SlideShare una empresa de Scribd logo
1 de 39
Descargar para leer sin conexión
Structural identification of the cable-stayed bridge
of the ANCRiSST SHM benchmark problem
Sapienza University of Rome – StroNGER s.r.l.
S. Arangio, S. Mannucci, F. Bontempi
email: stefania.arangio@uniroma1.it, franco.bontempi@uniroma1.it
New York, June19th 2013
2/38
PartIPartIIConclusionsIntroduction
STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE
OF THE ANCRISST SHM BENCHMARK PROBLEM
S. Arangio, S. Mannucci F. Bontempi
Introduction
Part I
Conclusions
Part II
The ANCRiSST benchmark problem
Traditional ad soft computing approaches for
structural identification ad damage detection
Outline
Processing of monitoring data with
Enhanced Frequency Domain Decomposition and Bayesian neural networks
3/38
PartIPartIIConclusionsIntroduction
STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE
OF THE ANCRISST SHM BENCHMARK PROBLEM
S. Arangio, S. Mannucci F. Bontempi
Introduction
Part I
Conclusions
Part II
The ANCRIiST benchmark problem
Traditional ad soft computing approaches for
structural identification ad damage detection
Outline
Processing of monitoring data with
Enhanced Frequency Domain Decomposition and Bayesian neural networks
4/38
PartIPartIIConclusionsIntroduction
STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE
OF THE ANCRISST SHM BENCHMARK PROBLEM
S. Arangio, S. Mannucci F. Bontempi
Introduction
The ANCRiSST benchmark problem
• Consortium of 20 research institutions
• Established in 2002 with the purpose of:
• assessing current progresses on smart materials and structures technology
• Developing synergies that facilitate joint research projects that cannot easily carried
out by individual centers
In October 2011 they opened for
researchers in the SHM community a
benchmark problem based on a real
bridge: the TianjinYonghe bridge
http://smc.hit.edu.cn/
5/38
PartIPartIIConclusionsIntroduction
STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE
OF THE ANCRISST SHM BENCHMARK PROBLEM
S. Arangio, S. Mannucci F. Bontempi
Description of the TianjinYonghe bridge
Tianjin Hangu
25.15 99.85 260 99.85 25.15
• Cable-stayed bridge
• Opened to traffic since December 1987
• After 19 years of operation damages were detected and the bridge was
retrofitted
• A sophisticated SHM system has been designed and implemented by the
Research Center of Structural Health Monitoring and Control of the Harbin
Institute of Technology
Introduction
6/38
PartIPartIIConclusionsIntroduction
STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE
OF THE ANCRISST SHM BENCHMARK PROBLEM
S. Arangio, S. Mannucci F. Bontempi
Structural Health Monitoring System
Tianjin Hangu
2515 5600 5885 5900 5600 5600 5900 5885 5600 2515
1 (3) 2 (4) 3 (5) 7 (9) 9 (10) 11 (12) 13 (14)
Uniaxial/biaxial accelerometers
Hygrothermograph
Anemometer
1, 3, 5, 7, 9 11, 13 2, 4, 6, 8, 10, 12, 14
During 2008:
• Continuous monitoring system
• 14 uniaxial accelerometers on the bridge deck (downward and upward)
• On the top of the tower: 1 biaxial accelerometer; 1 anemometer; 1 temperature
sensor
downward and upward
Introduction
7/38
PartIPartIIConclusionsIntroduction
STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE
OF THE ANCRISST SHM BENCHMARK PROBLEM
S. Arangio, S. Mannucci F. Bontempi
damaged area
Damage situation 1
Cracks at the closure segment
at both side spans
August 2008:
2 damages are detected
Introduction
8/38
PartIPartIIConclusionsIntroduction
STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE
OF THE ANCRISST SHM BENCHMARK PROBLEM
S. Arangio, S. Mannucci F. Bontempi
Damage situation 2
Damaged piers
Introduction
9/38
PartIPartIIConclusionsIntroduction
STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE
OF THE ANCRISST SHM BENCHMARK PROBLEM
S. Arangio, S. Mannucci F. Bontempi
Available data set
Health condition Damaged condition
• Time histories of the accelerations
recorded at the 14 deck sensors
on January 1st and January 17th 2008
(registration of 1 h for 24 h )
• Environmental information
(wind, temperature)
• Biaxial accelerations at the top of the
tower
• Time histories of the accelerations
recorded at the same 14 deck sensors
on July 30th 2008
(registration of 1 h for 24 h)
• Accelerations collected by field testing
August 7th to 10th 2008
Introduction
10/38
PartIPartIIConclusionsIntroduction
STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE
OF THE ANCRISST SHM BENCHMARK PROBLEM
S. Arangio, S. Mannucci F. Bontempi
Introduction
Part I
Conclusions
Part II
The ANCRIiST benchmark problem
Traditional ad soft computing approaches for
structural identification ad damage detection
Outline
Processing of monitoring data with
Enhanced Frequency Domain Decomposition and Bayesian neural networks
11/38
PartIPartIIConclusionsIntroduction
STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE
OF THE ANCRISST SHM BENCHMARK PROBLEM
S. Arangio, S. Mannucci F. Bontempi
PartI
Methods for structural identification ad damage detection
Input – output
techniques
• The structure has to be artificially excited
and in case of large structures it is not always
possible
• The operation of the structure has to be
interrupted
Only output
Techniques
• The excitation is given by the ambient
vibration
• Measurements in real operational
conditions
Traditional
methods
Soft computing
methods
• Time domain
approaches
• Frequency
domain
approaches
• Neural
networks
• Genetic
algorithms
• Fuzzy Logic
12/38
PartIPartIIConclusionsIntroduction
STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE
OF THE ANCRISST SHM BENCHMARK PROBLEM
S. Arangio, S. Mannucci F. Bontempi
PartI
Enhanced Frequency Domain Decomposition
• Data collection and signal preprocessing
• Construction of the the Power Spectral
Density matrix (PSD)
• Whelch averaged modified periodgram method
• 50 % overlapping and periodic Hamming windowing
• Singular Value Decomposition (SVD) of the PSD
• Identification of modal frequencies and mode shapes
• Evaluation of the damping
13/38
PartIPartIIConclusionsIntroduction
STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE
OF THE ANCRISST SHM BENCHMARK PROBLEM
S. Arangio, S. Mannucci F. Bontempi
PartII
Nonlinear feed-forward basis functions
( ) 







+








+= ∑∑ ==
)2(
0
1
)1(
0
)1(
1
)2(
, k
D
j
jiji
M
j
kjk bbxwgwfy wx
∑=
+=
D
i
jijij bxwa
1
)1(
0
)1(
( )kk afy =
∑=
+=
M
j
kjkjk bzwa
1
)2(
0
)2(
( )jj agz =
NEURAL NETWORK
MODEL
( ) ( )







= ∑=
M
j
jjwfy
1
, xwx φ
output units
hidden units
activations
weights
bias
Neural network model
14/38
PartIPartIIConclusionsIntroduction
STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE
OF THE ANCRISST SHM BENCHMARK PROBLEM
S. Arangio, S. Mannucci F. Bontempi
Traditional learning
t
ij
t
ij
t
ij www ∆+= −
η1
ij
ij
W
E
W
∂
∂
−=∆
η Learning rate
Weights updating
Minimization of a
sum of squares error function
Model fitting is obtained by modifications of the coefficients w
t = correct value
y = network value
( ) 







