www.nature.

com/scientificreports

OPEN Wireless vibration testing and

bridge deck damage identification
using underneath maintenance
walkway
Liang Li1 & Takaaki Ohkubo2
Bridges will inevitably experience structural damage during long-term operation, which will decrease
their structural load bearing capacity and durability. The vibration characteristics of bridge structures
can be used as evaluation indicators for accurately diagnosing their deterioration. This study aimed at
developing a vibration testing method for identifying bridge deck damage that does not affect traffic
and is easy to implement. The vibration tests on the in-service bridge revealed it was mainly subjected
to vertical vibrations, with the accelerations in the vertical direction about twice those in the horizontal
direction. Based on fast Fourier transform analysis, the dominant vertical vibration frequencies were
2.34 Hz, 4.10 Hz, 10.30 Hz, 14.45 Hz and 18.75 Hz. A series of 0.1–0.3 Hz bandpass filters were used
with those frequencies at the centers for modal analysis. The experimental results were then compared
to the finite element analysis results. This study proposed a vibration-based damage identification
method based on the differences between the bridge deck and both main steel girders. The study
confirmed the convenience and effectiveness of using underneath maintenance walkways for wireless
vibration testing.

Keywords Mid-span bridges, Inspection walkway, Wireless testing, Traffic loads, Vibration modes,
Displacement ratios

Natural disasters occur frequently worldwide. Hence, improving the disaster resistance of highways and bridges,
ensuring their safety, and fully utilizing roads and bridges is of concern for the industry and broader society.
Approximately 50% of the total number of Japanese bridges were built during past periods of fast economic
growth. The extensive construction of roads and bridges contributes significantly to the convenience of urban life.
However, with the rapid development of the economy and extensive utilization of heavy and fast vehicles, bridge
structures are often exposed to significant dynamic loads. After half a century in service, bridge deterioration and
damage problems have become increasingly prominent1. To achieve efficient bridge maintenance, it is necessary
to understand their usage status and structural performance. However, population ageing in Japan is becoming
increasingly severe and thus a decrease in the numbers of civil engineers and reduction in social infrastructure
maintenance and management budgets inevitable. The existing damage detection methods need be made
cheaper and less labor-intensive. Therefore, the demand for effective management methods that reduce labor
and maintenance costs will accelerate2. In previous studies, response analyses based on the wavelet transform
for bridge degradation diagnosis and safety assessment were conducted3,4. Various testing methods have
been proposed utilizing technologies such as intelligent materials, machine learning and drones5–9. However,
applications to practical structures require the understanding of the modal characteristics of the structural
systems in advance, numerical analyses and modal experiments10,11.
In recent years, a significant amount of attention has been devoted to the safety and reliability of existing
structures in Japan12,13. The Ministry of Land and Transport of Japan established the Social Infrastructure
Monitoring Technology Promotion Committee in 2013, promoting the adoption of detection technologies to
civil engineering infrastructure. The existing methods utilize visual inspections, long-term monitoring and
detailed inspections to diagnose existing bridge structures, and research on improving their accuracy is also
being strongly promoted14. Regular checks must ensure safety during periods between inspections, which is
expensive and labor intense. There are issues with improper use during inspections. In addition, Japan is prone
to earthquakes and typhoons, and vibrations under complex wind fields have stricter requirements for structural

1School of Civil Engineering, Tianjin Ren’ai College, Tianjin, China. 2Department of Architecture, Graduate School of

Engineering, Hiroshima University, Hiroshima, Japan. email: 223035@tjrac.edu.cn

Scientific Reports | (2024) 14:25247 | https://doi.org/10.1038/s41598-024-77179-y 1

