INTELLIGENT STRAIN MONITORING OF THE MAIN STEEL BOX GIRDER OF THE JIANGYIN YANGTZSE RIVER BRIDGE USING THE FIBER BRAGG GRATING TECHNOLOGY

dearmeltedUrban and Civil

Nov 25, 2013 (3 years and 9 months ago)

87 views

- 353 -

INTELLIGENT STRAIN MONITORING OF THE MAIN
STEEL BOX GIRDER OF THE JIANGYIN YANGTZSE
RIVER BRIDGE USING THE FIBER BRAGG GRATING
TECHNOLOGY
Yehua Fan* & Xiongfei Chen **
* Ph.D., Senior Engineer, Management Center of The Jiangyin Yangtze River Bridge,
Jiangsu Yangtze River Bridge Company Ltd,69Shigu St., Jiangsu Transportation House,
Nanjing 210004, P.R. China,fanyehua@163.com
** Senior Engineer, Management Center of The Jiangyin Yangtze River Bridge, Jiangsu
Yangtze River Bridge Company Ltd,69Shigu St., Jiangsu Transportation House, Nanjing
210004, P.R. China, cxfy@vip.sina.com

Key words: fiber Bragg grating (FBG), Jiangyin Bridge, intelligent monitoring, structural
strains, health and safety assessment.
Abstract: The Jiangyin Bridge has been experiencing the heavy traffic since it was opened
to traffic in 1999. The main steel box deck, which is directly subjected to traffic loads, is
prone to fatigue damage. One kind of the measurable responses of the deck girder is the
internal strains of the structure, by which means the dynamic structural responses can be
acquired. In this paper, the fiber Bragg grating technique has been used to monitor the
strains of the critical regions of the main girder. The stress and strains of the critical
sections of the main bridge and of the links are examined. The safety assessment of the
main girder has been carried out based on the measured strain data of the main structure.



- 354 -

Fan & Chen: INTELLIGENT STRAIN MONITORING OF THE MAIN STEEL BOX GIRDER OF THE JIANGYIN YANGTZSE RIVER BRIDGE...
1. INTRODUCTION
Because of the excessive loads, fatigue effects, external shocks and impacts, and the
degradations of the materials during service, a bridge structure might be disabled earlier
than its designed service life. When accumulated to a certain degree, above deteriorations
will cause a sudden accident and thus it is necessary to implement health monitoring and
evaluation in order to ensure the safety, applicability and durability of the bridge structure

[1, 2]
. Naturally, the strain is one of the most important performance indices which reflect the
health status of a bridge. Intelligent strain monitoring is helpful for evaluating the quality
and health status of the bridge and can be used as the basis for the assessment of the
remaining life. The fiber Bragg grating (FBG) sensor is a new-type sensor developed in the
1970’s. Small changes in the external physical quantity can be detected by the movement of
the grid-reflected wavelength. The FBG technology not only has the merits of the high
sensitivity, corrosion resistance, electromagnetic interference and large frequency
bandwidth, but also processes a high degree of linearity and repeatability. This technology
can be used to accurately measure the structural stress and strains with absolute and quasi-
distributed digital modes. Therefore, FBG sensors have been widely used in structural
health monitoring and intelligent assessment
[3-6]
.
2. INTELLIGENT STRAIN MONITORING USING THE FBG
The purpose of intelligent structural monitoring is to provide the basis for structural
damage detection, fatigue assessment and structural health state assessment of a bridge. By
collecting the strain data at critical positions of the bride and analyzing the distributions of
the stress and strains under various loads, such purpose can be achieved. Damage can be
detected by the variations of the stress and strains. In the stress-strain testing, traditional
monitoring sensors are referred to the resistance strain gauges and strain gauge steel strings
playing an important role in the testing. However, such sensors have some drawbacks, for
example, they are not suitable for the working conditions such as electromagnetic
interference, corrosion and moisture. The monitoring system using traditional sensors
cannot satisfy the requirement in the stability, durability and distribution of an engineering
structure, resulting in the lack of reliable experimental data for the bridge safety
assessment. However, compared with the traditional sensors, the FBG sensors has several
advantages such as high sensitivity, corrosion resistance, electrical insulation, waterproof,
anti-electromagnetic interference, optical flexible paths, easy connections to computer
modules, remote sensing. Moreover, their simple structural forms, small sizes, light weights
and wide frequency bandwidths can be used to the distributed measurements of
temperature, strains, pressure and other parameters. Therefore, the FBG sensors are
preferable in health monitoring of civil structures, particularly bridge structures.
Considering the necessity of long-term monitoring and signal transmission, the Jiangyin
Bridge structural health intelligent monitoring system adopted the FBG sensors for the
strain monitoring of the main beam.
2.1 Arrangement of strain monitoring points
Because the strain monitoring sensors were mounted on some critical sections of the bridge,
the determination of their locations is very important. The structural strain monitoring