+








+= ∑∑ ==
)2(
0
1
)1(
0
)1(
1
)2(
, k
D
j
jiji
M
j
kjk bbxwgwfy wx
Gradient descent algorithm [traingd]
Conjugate gradient algorithm [traincg]
Quasi – Newton algorithm [trainbfg]
Levenberg – Marquardt algorithm [trainlm]
( ){ } ∑∑∑
== =
+−=
W
i
i
N
n
oN
t
t
n
t
n wxytE
1
2
1 1
2
2
;
2
1 α
w
15/38
PartIPartIIConclusionsIntroduction
STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE
OF THE ANCRISST SHM BENCHMARK PROBLEM
S. Arangio, S. Mannucci F. Bontempi
Probabilistic interpretation
( ){ } 



−−∝ 2
;
2
exp),,,( ww nn xytMxtp
β
β
1) Probabilistic interpretation
of the network output
2) Probability model
for the prediction error
);( wxyt −=ε
Gaussian µ = 0
σD
2 = 1/β
3) Predictive PDF
The output
approximates the
conditional average of
the target data
hyperparameter
4) Prior PDF
( )






−=
2
2
exp
1
),( w
Z
Mwp
W
α
α
α
Gaussian µ = 0
σw
2 = 1/α
hyperparameter
16/38
PartIPartIIConclusionsIntroduction
STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE
OF THE ANCRISST SHM BENCHMARK PROBLEM
S. Arangio, S. Mannucci F. Bontempi
Network learning as inference
16
=),,( MwDp β
( )
( ){ }








−− ∑∑=
N
n
N
t
n
t
t
n
D
o
xyt
Z 1
2
;
2
exp
1
w
β
β
Likelihood
( ) ( )
( )MDp
MwpMwDp
Dwp
,,
,,,
)M,,,(
βα
αβ
βα =Bayes
theorem
evidence
priorxlikelihood
posterior =
( ) ( ){ } ∑∑∑ ==
+−=
W
i
i
N
n
N
t
n
t
t
n wxytwE
o
1
2
1
2
2
;
2
αβ
w
( ) ( ){ }∑∑=
−=−
N
n
N
t
n
t
t
n
o
xytMwDp
1
2
;
2
,,log w
β
β
( ) ∑=
=−
W
i
iwMwp
1
2
2
,log
α
α
max (posterior) = min (negative log posterior)
=− )M,,,(log βαDwp ( ) ( )=−− MwpMwDp ,log,,log αβ
( )






−=
2
2
exp
1
),( w
Z
Mwp
W
α
α
α
Prior
( )Mwp
( )MDwp ,
( )
( )
( ){ }








−−= ∑∑=
N
n
N
t
n
t
t
n
D
o
xyt
Z
MwDp
1
2
;
2
exp
1
,, w
β
β
β
( )