www.nature.com/scientificreports/

safety15. In this context, research on vibration testing of in-service bridges has received much attention16–18.
This research studies an application of a wireless acceleration testing system to evaluate the performance of
bridge superstructures. The majority of existing studies have proposed quantitative evaluation methods of
bridge condition based on changes in vibration frequencies. However, the sensitivity of natural frequencies to
stiffness changes caused by structural degradation is weak, while multiple environmental factors can also cause
vibration frequency shifts19. In addition, obtaining precise and unique vibration modes requires simultaneous
measurements at numerous points, hence the effectiveness of some testing methods need be improved20.
Vibration testing and system identification based on moving vehicular loads have recently become an important
research focus21,22.
Researchers have developed various testing techniques, including vibration testing and anomaly detection
methods, to evaluate the soundness of infrastructures23,24. When a bridge structure is damaged, the natural
periods, vibration modes, damping and other vibration characteristics will change. However, more research
is still needed on how to apply dynamic loads required for vibration testing to full-scale bridges. For example,
large vibration-forcing devices can help induce vibration of bridge structures for structural identification, but
they are difficult to use because they obstruct traffic. Furthermore, the cost of using vibration-inducing devices
is relatively high.
On the other hand, experiments at low response amplitudes are prone to detrimental effects of external
disturbances and signal noise, and additional numerical structural simulations may be required to infer vibration
characteristics. Any changes in structural vibration characteristics caused by damage may be minimal, and
precise methods for excluding factors other than damage need to be established. For medium and short-span
bridges, vehicular traffic can excite vibrations. Suppose daily vehicular traffic can be used to obtain the vibration
characteristics of a bridge for the purpose of detecting anomalies. In that case, we can conduct vibration testing
of a bridge without using vibration-excitation devices. However, due to the irregularity of natural traffic, it is
necessary to consider the impact of traffic on the bridge vibrations for reliable damage diagnostics using traffic-
induced vibrations. For example, the speed and type of vehicles affect the propagation of bridge vibrations
and low-frequency noise25. Thus, the variability of bridge vibrations caused by vehicles during testing must be
addressed.
In the presented context, this study developed a simple vibration testing system and studied a damage
identification method for bridge decks. Vibration testing was conducted on small and medium-sized span bridges
under vehicular traffic using the developed wireless acceleration testing system. Vibration characteristics, such
as natural frequencies and modes shapes, based on the accelerations obtained during vibration testing were
analyzed. The occurrence of damage to the bridge deck was identified based on the changes in the vibration
characteristics and a comparison of those of the main steel girders. Based on the research results, the effectiveness
and convenience of the wireless system and vibration testing proposed in this study and the use of maintenance
walkways under the bridge deck have been demonstrated.

Experimental program
Sensor locations
Most studies measure vibrations by placing sensors on the bridge deck, as shown in Fig. 1a but doing so provokes
safety concerns and may block traffic. In addition, while choosing individual measurement points, bridge
deck pavement and other factors, need to be considered. As shown in Fig. 1b, to avoid the above issues, this
study conducted vibration testing using the maintenance walkway located under the bridge. The experimental
personnel attached the acceleration sensors on the bridge deck soffit or main steel girders from the maintenance
walkway. The sensors were fixed to the bridge deck soffit using a strong adhesive to ensure accurate sensing of the
vibrations of the bridge structure. Figure 1c shows the developed wireless vibration testing system that did not
require a wired power supply. Due to the vast open space around the bridge structure, wireless communication
was efficient. The preparation time for testing, such as sensor settings for a single test condition, was less than
30 min, which demonstrates the efficiency and convenience of adopting the wireless testing system. Moreover,
this study used the maintenance walkway under the bridge for testing. Due to the width of the medium and the
bridge height, the safety needed to be considered during the experiments.

Testing system
This study used a wireless acceleration vibration testing system jointly developed by the Hiroshima University
and a company in Japan for vibration testing and data analysis. As shown in Fig. 2, the testing system consists of
an acceleration sensor, a developed wireless signal transmission unit, data storage and a real-time data analysis
system. The system signal receiving and storage unit (a PC equipped with testing software) does not require
a wired connection to the signal transmission unit, providing great convenience for in-situ experiments. The
analysis software developed simultaneously allows for real-time waveform analysis at the testing site. The saved
data does not need to be transformed, ensuring the accuracy of the research. Table 1 provides the specifications
of the wireless testing systems used in this study. The MEMS 3-axis acceleration sensor weighed 500 g and was
fixed under the bridge deck using a strong adhesive to ensure stability. This wireless testing system can accurately
measure structural vibrations in a wide range of accelerations of ± 3,000 gal. The resolution is 24 bits (at 100,200
sps) and the sampling accuracy reaches ± 1 ppm@25 °C for high-performance and high-precision vibration
testing. In addition, the maximum communication distance of the Bluetooth technology used for wireless
communication is about 100 m. It can connect to seven sensors with three sensitive axes simultaneously. Indeed,
21 channels of acceleration were used in the reported experiments. When conducting bridge vibration tests,
considering the vibration characteristics of the bridge, seven sensors were arranged in two layouts, one covering
the entire bridge deck and another a specific localized deck area.

Scientific Reports | (2024) 14:25247 | https://doi.org/10.1038/s41598-024-77179-y 2

www.nature.com/scientificreports/

Fig. 1. Selection of vibration testing locations: (a) bridge deck, (b) under bridge maintenance walkway, (c)
wireless vibration testing system in operation.

Fig. 2. Wireless vibration testing system.