- 355 -

Fan & Chen: INTELLIGENT STRAIN MONITORING OF THE MAIN STEEL BOX GIRDER OF THE JIANGYIN YANGTZSE RIVER BRIDGE...
sections were optimized by means of genetic algorithms. Since the outer strains of the
girders and piers are proportional to the corresponding curvatures, the optimal sensor
locations were determined using the curvature modes and the fitness has been adopted
based on the deformation bending energy
[7]
.
 
n
rj
n
ri
r
r
n
j
n
i
s
IEf 



11
(1)
Where E and I represent the elasticity modulus and moment of inertia of the beam (column),
respectively;
n
rj

is the component r with respect to the curvature mode j obtained by the
second difference of the displacement mode;

isth攠coll散ei潮o渭m敡獵牥e⁰潩nt猠
慮搠
r
denotes the component r restricted to all non-measured points. And
smaller
fitness value
s
f
is always desirable
.

The genetic algorithm used for the determination of the stress monitoring sections could be
used to basically determine the monitoring points. But the large and repeated stress sections
were also taken into account in the final layout of the monitoring points. The optimized
strain monitoring layout of the main girder is shown in Figure 1. Monitoring points are
located at the eight (9 in total) sections of the main girder in the longitudinal direction.
Each section had eight FBG strain sensors, while another four FBG temperature sensors
were used for the temperature compensation and the testing of the temperature field inside
the steel box girders.


Figure1: Arrangement of strain monitoring points on the main girder
Each FBG strain monitoring station was connected with the overall fiber optic network by
an optical transceiver station and thus became a node of the network. The optical fiber

- 356 -

Fan & Chen: INTELLIGENT STRAIN MONITORING OF THE MAIN STEEL BOX GIRDER OF THE JIANGYIN YANGTZSE RIVER BRIDGE...
strain testing data were sent to the data processing and control station through the network.
Then the station implemented the pre-processing and buffer storage of the monitoring data.
2.2 Analysis of the strain monitoring data
After the FBG strain monitoring system was established, the strains were measured under
normal traffic conditions. A period of the monitoring data were randomly selected and
analyzed. The measured strain curves of certain section are shown in Figure 2.


-30
-10
10
30
50
70
90
110
0 20 40 60 80 100
Time(s)
Strain(με)
FBG-n-1
FBG-n-2
FBG-n-3
FBG-n-4
FBG-n-5
FBG-n-6
FBG-n-7
FBG-n-8

Fig.2: The measured strain curves

It can be seen from figure 2 that (1) the vehicle impact effects on the top and bottom of the
main girder are basically opposite, which demonstrates the pressure trends on the top deck
and the tension trends on the bottom deck; (2) the tested strains are within the normal range
indicating the reliability of the monitoring data.
According to the static testing report of “The Jiangyin Yangtze River Highway Bridge
Completion Test”, the strain level of the top deck of the main girder is about -150 με under
the test loads, while the strain level of the bottom deck is between 150 and 250 με.
Compared with the testing results in the report, the strain level (in this paper) is reasonable.
And combined with the load-level difference between the normal traffic state and the static
testing state, it can be initially found that the strain monitoring results of the main girder
agree well with those obtained in the actual situation.


- 357 -

Fan & Chen: INTELLIGENT STRAIN MONITORING OF THE MAIN STEEL BOX GIRDER OF THE JIANGYIN YANGTZSE RIVER BRIDGE...
3. SAFETY ASSESSMENT OF THE MAIN GIRDER
3.1 Failure probability and safety indicators
The safety of the main girder can be evaluated using the reliability analysis of the structure.
The failure probability
f
P
(or the reliability index

⤠潦⁴h攠et牵捴畲攠楳r摥晩湥搠慳⁴桥
pr潢慢楬楴礠if⁴ 攠汯慤⁥晦散琠
S
exceeding the structural resistance effect
R
during a
period of time:

 



0
)()(0
sr
SRf
drdssfrfSRPP
(2)
where
S
and
R
are both independent random variables;


rf
R
and
 
sf
S
are the
probability density functions of the structural resistance effect and the load effect,
respectively.
The probability density function


sf
S
of
S
(stress, bending moments, shear forces, etc.)
can be obtained directly from the monitoring system. And the probability density function
 
rf
R
of
R
can be described as a Gaussian probability density function with the mean
R

and the variance
R

.
R

and
R

can be taken according to the design codes or from
the field material testing. But the measured stress (or stress ratio) distribution generally
does not necessarily satisfy the Gaussian distribution. Therefore, it is difficult to directly
estimate the failure probability. When the measured probability density distribution of the
load effect does not meet the Gaussian distribution, the Parzen window can be used to
estimate the probability density function
[7]
:










m
i
i
xx
m
xf
1
2
2
21
2
))(
exp
)2(
1
)(

(3)
After the probability density function of the load effect is represented by several Gaussian
probability density functions, the failure probability
f
P
and the reliability index