−=
2
2
exp
1
),( w
Z
Mwp
W
α
α
α
( ){ }∑ ∑ ∑= =
+−
=−
N
n
N
t
W
i
i
n
t
t
n
o
wxyt
Dwp
1 1
2
2
;
2
),,,(log
αβ
βα
w
M
DATA PRE- PROCESSING
OUTPUT
NETWORK ARCHITECTURE
n°INPUT
n°UNIT IN THE HIDDEN LAYERS
POSTERIOR: BAYES’ THEOREM
( ) ( )
( )MDp
MwpMwDp
Dwp
,,
,,,
),,,(
βα
αβ
βα =M
w = wMAP?
yes
INFERENCE OF NEW DATA
DATA POST PROCESSING
PROBABILISTIC MODEL
• NOISE MODEL
• PREDICTIVE PDF
• LIKELIHOOD
• PRIOR
),,,( Mxtp βw
( )MwDp ,, β
),( Mwp α
OPTIMIZATION
(MINIMUM OF )),,,(log MβαDwp−
no
INPUT
ED EW
( )Mwp
( )MDwp ,
1) Model fitting
18/38
PartIPartIIConclusionsIntroduction
STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE
OF THE ANCRISST SHM BENCHMARK PROBLEM
S. Arangio, S. Mannucci F. Bontempi
Bayesian techniques for neural networks
• Level 1 Model fitting: inferring the model parameters given the
model and the data
• Level 2 Optimization of the hyperparameters α and β
• Level 3 Model class selection: optimal model complexity
• Level 4 Automatic relevance determination (ARD):
evaluation of the relative importance of different inputs
Network learning as inference (model fitting) is only one level in
which Bayesian inference can be applied in the neural network
field
Hierarchical multi-level approach
PartI
19
POSTERIOR FOR α, β
TRAINING: OPTIMIZATION
w = wMAP?
?( ) ( )DMEVDMEV ii 1−>
INFERENCE OF NEW DATA
CHOOSE MODEL Mi-1
?
POSTERIOR FOR Mi
α, β = αMP, βMP
DATA PRE- PROCESSING
OUTPUT
NETWORK MODEL Mi
N HIDDEN = i
N INPUT = k
POSTERIOR FOR w
yes
DATA POST PROCESSING
PROBABILISTIC MODEL
no
INPUT
CHOOSE INITIAL α, β
INITIALIZE WEIGHTS w
RE-ESTIMATION OF α, β
yes
no
Wγ ≈
yes
no
i= i+1
is α1,…,αk
‘very large’?
k= k-1
yes
no
( ) ( )
( )MDp
MwpMwDp
Dwp
,,
,,,
),,,(
βα
αβ
βα =M
1st level
Model fitting
20
POSTERIOR FOR α, β
TRAINING: OPTIMIZATION
w = wMAP?
?( ) ( )DMEVDMEV ii 1−>
INFERENCE OF NEW DATA
CHOOSE MODEL Mi-1
?
POSTERIOR FOR Mi
α, β = αMP, βMP
DATA PRE- PROCESSING
OUTPUT
NETWORK MODEL Mi
N HIDDEN = i
N INPUT = k
POSTERIOR FOR w
yes
DATA POST PROCESSING
PROBABILISTIC MODEL
no
INPUT
CHOOSE INITIAL α, β
INITIALIZE WEIGHTS w
RE-ESTIMATION OF α, β
yes
no
Wγ ≈
yes
no
i= i+1
is α1,…,αk
‘very large’?
k= k-1
yes
no
( ) ( )
( )MDp
MwpMwDp
Dwp
,,
,,,
),,,(
βα
αβ
βα =M
1st level
Model fitting
2nd level
Evaluating the hyperparameters α, β
( ) ( )
( )MDp
MpMDp
Dp
βαβα
βα
,,,
),,( =M
21/38
PartIPartIIConclusionsIntroduction
STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE
OF THE ANCRISST SHM BENCHMARK PROBLEM
S. Arangio, S. Mannucci F. Bontempi
Issues in neural network design: selection of the optimal model
RULES OF THUMBS
-…between the input layer size and the output
layer size (Blum, 1992)
- (Software Neuroshell, 2000)
- (Berry and Lynoff, 1997)
- n = dimension needed to capture 70-80% of the
variance
(Boger and Guterman, 1997)
OPTIMAL NUMBER OF UNITS
)(
3
2
oI NNn +=
INn ⋅<2
examplesNn ⋅<
30
1
They aren’t rigorous methods
INPUT
LAYER
OUTPUT
LAYER
HIDDEN
LAYERS
PartI
22
POSTERIOR FOR α, β
TRAINING: OPTIMIZATION
w = wMAP?
?( ) ( )DMEVDMEV ii 1−>
INFERENCE OF NEW DATA
CHOOSE MODEL Mi-1
?
POSTERIOR FOR Mi
α, β = αMP, βMP
DATA PRE- PROCESSING
OUTPUT
NETWORK MODEL Mi
N HIDDEN = i
N INPUT = k
POSTERIOR FOR w
yes
DATA POST PROCESSING
PROBABILISTIC MODEL
no
INPUT
CHOOSE INITIAL α, β
INITIALIZE WEIGHTS w
RE-ESTIMATION OF α, β
yes
n
o
Wγ ≈
yes
no
i= i+1
is α1,…,αk
‘very large’?
k= k-1
yes
no
( ) ( )
( )MDp
MwpMwDp
Dwp
,,
,,,
),,,(
βα
αβ
βα =M
1st level
Model fitting
2nd level
Evaluating the hyperparameters α, β
3rd level
Model class selection
( ) ( )MpMDpDMp ∝)(
prior = constantevidence
( ) ( )
( )MDp
MpMDp
Dp
βαβα
βα
,,,
),,( =M
POSTERIOR FOR α, β
TRAINING: OPTIMIZATION
w = wMAP?
?( ) ( )DMEVDMEV ii 1−>
INFERENCE OF NEW DATA
CHOOSE MODEL Mi-1
?
POSTERIOR FOR Mi
α, β = αMP, βMP
DATA PRE- PROCESSING
OUTPUT
NETWORK MODEL Mi
N HIDDEN = i
N INPUT = k
POSTERIOR FOR w
yes
DATA POST PROCESSING
PROBABILISTIC MODEL
no
INPUT
CHOOSE INITIAL α, β
INITIALIZE WEIGHTS w
RE-ESTIMATION OF α, β
yes
n
o
Wγ ≈
yes
no
i= i+1
is α1,…,αk
‘very large’?
k= k-1
yes
no
( ) ( )
( )MDp
MwpMwDp
Dwp
,,
,,,
),,,(
βα
αβ
βα =M
1st level
Model fitting
2nd level
Evaluating the hyperparameters α, β
3rd level
Model class selection
( ) ( )MpMDpDMp ∝)(
prior = constantevidence
is α1,…,αk
‘very large’?
4th level
Automatic Relevance Determination
( ) ( )
( )MDp
MpMDp
Dp
βαβα
βα
,,,
),,( =M
24/38
PartIPartIIConclusionsIntroduction
STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE
OF THE ANCRISST SHM BENCHMARK PROBLEM
S. Arangio, S. Mannucci F. Bontempi
Introduction
Part I
Conclusions
Part II
The ANCRIiST benchmark problem
Traditional ad soft computing approaches for
structural identification ad damage detection
Outline
Processing of monitoring data with
Enhanced Frequency Domain Decomposition and Bayesian neural networks
25/38
PartIPartIIConclusionsIntroduction
STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE
OF THE ANCRISST SHM BENCHMARK PROBLEM
S. Arangio, S. Mannucci F. Bontempi
0
0,2
0,4
0,6
0 0,5 1 1,5 2 2,5 3
H6
H11
H15
H17
H19
H21
SingularValues(health)
f [Hz]
0
0,1
0,2
0,3
0 0,5 1 1,5 2
AverageSingularValues(health)
f [Hz]
EFDD: Singular Values Decomposition (undamaged)
PartII
26/38
PartIPartIIConclusionsIntroduction
STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE
OF THE ANCRISST SHM BENCHMARK PROBLEM
S. Arangio, S. Mannucci F. Bontempi
0
0,5
1
1,5
2
0 0,5 1 1,5 2
AverageSingularValues(damaged)
f [Hz]
0
0,5
1
1,5
2
0 0,5 1 1,5 2 2,5 3
H6
H9
H12
H15
H18
H20
H22
H23
H24
EFDD: Singular Values Decomposition (damaged)
PartII
27/38
PartIPartIIConclusionsIntroduction
STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE
OF THE ANCRISST SHM BENCHMARK PROBLEM
S. Arangio, S. Mannucci F. Bontempi
Comparison of the mode shapes
PartII
28/38
PartIPartIIConclusionsIntroduction
STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE
OF THE ANCRISST SHM BENCHMARK PROBLEM
S. Arangio, S. Mannucci F. Bontempi
PartII
Procedure for neural network training
time history of the
acceleration recorded at
sensor #
Structural system
Ambient excitation
1+−dtf 2−tf tf1−tf 1+tfTraining of the neural
network model in
undamaged condition
2+tf
Test of the trained neural
network model on a new time
history
29/38
PartIPartIIConclusionsIntroduction
STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE
OF THE ANCRISST SHM BENCHMARK PROBLEM
S. Arangio, S. Mannucci F. Bontempi
PartII
Neural network based damage detection strategy
14 groups of networks have been created
(one for each measurement point e one for each hour of measurements)
14 (points) x24 (hours) = 336 neural network models
Tianjin Hangu
1 (3) 2 (4) 3 (5) 7 (9) 9 (10) 11 (12) 13 (14)
accelerometers
30/38
PartIPartIIConclusionsIntroduction
STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE
OF THE ANCRISST SHM BENCHMARK PROBLEM
S. Arangio, S. Mannucci F. Bontempi
PartII
Detection of anomalies
If ∆e ≈ 0
the structure is considered as undamaged
If ∆e is large an anomaly is detected
31/38
PartIPartIIConclusionsIntroduction
STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE
OF THE ANCRISST SHM BENCHMARK PROBLEM
S. Arangio, S. Mannucci F. Bontempi
PartII
Damaged area
Error in the approximation of the accelerations in the undamaged sections
Training Undamaged
Damage detection
Tianjin Hangu
32/38
PartIPartIIConclusionsIntroduction
STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE
OF THE ANCRISST SHM BENCHMARK PROBLEM
S. Arangio, S. Mannucci F. Bontempi
PartII
Bayesian model class selection
The most plausible class can be obtained applying Bayes’ Theorem:
( ) ( )( , ) |j jj
p M D p D M p M∝M M
prior = costevidence
provided by D
The various model can be compared by evaluating their evidence
The chosen model has 3 hidden units:
N hidden units 1 2 3 4 5
evidence 20756 22603 24922 21944 23240
33/38
PartIPartIIConclusionsIntroduction
STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE
OF THE ANCRISST SHM BENCHMARK PROBLEM
S. Arangio, S. Mannucci F. Bontempi
PartII
BAYESIANMODELSELECTION
34/38
PartIPartIIConclusionsIntroduction
STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE
OF THE ANCRISST SHM BENCHMARK PROBLEM
S. Arangio, S. Mannucci F. Bontempi
PartII
Error in the approximation of the undamaged conditions
downriver
upriver
∆e at the various locations
Data for training: January 1st 2008 (H1 to H24)
Data for testing: January 17th 2008 (H1 to H24)
Undamaged conditions
35/38
PartIPartIIConclusionsIntroduction
STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE
OF THE ANCRISST SHM BENCHMARK PROBLEM
S. Arangio, S. Mannucci F. Bontempi
PartII
Error in the approximation of the damaged conditions
∆e at the various locations
Data for training: January 1st 2008 (H1 to H24)
Data for testing: July 30th 2008 (H1 to H24
Damaged conditions!
36/38
PartIPartIIConclusionsIntroduction
STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE
OF THE ANCRISST SHM BENCHMARK PROBLEM
S. Arangio, S. Mannucci F. Bontempi
PartII
Difference of the errors
The difference of error in the approximation suggests the presence of structural
anomalies around sensor #10
37/38
PartIPartIIConclusionsIntroduction
STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE
OF THE ANCRISST SHM BENCHMARK PROBLEM
S. Arangio, S. Mannucci F. Bontempi
Introduction
Part I
Conclusions
Part II
The ANCRIiST benchmark problem
Description of the bridge and available monitoring data
Outline
Neural network based damage detection strategy
Results
38/38
PartIPartIIConclusionsIntroduction
STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE
OF THE ANCRISST SHM BENCHMARK PROBLEM
S. Arangio, S. Mannucci F. Bontempi
Conclusions
In the Performance based Design framework, monitoring is an
important tool for the verification of the accomplishment of the
expected performance during the entire life cycle
Different approaches for processing monitoring data exist.
A traditional approach (Enhanced Frequency Domain
Decomposition) and a soft computing approach (neural networks)
have been applied on the same data coming from the bridge of the
ANCRiSST SHM benchmark problem and both methods detected
the occurrence of an anomaly.
This work shows that the use of different methods is very important for
the cross validation of the obtained results
The current work is focused on the development of methods for the
localization of the detected damage
Conclusions
POSTERIOR FOR α, β
TRAINING: OPTIMIZATION
w = wMAP?
?( ) ( )DMEVDMEV ii 1−>
INFERENCE OF NEW DATA
CHOOSE MODEL Mi-1
?
POSTERIOR FOR Mi
α, β = αMP, βMP
DATA PRE- PROCESSING
OUTPUT
NETWORK MODEL Mi
N HIDDEN = i
N INPUT = k
POSTERIOR FOR w
yes
DATA POST PROCESSING
PROBABILISTIC MODEL
no
INPUT
CHOOSE INITIAL α, β
INITIALIZE WEIGHTS w
RE-ESTIMATION OF α, β
yes
n
o
Wγ ≈
yes
no
i= i+1
is α1,…,αk
‘very large’?
k= k-1
yes
no
email: stefania.arangio@uniroma1.it
stefania.arangio@stronger2012.com
This research was partially supported by StroNGER s.r.l. from
the fund “FILAS - POR FESR LAZIO 2007/2013 - Support for
the research spin off”.