Testing and data analysis methods

To obtain the vibration characteristics of the bridge and identify the damage location on the bridge deck,
vibration tests were conducted in multiple operating conditions. The sampling frequency used in the vibration
tests was 100 Hz and the acceleration input type was DC. Five-minute long intervals of measured data were
selected for Fourier spectral analysis. In addition, the vibration modes of typical accelerations were extracted
using filters and Lissajous functions (three-dimensional synchronous wave displays). To accurately identify
vibration patterns, a finite impulse response digital filter was used.

Scientific Reports | (2024) 14:25247 | https://doi.org/10.1038/s41598-024-77179-y 3

www.nature.com/scientificreports/

Property Specifications
Acceleration transducer: 75(W) × 80(L) × 57(H) mm
Dimension
Wireless signal transmission unit: 58(W) × 111(L) × 39(H) mm
Acceleration transducer: 500 g
Weight
Wireless signal transmission unit: 300 g
Measurement range ± 3000 gal
Sampling frequency 100 sps, 200 sps, 500 sps, 1 ksps, 2 ksps
Sampling precision ± 1 ppm@25 °C (ultra-high precision)
Resolution 24 bits (100 sps, 200 sps)
Measurement directions 3-Axis, simultaneous AD conversion
Wireless communication Bluetooth Class 1, USB
Time synchronization Can
Echolocation Vertical axis 1 and horizontal axis 2

Table 1. Specifications of wireless testing system.

Fig. 3. The typical 3D views of the bridge.

Construction site Shimane

Construction year 2010
Structural arrangement Two three-span steel continuous steel girders with non- composite concrete deck
Bridge deck Prestressed cast-in-place slab (thickness 300 mm)
Bridge length 143.0 m
Effective width 10.28 m

Table 2. Specifications of tested bridge.

Description of tested bridge

Overview of tested bridge
Figure 3 shows the typical 3D views of the bridge. Table 2 list the basic specifications of the tested bridge. It was
built in Shimane, located in the southwest of Japan mainland and close to the Sea of Japan. The construction
was completed in 2010, but cracks were found in 2011. The superstructure comprises three-spans made of two
continuous steel girders with a non-composite cast-in-place concrete deck. The length of the bridge is 143.0 m
and the width is 10.28 m. In May 2011, repair works were carried out on the cracks noticed in the bridge deck.
Considering long-term deterioration issues, it is necessary to regularly inspect the bridge for safety and any
potential damage to its components. In this context, the reported vibration tests were conducted on the entire
bridge deck and locally at the crack location.

Measurement cases
Figure 4 shows a schematic diagram of the location of the wireless acceleration sensors on the tested bridge.
Visual inspection indicated that the cracks in the bridge deck were mainly concentrated near pier P2, as shown
in Fig. 4 (d). The cracks near pier P2 were primarily perpendicular to the bridge axis. Figure 4 (a) shows three
sensor layouts that were adopted for measuring the overall vibration characteristics of the bridge and identifying
local damage. CASE-1 had wireless acceleration sensors installed at pier P1 and P2 on the bridge axis. As shown
in Fig. 4 (b), measurement nodes M2 and M5 were located at the bridge axis on the bridge deck soffit. Figure 4

Scientific Reports | (2024) 14:25247 | https://doi.org/10.1038/s41598-024-77179-y 4

www.nature.com/scientificreports/

Fig. 4. Overall bridge view, damage and sensor layouts: (a) sensor layouts (CASE-1: (Open square) M1-M6,
CASE-2: (Open triangle) Q1-Q6, CASE-3: (filled circle) N1-N6), (b) sensor attached to bridge deck soffit, (c)
sensor attached to steel girder, (d) deck cracks near pier P2.

(c) shows nodes M1, M3, M4 and M6 located on the main steel girders. CASE-2 had sensors at pier P2, where the
cracks occurred, and at abutment A2 bridge end, using the same method as CASE-1. The sensors of CASE-1 and
CASE-2 were arranged perpendicularly to the bridge axis and approximately 1.0 m from the inspection walkway.
To measure the overall vibration characteristics of the bridge, CASE-3 had sensors at the mid-spans and ends
along the entire length of the bridge axis. In the tests, in addition to placing sensors on the bridge deck, they
were also installed on both main steel girders to identify the occurrence of damage by studying the differences in
their vibration characteristics. The acceleration signals collected by the testing system can be double-integrated
to obtain displacements. To verify the double-integration method, a displacement meter was installed below an
acceleration sensor on the bridge deck.