瑨攠
獴su捴畲攠捡渠扥⁤e瑥tm楮敤⁢礠sin朠瑨攠獥捯湤n 潲oe爠ra瑲楸⁴h敯特⸠䉵琠楮⁰牡捴楣攬⁩e⁩猠
di晦楣f汴⁴漠捯lput攠瑨攠晡楬f牥rprobab楬楴i⁵獩湧⁥煵a瑩潮 ⠳⤮⁔敲敦潲攬⁡e湵mb敲eof
慰灲潸業a瑥tme瑨潤猠桡癥⁢敥n⁤ 癥汯灥搬v such⁡猠 h攠䙩牳琭佲 摥爠剥汩慢楬楴 礠䵥瑨潤
⡆佒䴩Ⱐ(h攠卥捯dⵏ牤敲e剥汩慢楬楴礠䵥瑨潤
协前慮搠瑨攠eont攠䍡牬e⁳業u污瑩onⰠ
慭潮o⁷桩c栠hh攠FOR⁨ 猠扥bn mo獴 睩摥dy⁵獥 ⸠周攠䙏. 晩牳rly 捯湶n牴猠sh攠
牡rdom⁶慲慢ae⁘⁴o⁡n⁩n摥pe湤n湴⁳a湤nrd湯牭慬⁲慮om⁶慲慢ae⁕湡ne汹㨠
 
x
T u
(4)
Then the limit-state function can be described as the linear hyper-plane of the standard
normal distribution of random variables in the approximate space, and the minimum
distance between the origin and the limit-state hyper-planes is defined as the safety
indicator, namely:
 
1
* *
2
T
u u  
(5)
If the limit state function is enough precise, the failure probability can be approximately
obtained as follows:

- 358 -

Fan & Chen: INTELLIGENT STRAIN MONITORING OF THE MAIN STEEL BOX GIRDER OF THE JIANGYIN YANGTZSE RIVER BRIDGE...



f
P

  
(6)
where
Ф is
the probability density function of the standard normal distribution variables.
The failure probability and safety indicators of the structural components can provide direct
quantitative indicators for the integrated management system to the inspection,
maintenance and management of the bridge. Table 1 gives the relationship between the
safety indicators and maintenance decision-making strategy for the structural health
monitoring and safety evaluation system of the Jiangyin Bridge.
Safety level 5 4 3 2 1
Safety indicator β>9.0 9.0>β>8.0 8.0>β>6.0 6.0>β>4.6 4.6>β
Safety
Characterization
Excellent Very Good Good Fair unacceptable
Maintenance
Strategy
Concern,
without
checking
Protective
Inspection
Maintenance
checking
Appropriate
reinforcement
Repairing
Table 1: Relationship between safety indicators and decision-making strategy
3.2 Safety evaluation of the steel box girder
The measured strain data of the steel box girder were used to assess the safety indicators of
the main girder of the Jiangyin bridge. Firstly, the stress data were obtained by the
measured strains. And the mean, standard deviation and probability distribution of the
stress were calculated based on the principles of probability and statistics. Secondly, the
standard mean, standard deviation and probability distribution of structural components
were determined according to the
design codes
. Finally, the failure probability and security
indicators of structural components were calculated and assessed according to the relevant
principles of structural reliability.
The measured stress data of the steel box girder during the period of January 2006 to April
were randomly extracted and processed by statistical analysis. Each measuring point of the
cross section 1 # to section 4 # of the main girder was selected to analyze their safety
including "FBG-1-8-s", "FBG-2-6-s", "FBG-3-6 -s "and" FBG-4-7-s ". Figure 3 illustrates
the stress probability distribution of the measurement points.
0
5
10
15
20
25
30
35
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
Stress (MPa)
PDF


February
March
April

0
10
20
30
40
50
60
70
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
Stress (MPa)
PDF


February
March
April

(a) FBG-1-8-s (b) FBG-2-6-s

- 359 -

Fan & Chen: INTELLIGENT STRAIN MONITORING OF THE MAIN STEEL BOX GIRDER OF THE JIANGYIN YANGTZSE RIVER BRIDGE...
0
5
10
15
20
25
30
35
40
45
50
0
0.02
0.04
0.06
0.08
0.10
0.12
0.14
Stress (MPa)
PDF