Más contenido relacionado

Similar a 2013 icossar sa_sm_fb_upload

3 d fnal_gd
3 d fnal_gd3 d fnal_gd
3 d fnal_gd
altaf49
 
GAS@IGARSS2011.ppt
GAS@IGARSS2011.pptGAS@IGARSS2011.ppt
GAS@IGARSS2011.ppt
grssieee
 

Similar a 2013 icossar sa_sm_fb_upload (20)

Neural network-based techniques for the damage identification of bridges: a r...
Neural network-based techniques for the damage identification of bridges: a r...Neural network-based techniques for the damage identification of bridges: a r...
Neural network-based techniques for the damage identification of bridges: a r...
 
Corso di dottorato in Ottimizzazione Strutturale: applicazione mensola strall...
Corso di dottorato in Ottimizzazione Strutturale: applicazione mensola strall...Corso di dottorato in Ottimizzazione Strutturale: applicazione mensola strall...
Corso di dottorato in Ottimizzazione Strutturale: applicazione mensola strall...
 
Corso di dottorato in Ottimizzazione Strutturale: applicazione mensola strall...
Corso di dottorato in Ottimizzazione Strutturale: applicazione mensola strall...Corso di dottorato in Ottimizzazione Strutturale: applicazione mensola strall...
Corso di dottorato in Ottimizzazione Strutturale: applicazione mensola strall...
 
Corso di dottorato in Ottimizzazione Strutturale: Applicazione mensola strall...
Corso di dottorato in Ottimizzazione Strutturale: Applicazione mensola strall...Corso di dottorato in Ottimizzazione Strutturale: Applicazione mensola strall...
Corso di dottorato in Ottimizzazione Strutturale: Applicazione mensola strall...
 
Ldb mapperò di prinzio 01
Ldb mapperò  di prinzio 01Ldb mapperò  di prinzio 01
Ldb mapperò di prinzio 01
 
The optimal tight oil and shale gas development based on pre-existing fractur...
The optimal tight oil and shale gas development based on pre-existing fractur...The optimal tight oil and shale gas development based on pre-existing fractur...
The optimal tight oil and shale gas development based on pre-existing fractur...
 
HAMS weekly update @ 2016.04.29
HAMS weekly update @ 2016.04.29HAMS weekly update @ 2016.04.29
HAMS weekly update @ 2016.04.29
 
10 Important AI Research Papers.pdf
10 Important AI Research Papers.pdf10 Important AI Research Papers.pdf
10 Important AI Research Papers.pdf
 
Image denoising using curvelet transform
Image denoising using curvelet transformImage denoising using curvelet transform
Image denoising using curvelet transform
 
EMCLO PROJECT: EMC DESIGN METHODOLOGY FOR LAYOUT OPTIMIZATION
EMCLO PROJECT: EMC DESIGN METHODOLOGY FOR LAYOUT OPTIMIZATIONEMCLO PROJECT: EMC DESIGN METHODOLOGY FOR LAYOUT OPTIMIZATION
EMCLO PROJECT: EMC DESIGN METHODOLOGY FOR LAYOUT OPTIMIZATION
 
EFFINET - Initial Presentation
EFFINET - Initial PresentationEFFINET - Initial Presentation
EFFINET - Initial Presentation
 
3 d fnal_gd
3 d fnal_gd3 d fnal_gd
3 d fnal_gd
 
Behalf Of Pamela Collaboration
Behalf Of Pamela CollaborationBehalf Of Pamela Collaboration
Behalf Of Pamela Collaboration
 
Model-counting Approaches For Nonlinear Numerical Constraints
Model-counting Approaches For Nonlinear Numerical ConstraintsModel-counting Approaches For Nonlinear Numerical Constraints
Model-counting Approaches For Nonlinear Numerical Constraints
 
Definition and Validation of Scientific Algorithms for the SEOSAT/Ingenio GPP
Definition and Validation of Scientific Algorithms for the SEOSAT/Ingenio GPPDefinition and Validation of Scientific Algorithms for the SEOSAT/Ingenio GPP
Definition and Validation of Scientific Algorithms for the SEOSAT/Ingenio GPP
 
Multivariate dimensionality reduction in cross-correlation analysis
Multivariate dimensionality reduction in cross-correlation analysis Multivariate dimensionality reduction in cross-correlation analysis
Multivariate dimensionality reduction in cross-correlation analysis
 
Image De-noising on Strip Steel Surface Defect Using Improved Compressive Sen...
Image De-noising on Strip Steel Surface Defect Using Improved Compressive Sen...Image De-noising on Strip Steel Surface Defect Using Improved Compressive Sen...
Image De-noising on Strip Steel Surface Defect Using Improved Compressive Sen...
 
Pm8000 series technical datasheet (web).2016.04
Pm8000 series technical datasheet (web).2016.04Pm8000 series technical datasheet (web).2016.04
Pm8000 series technical datasheet (web).2016.04
 
Pres eucome 2016_v3
Pres eucome 2016_v3Pres eucome 2016_v3
Pres eucome 2016_v3
 
GAS@IGARSS2011.ppt
GAS@IGARSS2011.pptGAS@IGARSS2011.ppt
GAS@IGARSS2011.ppt
 

Más de StroNGER2012

L’investigazione antincendio sugli aspetti strutturali: una proposta di codifica
L’investigazione antincendio sugli aspetti strutturali: una proposta di codificaL’investigazione antincendio sugli aspetti strutturali: una proposta di codifica
L’investigazione antincendio sugli aspetti strutturali: una proposta di codifica
StroNGER2012
 
IF CRASC'15 summary
IF CRASC'15 summaryIF CRASC'15 summary
IF CRASC'15 summary
StroNGER2012
 

Más de StroNGER2012 (20)

Corso di dottorato & Corso di formazione
Corso di dottorato & Corso di formazione Corso di dottorato & Corso di formazione
Corso di dottorato & Corso di formazione
 
I Restauri e la Città: l’esempio del Colosseo e della Casa di Augusto
I Restauri e la Città: l’esempio del Colosseo e della Casa di AugustoI Restauri e la Città: l’esempio del Colosseo e della Casa di Augusto
I Restauri e la Città: l’esempio del Colosseo e della Casa di Augusto
 
SISTEMILA RETE STRADALE URBANA:UN’EMERGENZA DEL QUOTIDIANO O UN’OPPORTUNITA’ ...
SISTEMILA RETE STRADALE URBANA:UN’EMERGENZA DEL QUOTIDIANO O UN’OPPORTUNITA’ ...SISTEMILA RETE STRADALE URBANA:UN’EMERGENZA DEL QUOTIDIANO O UN’OPPORTUNITA’ ...
SISTEMILA RETE STRADALE URBANA:UN’EMERGENZA DEL QUOTIDIANO O UN’OPPORTUNITA’ ...
 