Analysis of bridge vibration characteristics

Experimental vibration waveforms
This study focused on the vibration characteristics of bridges under actual traffic loads, thus it obtained
accelerations at various measurement points caused by vehicular traffic. Figures 5 and 6 show the corresponding
10-second long acceleration waveforms for CASE-1 layout at measurement node M5 on the bridge deck and
node M6 on a main steel girder, respectively, for the same vehicle passage. When the vehicle passed, the bridge
mainly vibrated in the vertical direction. As shown in Fig. 5, the acceleration magnitude in the vertical direction
was about twice that in the horizontal direction. The acceleration magnitude in the Y direction of the steep
bridge axis was about twice that of the bridge axis direction. In the flat order, it is mainly lateral vibration.
Comparing Figs. 5 and 6, it can be seen that there was little difference in the vertical or horizontal accelerations
between the bridge deck and the steel girder when for the same vehicle passage. Specifically, the maximum
difference in horizontal acceleration is 0.41 gal, and 3.64 gal in vertical acceleration. In the vertical direction,
the accelerations of the bridge deck were significantly higher than those of the main steel girder. Based on the
above vibration characteristics, this study focused on the vibration characteristics in the vertical direction and
attempted to identify damage using the relative vibration differences between main steel girders and bridge deck.

FFT analysis results

In this experiment, to obtain the vibration characteristics of the bridge and identify damage to the bridge deck,
a fast Fourier transform (FFT) analysis was conducted that focused on the vertical vibrations. In the CASE-1
layout shown in Fig. 4, a 300-second long experimental time history was divided into small sections of 20.48 s
with a 50% overlap between sections to perform an FFT analysis. Figure 7 shows the Fourier spectra of CASE-
1 layout. At the mid-spans, the observed natural frequencies were 2.34 Hz, 4.10 Hz, 10.30 Hz, 14.45 Hz and
18.75 Hz. However, it is difficult to identify the damage present near bridge pier P2 from Fig. 7.

Scientific Reports | (2024) 14:25247 | https://doi.org/10.1038/s41598-024-77179-y 5

www.nature.com/scientificreports/

Fig. 5. Acceleration waveforms of bridge deck: (a) node M5 in X-direction, (b) node M5 in Y-direction, (c)
node M5 in Z-direction.

Fig. 6. Acceleration waveforms of main steel girder: (a) node M6 in X-direction, (b) node M6 in Y-direction,
(c) node M6 in Z-direction.

Fig. 7. Fourier spectra.

Scientific Reports | (2024) 14:25247 | https://doi.org/10.1038/s41598-024-77179-y 6

www.nature.com/scientificreports/

Comparison of vibration modes and finite element analysis results

According to previous research, it is possible to quantitatively evaluate bridge damage from changes in natural
frequencies. However, it is also recognized that this method will struggle in situations with only minor or localized
damage. Many studies attempted to use the vibration characteristics for degradation diagnosis, focusing on the
changes in natural frequencies identified from vibrations measured under traffic loads to evolute the safety of
bridges. Therefore, determining the fundamental vibration characteristics of bridges, such as vibration modes,
displacements, etc., is important. In this experiment, traffic-induced vibration measurements were conducted on
the bridge and the its natural frequencies and global vibration mode shapes were examined.
Numerous methods have been proposed for mode shape identification26,27. This study used the natural
frequencies obtained from FFT analysis and adopted a bandpass filter to identify the vibration mode shapes at
each dominant frequency. The detailed procedure was as follows:

(1) A series of bandpass filters were set around the peaks of the FFT spectra to extract the corresponding sin-
gle-peak vibration signals.
(2) An inverse FFT was applied to the band-pass-filtered responses to obtain the accelerations around the dom-
inant frequencies.
(3) The filtered acceleration magnitudes at each measurement point obtained above were connected with
straight lines to construct mode shapes.

In the above-proposed scheme, the first two steps are commonly used, but the final visualization method for
vibration mode shapes is a new approach developed in this study. Figure 8 shows the visualized vertical vibration
mode shapes obtained by the proposed method of connecting the acceleration magnitudes at each measurement
point with straight lines. Based on the previous research results and the frequency values obtained from this test,
a bandpass filter width of 0.2 Hz (± 0.1 Hz) was adopted.

Bridge deck damage identification

This study proposed a method for detecting the deterioration of bridge decks, focusing on changes in vibration
characteristics caused by vehicle-induced vibrations. It is difficult to determine the location of damage in the
bridge deck from low-level vibration signals. However, when a known vehicle passes overt a bridge, the difference
between a healthy bridge deck and a deteriorated bridge deck can be identified based on vibration differences.
A test can be completed without obstructing traffic by deploying a wireless acceleration testing system via the
maintenance walkway underneath the bridge.

