B. Frankenstein et al., Fraunhofer IZFP

SHM in Publication• Internat., European and Asia-Pacific

Workshop on SHM• Encyclopedia of SHM; John Wiley &

Sons, 5. Vols., ca. 3,000 pages, 01/2009

• Structural Health Monitoring – An International Journal; SAGE Publ.

• Int. Journal on Structural Control and Monitoring; John Wiley & Sons

• Smart Materials and Structures; Inst. of Physics Publ.

• Intelligent Material Systems and Structures; SAGE Publ.

• Staszewski W.J., C. Boller and G.R. Tomlinson (Ed.s), 2003: Health Monitoring of Aircraft Structures; J

• Introduction and Definitions • Physical Monitoring Principles • Signal Processing • Simulation


54

Monitoring of Aircraft Structures; J. Wiley & Sons

• Balageas D., C.-P. Fritzen and A. Güemes, 2005: Structural Health Monitoring; ISTE Ltd.

• …

• Simulation • Sensors• Systems and System Design• Principles of SHM-Based Structural

Monitoring, Design and Maintenance• Aerospace Applications• Civil Engineering Applications • Other Applications• Standardization and Specification• Glossary

28

• Ageing structures must have a damage tolerance if they should be managed

• Structural design principles and geometry must be understood• Operational loads need to be known

Di it l d l f th t t i h l f l f d l ti

Conclusions

• Digital model of the structure is helpful for damage accumulation evaluation

• NDT methods available for in situ structural and materials characterisation

• Different sensing system principles to be ready for structural integration

• NDT phenomena around the areas to be monitored have to be known• SHM technologies considered provide sufficient maturity and

reliability


55

reliability• A lot of technology around for SHM based ageing infrastructure

management

Managing Civil Infrastructure on the Basis of Structural ......1 Managing Civil Infrastructure on...

Documents

Transcript of Managing Civil Infrastructure on the Basis of Structural ......1 Managing Civil Infrastructure on...