INFRASTRUTTURE IN AMBITO URBANO: COMPLESSITA’ DI PROGETTO E DURABILITA’
INFRASTRUTTURE IN AMBITO URBANO: COMPLESSITA’ DI PROGETTO E DURABILITA’INFRASTRUTTURE IN AMBITO URBANO: COMPLESSITA’ DI PROGETTO E DURABILITA’
INFRASTRUTTURE IN AMBITO URBANO: COMPLESSITA’ DI PROGETTO E DURABILITA’
 
61Resilienza dei centri urbani e rilievo delle costruzioni: un binomio indivi...
61Resilienza dei centri urbani e rilievo delle costruzioni: un binomio indivi...61Resilienza dei centri urbani e rilievo delle costruzioni: un binomio indivi...
61Resilienza dei centri urbani e rilievo delle costruzioni: un binomio indivi...
 
Roma e le sue acque:il punto di vista della Protezione Civile
Roma e le sue acque:il punto di vista della Protezione CivileRoma e le sue acque:il punto di vista della Protezione Civile
Roma e le sue acque:il punto di vista della Protezione Civile
 
Una visione ampia dei sistemi: robustezza e resilienza.
Una visione ampia dei sistemi: robustezza e resilienza.Una visione ampia dei sistemi: robustezza e resilienza.
Una visione ampia dei sistemi: robustezza e resilienza.
 
L’investigazione antincendio sugli aspetti strutturali: una proposta di codifica
L’investigazione antincendio sugli aspetti strutturali: una proposta di codificaL’investigazione antincendio sugli aspetti strutturali: una proposta di codifica
L’investigazione antincendio sugli aspetti strutturali: una proposta di codifica
 
Progetto e analisi di ospedali come costruzioni strategiche: visione di siste...
Progetto e analisi di ospedali come costruzioni strategiche: visione di siste...Progetto e analisi di ospedali come costruzioni strategiche: visione di siste...
Progetto e analisi di ospedali come costruzioni strategiche: visione di siste...
 
norme tecniche di prevenzione incendi
norme tecniche di prevenzione incendinorme tecniche di prevenzione incendi
norme tecniche di prevenzione incendi
 
Arangio sapienza-may2015
Arangio sapienza-may2015Arangio sapienza-may2015
Arangio sapienza-may2015
 
Petrini sapienza-may2015
Petrini sapienza-may2015Petrini sapienza-may2015
Petrini sapienza-may2015
 
Erdogmus sapienza-may2015
Erdogmus sapienza-may2015Erdogmus sapienza-may2015
Erdogmus sapienza-may2015
 
Avila sapienza-may2015
Avila sapienza-may2015Avila sapienza-may2015
Avila sapienza-may2015
 
Uso delle fibre di basalto nel risanamento degli edifici storici
Uso delle fibre di basalto nel risanamento degli edifici storiciUso delle fibre di basalto nel risanamento degli edifici storici
Uso delle fibre di basalto nel risanamento degli edifici storici
 
Programma IF CRASC 15
Programma IF CRASC 15Programma IF CRASC 15
Programma IF CRASC 15
 
IDENTIFICAZIONE STRUTTURALE DEL COMPORTAMENTO SPERIMENTALE DI CENTINE INNOVAT...
IDENTIFICAZIONE STRUTTURALE DEL COMPORTAMENTO SPERIMENTALE DI CENTINE INNOVAT...IDENTIFICAZIONE STRUTTURALE DEL COMPORTAMENTO SPERIMENTALE DI CENTINE INNOVAT...
IDENTIFICAZIONE STRUTTURALE DEL COMPORTAMENTO SPERIMENTALE DI CENTINE INNOVAT...
 
IF CRASC'15 summary
IF CRASC'15 summaryIF CRASC'15 summary
IF CRASC'15 summary
 
Corso Ottimizzazione Strutturale Sapienza 2015
Corso Ottimizzazione Strutturale Sapienza 2015Corso Ottimizzazione Strutturale Sapienza 2015
Corso Ottimizzazione Strutturale Sapienza 2015
 
MIGLIORAMENTO ED ADEGUAMENTO SISMICO DI STRUTTURE ESISTENTI ATTRAVERSO L’UTIL...
MIGLIORAMENTO ED ADEGUAMENTO SISMICO DI STRUTTURE ESISTENTI ATTRAVERSO L’UTIL...MIGLIORAMENTO ED ADEGUAMENTO SISMICO DI STRUTTURE ESISTENTI ATTRAVERSO L’UTIL...
MIGLIORAMENTO ED ADEGUAMENTO SISMICO DI STRUTTURE ESISTENTI ATTRAVERSO L’UTIL...
 

Último

Artificial Intelligence: Facts and Myths
Artificial Intelligence: Facts and MythsArtificial Intelligence: Facts and Myths
Artificial Intelligence: Facts and Myths
Joaquim Jorge
 
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
?#DUbAI#??##{{(☎️+971_581248768%)**%*]'#abortion pills for sale in dubai@
 

Último (20)

How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerHow to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected Worker
 
Automating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps ScriptAutomating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps Script
 
Apidays New York 2024 - The value of a flexible API Management solution for O...
Apidays New York 2024 - The value of a flexible API Management solution for O...Apidays New York 2024 - The value of a flexible API Management solution for O...
Apidays New York 2024 - The value of a flexible API Management solution for O...
 
Developing An App To Navigate The Roads of Brazil
Developing An App To Navigate The Roads of BrazilDeveloping An App To Navigate The Roads of Brazil
Developing An App To Navigate The Roads of Brazil
 
Artificial Intelligence: Facts and Myths
Artificial Intelligence: Facts and MythsArtificial Intelligence: Facts and Myths
Artificial Intelligence: Facts and Myths
 
HTML Injection Attacks: Impact and Mitigation Strategies
HTML Injection Attacks: Impact and Mitigation StrategiesHTML Injection Attacks: Impact and Mitigation Strategies
HTML Injection Attacks: Impact and Mitigation Strategies
 
Data Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonData Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt Robison
 
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
 
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
 
2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...
 
Handwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed textsHandwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed texts
 
Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf
 
GenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day PresentationGenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day Presentation
 
A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)
 
How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerHow to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected Worker
 
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, AdobeApidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
 
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
 
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
 
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemkeProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
 

2013 icossar sa_sm_fb_upload

  • 1. Structural identification of the cable-stayed bridge of the ANCRiSST SHM benchmark problem Sapienza University of Rome – StroNGER s.r.l. S. Arangio, S. Mannucci, F. Bontempi email: stefania.arangio@uniroma1.it, franco.bontempi@uniroma1.it New York, June19th 2013
  • 2. 2/38 PartIPartIIConclusionsIntroduction STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE OF THE ANCRISST SHM BENCHMARK PROBLEM S. Arangio, S. Mannucci F. Bontempi Introduction Part I Conclusions Part II The ANCRiSST benchmark problem Traditional ad soft computing approaches for structural identification ad damage detection Outline Processing of monitoring data with Enhanced Frequency Domain Decomposition and Bayesian neural networks
  • 3. 3/38 PartIPartIIConclusionsIntroduction STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE OF THE ANCRISST SHM BENCHMARK PROBLEM S. Arangio, S. Mannucci F. Bontempi Introduction Part I Conclusions Part II The ANCRIiST benchmark problem Traditional ad soft computing approaches for structural identification ad damage detection Outline Processing of monitoring data with Enhanced Frequency Domain Decomposition and Bayesian neural networks
  • 4. 4/38 PartIPartIIConclusionsIntroduction STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE OF THE ANCRISST SHM BENCHMARK PROBLEM S. Arangio, S. Mannucci F. Bontempi Introduction The ANCRiSST benchmark problem • Consortium of 20 research institutions • Established in 2002 with the purpose of: • assessing current progresses on smart materials and structures technology • Developing synergies that facilitate joint research projects that cannot easily carried out by individual centers In October 2011 they opened for researchers in the SHM community a benchmark problem based on a real bridge: the TianjinYonghe bridge http://smc.hit.edu.cn/
  • 5. 5/38 PartIPartIIConclusionsIntroduction STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE OF THE ANCRISST SHM BENCHMARK PROBLEM S. Arangio, S. Mannucci F. Bontempi Description of the TianjinYonghe bridge Tianjin Hangu 25.15 99.85 260 99.85 25.15 • Cable-stayed bridge • Opened to traffic since December 1987 • After 19 years of operation damages were detected and the bridge was retrofitted • A sophisticated SHM system has been designed and implemented by the Research Center of Structural Health Monitoring and Control of the Harbin Institute of Technology Introduction
  • 6. 6/38 PartIPartIIConclusionsIntroduction STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE OF THE ANCRISST SHM BENCHMARK PROBLEM S. Arangio, S. Mannucci F. Bontempi Structural Health Monitoring System Tianjin Hangu 2515 5600 5885 5900 5600 5600 5900 5885 5600 2515 1 (3) 2 (4) 3 (5) 7 (9) 9 (10) 11 (12) 13 (14) Uniaxial/biaxial accelerometers Hygrothermograph Anemometer 1, 3, 5, 7, 9 11, 13 2, 4, 6, 8, 10, 12, 14 During 2008: • Continuous monitoring system • 14 uniaxial accelerometers on the bridge deck (downward and upward) • On the top of the tower: 1 biaxial accelerometer; 1 anemometer; 1 temperature sensor downward and upward Introduction
  • 7. 7/38 PartIPartIIConclusionsIntroduction STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE OF THE ANCRISST SHM BENCHMARK PROBLEM S. Arangio, S. Mannucci F. Bontempi damaged area Damage situation 1 Cracks at the closure segment at both side spans August 2008: 2 damages are detected Introduction
  • 8. 8/38 PartIPartIIConclusionsIntroduction STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE OF THE ANCRISST SHM BENCHMARK PROBLEM S. Arangio, S. Mannucci F. Bontempi Damage situation 2 Damaged piers Introduction
  • 9. 9/38 PartIPartIIConclusionsIntroduction STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE OF THE ANCRISST SHM BENCHMARK PROBLEM S. Arangio, S. Mannucci F. Bontempi Available data set Health condition Damaged condition • Time histories of the accelerations recorded at the 14 deck sensors on January 1st and January 17th 2008 (registration of 1 h for 24 h ) • Environmental information (wind, temperature) • Biaxial accelerations at the top of the tower • Time histories of the accelerations recorded at the same 14 deck sensors on July 30th 2008 (registration of 1 h for 24 h) • Accelerations collected by field testing August 7th to 10th 2008 Introduction
  • 10. 10/38 PartIPartIIConclusionsIntroduction STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE OF THE ANCRISST SHM BENCHMARK PROBLEM S. Arangio, S. Mannucci F. Bontempi Introduction Part I Conclusions Part II The ANCRIiST benchmark problem Traditional ad soft computing approaches for structural identification ad damage detection Outline Processing of monitoring data with Enhanced Frequency Domain Decomposition and Bayesian neural networks
  • 11. 11/38 PartIPartIIConclusionsIntroduction STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE OF THE ANCRISST SHM BENCHMARK PROBLEM S. Arangio, S. Mannucci F. Bontempi PartI Methods for structural identification ad damage detection Input – output techniques • The structure has to be artificially excited and in case of large structures it is not always possible • The operation of the structure has to be interrupted Only output Techniques • The excitation is given by the ambient vibration • Measurements in real operational conditions Traditional methods Soft computing methods • Time domain approaches • Frequency domain approaches • Neural networks • Genetic algorithms • Fuzzy Logic
  • 12. 12/38 PartIPartIIConclusionsIntroduction STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE OF THE ANCRISST SHM BENCHMARK PROBLEM S. Arangio, S. Mannucci F. Bontempi PartI Enhanced Frequency Domain Decomposition • Data collection and signal preprocessing • Construction of the the Power Spectral Density matrix (PSD) • Whelch averaged modified periodgram method • 50 % overlapping and periodic Hamming windowing • Singular Value Decomposition (SVD) of the PSD • Identification of modal frequencies and mode shapes • Evaluation of the damping