February
March
April
0
5
10
15
20
25
30
35
40
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.10
Stress (MPa)
PDF


February
March
April

(c) FBG-3-6-s (d) FBG-4-7-s

Figure 3: Probability distributions of the stress
The safety indicators of the steel box girder are confirmed according to the corresponding
relationship between identified safety indicators and maintenance decision-making strategy,
as shown in Table 1. And the specific analysis results are given in Table 2.
FBG
number
Month Mean Variance


f
P

Assessment
FYB-1-
8-s
Feb 6.843 2.777 8.922 2.076e-18 Very Good
Mar 6.298 3.959 8.638 2.499e-17 Very Good
Apr 4.059 3.760 8.917 2.154e-18 Very Good
FYB-2-
6-s
Feb 5.699 5.927 8.004 4.903e-15 Very Good
Mar 8.843 8.402 6.832 2.921e-11 Good
Apr 9.066 9.322 6.497 2.726e-10 Good
FYB-3-
6-s
Feb 5.900 3.100 8.931 1.902e-18 Very Good
Mar 7.394 5.979 7.835 1.869e-14 Good
Apr 8.089 8.332 6.916 1.638e-11 Good
FYB-4-
7-s
Feb 7.775 4.233 8.409 1.763e-16 Very Good
Mar 7.859 5.690 7.898 1.134e-14 Good
Apr 5.870 6.062 7.938 8.250e-15 Good
Table 2: Safety indicators and evaluation of the main steel beam
It can be seen from Table 2 that: (1) the stress probability distribution of the steel box
girder of the Jiangyin bridge follows the basic normal distribution; (2) The safety indicators
of the main structural components are within a satisfactory level. And the local positions
are in good grades, indicating that most of the components have certain safety reservation.
4. CONCLUSIONS
This paper utilizes the FBG technology to monitor the stress and strain of the bridge
structure. Self-perception and diagnostic capabilities, containing smart features, have been
applied to the main steel girder of the Jiangyin Bridge. Structures possessing smart

- 360 -

Fan & Chen: INTELLIGENT STRAIN MONITORING OF THE MAIN STEEL BOX GIRDER OF THE JIANGYIN YANGTZSE RIVER BRIDGE...
characteristics are known as intelligent structures, such as intelligent bridge structures. To
be more specific, if materials or structures are further equipped with the capability of
automatically processing structural information, one may say that these materials or
structures have life features. The FBG sensors for monitoring the stress and strain of the
main girder are introduced in the structural health monitoring system of the Jiangyin
Bridge. The measured data are useful for structural damage detection, fatigue damage
assessment and structural state analysis. The structural stress and strains of the main steel
girder are obtained based on the FBG intelligent monitoring system and the following
conclusions can be drawn: (1) The strain data of the main girder are reliable, in good
agreement with real readings; (2) The principles and methods of the safety evaluation of the
Jiangyin Bridge are presented, based on the structural reliability theory, and then the safety
index and failure probabilities of the main girder can be calculated; (3) The probability
distribution of the dynamic stress follows the basic normal distribution, and the safety
indicators of the main components are within satisfactory levels, indicating that most of the
components have a certain safety reserve.

REFERENCES
[1] Pines, D. & Aktan, A.E. 2002. Status of structural health monitoring of long-span
bridges in the United States. Progress in Structural Engineering and Materials 4:372-
380.
[2] Aktan, A.E, Catbas, F.N, Grimmelsman, K., et al. 2002. Health monitoring for
effective management of infrastructure. S.C. Liu and D.J. Pines (eds.), Smart
Structures and Materials 2002: Smart Systems for Bridges, Structures, and Highways,
SPIE, Vol. 4696: 17-29.
[3] Zhang X.S., Du Y.L. & Sun B.C. 2003. The Fiber Bragg grating sensors used to real-
time monitor the cable tension of Cable-stayed Bridge. China Safety Science Journal
13 (7): 77-79.
[4] Melle, S.M., Liu, K.X. & Measures, R.M. (1993). Practical fiber-optic Bragg grating
strain gauge system. Applied Optics 32(19):3601-3609.
[5] Mastro, S. 2000. The effect of transverse load on fiber Bragg grating measurements”.
Electrical and computer engineering, Drexel University.
[6] Li, P., Zhong, Z.L. & Ping, Y,F. 2000. The application of optical fiber grating (OFG)
pressure sensor on train real-time tracing system. Proceedings of SPIE 4221: 67-69.
[7] Jiangsu Yangtze River Bridge Co., Ltd, Jiangsu Transportation Research Institute.
2008. Structural Health Monitoring and Safety Evaluation System of Jiangyin Bridge.
Nanjing: Jiangsu Yangtze River Bridge Co., Ltd.