  • 13. 13/38 PartIPartIIConclusionsIntroduction STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE OF THE ANCRISST SHM BENCHMARK PROBLEM S. Arangio, S. Mannucci F. Bontempi PartII Nonlinear feed-forward basis functions ( )         +         += ∑∑ == )2( 0 1 )1( 0 )1( 1 )2( , k D j jiji M j kjk bbxwgwfy wx ∑= += D i jijij bxwa 1 )1( 0 )1( ( )kk afy = ∑= += M j kjkjk bzwa 1 )2( 0 )2( ( )jj agz = NEURAL NETWORK MODEL ( ) ( )        = ∑= M j jjwfy 1 , xwx φ output units hidden units activations weights bias Neural network model
  • 14. 14/38 PartIPartIIConclusionsIntroduction STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE OF THE ANCRISST SHM BENCHMARK PROBLEM S. Arangio, S. Mannucci F. Bontempi Traditional learning t ij t ij t ij www ∆+= − η1 ij ij W E W ∂ ∂ −=∆ η Learning rate Weights updating Minimization of a sum of squares error function Model fitting is obtained by modifications of the coefficients w t = correct value y = network value ( )         +         += ∑∑ == )2( 0 1 )1( 0 )1( 1 )2( , k D j jiji M j kjk bbxwgwfy wx Gradient descent algorithm [traingd] Conjugate gradient algorithm [traincg] Quasi – Newton algorithm [trainbfg] Levenberg – Marquardt algorithm [trainlm] ( ){ } ∑∑∑ == = +−= W i i N n oN t t n t n wxytE 1 2 1 1 2 2 ; 2 1 α w
  • 15. 15/38 PartIPartIIConclusionsIntroduction STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE OF THE ANCRISST SHM BENCHMARK PROBLEM S. Arangio, S. Mannucci F. Bontempi Probabilistic interpretation ( ){ }     −−∝ 2 ; 2 exp),,,( ww nn xytMxtp β β 1) Probabilistic interpretation of the network output 2) Probability model for the prediction error );( wxyt −=ε Gaussian µ = 0 σD 2 = 1/β 3) Predictive PDF The output approximates the conditional average of the target data hyperparameter 4) Prior PDF ( )       −= 2 2 exp 1 ),( w Z Mwp W α α α Gaussian µ = 0 σw 2 = 1/α hyperparameter
  • 16. 16/38 PartIPartIIConclusionsIntroduction STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE OF THE ANCRISST SHM BENCHMARK PROBLEM S. Arangio, S. Mannucci F. Bontempi Network learning as inference 16 =),,( MwDp β ( ) ( ){ }         −− ∑∑= N n N t n t t n D o xyt Z 1 2 ; 2 exp 1 w β β Likelihood ( ) ( ) ( )MDp MwpMwDp Dwp ,, ,,, )M,,,( βα αβ βα =Bayes theorem evidence priorxlikelihood posterior = ( ) ( ){ } ∑∑∑ == +−= W i i N n N t n t t n wxytwE o 1 2 1 2 2 ; 2 αβ w ( ) ( ){ }∑∑= −=− N n N t n t t n o xytMwDp 1 2 ; 2 ,,log w β β ( ) ∑= =− W i iwMwp 1 2 2 ,log α α max (posterior) = min (negative log posterior) =− )M,,,(log βαDwp ( ) ( )=−− MwpMwDp ,log,,log αβ ( )       −= 2 2 exp 1 ),( w Z Mwp W α α α Prior ( )Mwp ( )MDwp ,
  • 17. ( ) ( ) ( ){ }         −−= ∑∑= N n N t n t t n D o xyt Z MwDp 1 2 ; 2 exp 1 ,, w β β β ( )       −= 2 2 exp 1 ),( w Z Mwp W α α α ( ){ }∑ ∑ ∑= = +− =− N n N t W i i n t t n o wxyt Dwp 1 1 2 2 ; 2 ),,,(log αβ βα w M DATA PRE- PROCESSING OUTPUT NETWORK ARCHITECTURE n°INPUT n°UNIT IN THE HIDDEN LAYERS POSTERIOR: BAYES’ THEOREM ( ) ( ) ( )MDp MwpMwDp Dwp ,, ,,, ),,,( βα αβ βα =M w = wMAP? yes INFERENCE OF NEW DATA DATA POST PROCESSING PROBABILISTIC MODEL • NOISE MODEL • PREDICTIVE PDF • LIKELIHOOD • PRIOR ),,,( Mxtp βw ( )MwDp ,, β ),( Mwp α OPTIMIZATION (MINIMUM OF )),,,(log MβαDwp− no INPUT ED EW ( )Mwp ( )MDwp , 1) Model fitting
  • 18. 18/38 PartIPartIIConclusionsIntroduction STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE OF THE ANCRISST SHM BENCHMARK PROBLEM S. Arangio, S. Mannucci F. Bontempi Bayesian techniques for neural networks • Level 1 Model fitting: inferring the model parameters given the model and the data • Level 2 Optimization of the hyperparameters α and β • Level 3 Model class selection: optimal model complexity • Level 4 Automatic relevance determination (ARD): evaluation of the relative importance of different inputs Network learning as inference (model fitting) is only one level in which Bayesian inference can be applied in the neural network field Hierarchical multi-level approach PartI
  • 19. 19 POSTERIOR FOR α, β TRAINING: OPTIMIZATION w = wMAP? ?( ) ( )DMEVDMEV ii 1−> INFERENCE OF NEW DATA CHOOSE MODEL Mi-1 ? POSTERIOR FOR Mi α, β = αMP, βMP DATA PRE- PROCESSING OUTPUT NETWORK MODEL Mi N HIDDEN = i N INPUT = k POSTERIOR FOR w yes DATA POST PROCESSING PROBABILISTIC MODEL no INPUT CHOOSE INITIAL α, β INITIALIZE WEIGHTS w RE-ESTIMATION OF α, β yes no Wγ ≈ yes no i= i+1 is α1,…,αk ‘very large’? k= k-1 yes no ( ) ( ) ( )MDp MwpMwDp Dwp ,, ,,, ),,,( βα αβ βα =M 1st level Model fitting
  • 20. 20 POSTERIOR FOR α, β TRAINING: OPTIMIZATION w = wMAP? ?( ) ( )DMEVDMEV ii 1−> INFERENCE OF NEW DATA CHOOSE MODEL Mi-1 ? POSTERIOR FOR Mi α, β = αMP, βMP DATA PRE- PROCESSING OUTPUT NETWORK MODEL Mi N HIDDEN = i N INPUT = k POSTERIOR FOR w yes DATA POST PROCESSING PROBABILISTIC MODEL no INPUT CHOOSE INITIAL α, β INITIALIZE WEIGHTS w RE-ESTIMATION OF α, β yes no Wγ ≈ yes no i= i+1 is α1,…,αk ‘very large’? k= k-1 yes no ( ) ( ) ( )MDp MwpMwDp Dwp ,, ,,, ),,,( βα αβ βα =M 1st level Model fitting 2nd level Evaluating the hyperparameters α, β ( ) ( ) ( )MDp MpMDp Dp βαβα βα ,,, ),,( =M
  • 21. 21/38 PartIPartIIConclusionsIntroduction STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE OF THE ANCRISST SHM BENCHMARK PROBLEM S. Arangio, S. Mannucci F. Bontempi Issues in neural network design: selection of the optimal model RULES OF THUMBS -…between the input layer size and the output layer size (Blum, 1992) - (Software Neuroshell, 2000) - (Berry and Lynoff, 1997) - n = dimension needed to capture 70-80% of the variance (Boger and Guterman, 1997) OPTIMAL NUMBER OF UNITS )( 3 2 oI NNn += INn ⋅<2 examplesNn ⋅< 30 1 They aren’t rigorous methods INPUT LAYER OUTPUT LAYER HIDDEN LAYERS PartI
  • 22. 22 POSTERIOR FOR α, β TRAINING: OPTIMIZATION w = wMAP? ?( ) ( )DMEVDMEV ii 1−> INFERENCE OF NEW DATA CHOOSE MODEL Mi-1 ? POSTERIOR FOR Mi α, β = αMP, βMP DATA PRE- PROCESSING OUTPUT NETWORK MODEL Mi N HIDDEN = i N INPUT = k POSTERIOR FOR w yes DATA POST PROCESSING PROBABILISTIC MODEL no INPUT CHOOSE INITIAL α, β INITIALIZE WEIGHTS w RE-ESTIMATION OF α, β yes n o Wγ ≈ yes no i= i+1 is α1,…,αk ‘very large’? k= k-1 yes no ( ) ( ) ( )MDp MwpMwDp Dwp ,, ,,, ),,,( βα αβ βα =M 1st level Model fitting 2nd level Evaluating the hyperparameters α, β 3rd level Model class selection ( ) ( )MpMDpDMp ∝)( prior = constantevidence ( ) ( ) ( )MDp MpMDp Dp βαβα βα ,,, ),,( =M
  • 23. POSTERIOR FOR α, β TRAINING: OPTIMIZATION w = wMAP? ?( ) ( )DMEVDMEV ii 1−> INFERENCE OF NEW DATA CHOOSE MODEL Mi-1 ? POSTERIOR FOR Mi α, β = αMP, βMP DATA PRE- PROCESSING OUTPUT NETWORK MODEL Mi N HIDDEN = i N INPUT = k POSTERIOR FOR w yes DATA POST PROCESSING PROBABILISTIC MODEL no INPUT CHOOSE INITIAL α, β INITIALIZE WEIGHTS w RE-ESTIMATION OF α, β yes n o Wγ ≈ yes no i= i+1 is α1,…,αk ‘very large’? k= k-1 yes no ( ) ( ) ( )MDp MwpMwDp Dwp ,, ,,, ),,,( βα αβ βα =M 1st level Model fitting 2nd level Evaluating the hyperparameters α, β 3rd level Model class selection ( ) ( )MpMDpDMp ∝)( prior = constantevidence is α1,…,αk ‘very large’? 4th level Automatic Relevance Determination ( ) ( ) ( )MDp MpMDp Dp βαβα βα ,,, ),,( =M
  • 24. 24/38 PartIPartIIConclusionsIntroduction STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE OF THE ANCRISST SHM BENCHMARK PROBLEM S. Arangio, S. Mannucci F. Bontempi Introduction Part I Conclusions Part II The ANCRIiST benchmark problem Traditional ad soft computing approaches for structural identification ad damage detection Outline Processing of monitoring data with Enhanced Frequency Domain Decomposition and Bayesian neural networks
  • 25. 25/38 PartIPartIIConclusionsIntroduction STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE OF THE ANCRISST SHM BENCHMARK PROBLEM S. Arangio, S. Mannucci F. Bontempi 0 0,2 0,4 0,6 0 0,5 1 1,5 2 2,5 3 H6 H11 H15 H17 H19 H21 SingularValues(health) f [Hz] 0 0,1 0,2 0,3 0 0,5 1 1,5 2 AverageSingularValues(health) f [Hz] EFDD: Singular Values Decomposition (undamaged) PartII
  • 26. 26/38 PartIPartIIConclusionsIntroduction STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE OF THE ANCRISST SHM BENCHMARK PROBLEM S. Arangio, S. Mannucci F. Bontempi 0 0,5 1 1,5 2 0 0,5 1 1,5 2 AverageSingularValues(damaged) f [Hz] 0 0,5 1 1,5 2 0 0,5 1 1,5 2 2,5 3 H6 H9 H12 H15 H18 H20 H22 H23 H24 EFDD: Singular Values Decomposition (damaged) PartII
  • 27. 27/38 PartIPartIIConclusionsIntroduction STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE OF THE ANCRISST SHM BENCHMARK PROBLEM S. Arangio, S. Mannucci F. Bontempi Comparison of the mode shapes PartII
  • 28. 28/38 PartIPartIIConclusionsIntroduction STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE OF THE ANCRISST SHM BENCHMARK PROBLEM S. Arangio, S. Mannucci F. Bontempi PartII Procedure for neural network training time history of the acceleration recorded at sensor # Structural system Ambient excitation 1+−dtf 2−tf tf1−tf 1+tfTraining of the neural network model in undamaged condition 2+tf Test of the trained neural network model on a new time history
  • 29. 29/38 PartIPartIIConclusionsIntroduction STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE OF THE ANCRISST SHM BENCHMARK PROBLEM S. Arangio, S. Mannucci F. Bontempi PartII Neural network based damage detection strategy 14 groups of networks have been created (one for each measurement point e one for each hour of measurements) 14 (points) x24 (hours) = 336 neural network models Tianjin Hangu 1 (3) 2 (4) 3 (5) 7 (9) 9 (10) 11 (12) 13 (14) accelerometers
  • 30. 30/38 PartIPartIIConclusionsIntroduction STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE OF THE ANCRISST SHM BENCHMARK PROBLEM S. Arangio, S. Mannucci F. Bontempi PartII Detection of anomalies If ∆e ≈ 0 the structure is considered as undamaged If ∆e is large an anomaly is detected
  • 31. 31/38 PartIPartIIConclusionsIntroduction STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE OF THE ANCRISST SHM BENCHMARK PROBLEM S. Arangio, S. Mannucci F. Bontempi PartII Damaged area Error in the approximation of the accelerations in the undamaged sections Training Undamaged Damage detection Tianjin Hangu
  • 32. 32/38 PartIPartIIConclusionsIntroduction STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE OF THE ANCRISST SHM BENCHMARK PROBLEM S. Arangio, S. Mannucci F. Bontempi PartII Bayesian model class selection The most plausible class can be obtained applying Bayes’ Theorem: ( ) ( )( , ) |j jj p M D p D M p M∝M M prior = costevidence provided by D The various model can be compared by evaluating their evidence The chosen model has 3 hidden units: N hidden units 1 2 3 4 5 evidence 20756 22603 24922 21944 23240
  • 33. 33/38 PartIPartIIConclusionsIntroduction STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE OF THE ANCRISST SHM BENCHMARK PROBLEM S. Arangio, S. Mannucci F. Bontempi PartII BAYESIANMODELSELECTION
  • 34. 34/38 PartIPartIIConclusionsIntroduction STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE OF THE ANCRISST SHM BENCHMARK PROBLEM S. Arangio, S. Mannucci F. Bontempi PartII Error in the approximation of the undamaged conditions downriver upriver ∆e at the various locations Data for training: January 1st 2008 (H1 to H24) Data for testing: January 17th 2008 (H1 to H24) Undamaged conditions
  • 35. 35/38 PartIPartIIConclusionsIntroduction STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE OF THE ANCRISST SHM BENCHMARK PROBLEM S. Arangio, S. Mannucci F. Bontempi PartII Error in the approximation of the damaged conditions ∆e at the various locations Data for training: January 1st 2008 (H1 to H24) Data for testing: July 30th 2008 (H1 to H24 Damaged conditions!
  • 36. 36/38 PartIPartIIConclusionsIntroduction STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE OF THE ANCRISST SHM BENCHMARK PROBLEM S. Arangio, S. Mannucci F. Bontempi PartII Difference of the errors The difference of error in the approximation suggests the presence of structural anomalies around sensor #10
  • 37. 37/38 PartIPartIIConclusionsIntroduction STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE OF THE ANCRISST SHM BENCHMARK PROBLEM S. Arangio, S. Mannucci F. Bontempi Introduction Part I Conclusions Part II The ANCRIiST benchmark problem Description of the bridge and available monitoring data Outline Neural network based damage detection strategy Results
  • 38. 38/38 PartIPartIIConclusionsIntroduction STRUCTURAL IDENTIFICATION OF THE CABLE STAYED BRIDGE OF THE ANCRISST SHM BENCHMARK PROBLEM S. Arangio, S. Mannucci F. Bontempi Conclusions In the Performance based Design framework, monitoring is an important tool for the verification of the accomplishment of the expected performance during the entire life cycle Different approaches for processing monitoring data exist. A traditional approach (Enhanced Frequency Domain Decomposition) and a soft computing approach (neural networks) have been applied on the same data coming from the bridge of the ANCRiSST SHM benchmark problem and both methods detected the occurrence of an anomaly. This work shows that the use of different methods is very important for the cross validation of the obtained results The current work is focused on the development of methods for the localization of the detected damage Conclusions
  • 39. POSTERIOR FOR α, β TRAINING: OPTIMIZATION w = wMAP? ?( ) ( )DMEVDMEV ii 1−> INFERENCE OF NEW DATA CHOOSE MODEL Mi-1 ? POSTERIOR FOR Mi α, β = αMP, βMP DATA PRE- PROCESSING OUTPUT NETWORK MODEL Mi N HIDDEN = i N INPUT = k POSTERIOR FOR w yes DATA POST PROCESSING PROBABILISTIC MODEL no INPUT CHOOSE INITIAL α, β INITIALIZE WEIGHTS w RE-ESTIMATION OF α, β yes n o Wγ ≈ yes no i= i+1 is α1,…,αk ‘very large’? k= k-1 yes no email: stefania.arangio@uniroma1.it stefania.arangio@stronger2012.com This research was partially supported by StroNGER s.r.l. from the fund “FILAS - POR FESR LAZIO 2007/2013 - Support for the research spin off”.