Analysis of bridge deck deformations

Cracks and fatigue damage in the bridge deck cause the stiffness degradation and changes in the vertical
deformations of the bridge deck when it is subjected to vehicular loads. A bridge deck damage identification
method based on absolute displacements and on-site have been proposed. In this study, we focused on the
relationship between the maximum accelerations and displacements, and conducted bridge deck damage
location and severity identification. At the same vibration level, a bridge deck with stiffness reduced by damage
will experience more significant deformations than the bridge deck without damage. Before conducting a
comprehensive evaluation, based on the acceleration data obtained from vibration testing, low-frequency
noise below 0.1 Hz was removed, and vehicle displacements caused by passing vehicles were calculated via
double integration using a high-pass filter. Figure 9 shows the displacements at nodes N2 and N5 from CASE-1
testing under medium truck vehicular traffic. When the same vehicle passed over, the maximum displacement
of the bridge deck at node M2 (where cracks occurred) was 0.061 mm, and no cracked bridge deck at node
M5 (undamaged locations) was 0.053 mm. To verify the correctness of the adopted calculation method, a
comparison was made with the displacement results measured directly by a displacement meter. As shown in
Fig. 10, the maximum deflection measured by the node N3 displacement meter of CASE 1 when the medium
truck vehicle passed over was 0.038 mm, whereas the estimated result from experimental data was 0.036 mm,
i.e., very close. In addition, the two vertical displacement shapes were essentially the same. Thus, the correctness
of the displacement estimation method proposed in this study has been verified.
Figure 11 shows the maximum acceleration-displacement relationship for the bridge deck. In this experiment,
the bridge deck location over pier P2 with a high probability of damage was determined via visual inspection
and earlier regular inspections. A comparison was conducted between the signals recorded at node M5 (deck
without cracks) and node M2 (damaged deck). The diagonal section represents the location of the damage.
Based on the data obtained from vibration testing, the maximum displacements and corresponding maximum

Fig. 8. Experimental bridge vertical vibration modes.

Scientific Reports | (2024) 14:25247 | https://doi.org/10.1038/s41598-024-77179-y 7

www.nature.com/scientificreports/

Fig. 9. Comparison of displacements between bridge deck locations with and without cracks.

Fig. 10. Comparison of displacement meter measurements with results estimated from vibration test data.

Fig. 11. Relationship between maximum accelerations and displacements of bridge deck.

Scientific Reports | (2024) 14:25247 | https://doi.org/10.1038/s41598-024-77179-y 8

www.nature.com/scientificreports/

Fig. 12. Vertical displacement ratio of bridge deck slab.

Sensor position Band-pass filter range (Hz) Logarithmic decrement (%)

4.1 ± 0.1 5.32
N2 (cracked)
10.3 ± 0.1 4.27
4.1 ± 0.1 7.32
N5 (no cracks)
10.3 ± 0.1 3.93

Table 3. Logarithmic decay rates.

accelerations caused by 10 small and medium-sized vehicles passing over were extracted and a relationship
between them inferred. As shown in Fig. 11, the absolute values of acceleration and removal at the location
where cracks occurred were greater than those of the bridge deck without damage. This suggests a potential
influence of cracks on the local stiffness of the bridge deck, thereby potentially altering its vibration response.

Comparison of vibration characteristics between bridge deck and main steel girders
According to previous research, fatigue damage caused by traffic-induced vibrations can easily lead to functional
degradation of the bridge deck. The research results suggest that the center of the bridge deck in the center of
the bridge had significant cracks in the direction perpendicular to the bridge axis. Therefore, if differences in
vibration responses of the central part of the bridge deck are found to both sides, it can help effectively identify
the deterioration. This study compared the ratio of vertical displacements of the bridge deck at the middle
position in layout CASE-1 to those of the main steel girders on both sides to detect the deterioration of the bridge
deck. The average value for both main steel girders was used to eliminate the impact of the vehicle position on
structural vibrations. The displacement ratio was calculated as follows:
ωs
(ω BL + ω BR ) /2 (1)
∆ =

where ∆ is the displacement ratio, ω s is the average displacement within a 10 s interval when small and
medium-sized vehicles pass through, ω BL is the representative displacement value of the left main steel girder
within the corresponding time period, and ω BR is the representative displacement value of the right main steel
girder within the corresponding period.
Figure 12 shows the displacement ratio of the bridge deck of bridge P1 (no cracks) and P2 (cracks present)
relative to those of the two main steel girders for six arbitrary time intervals for CASE-1 layout. The vertical
axis represents the displacement ratio, while the horizontal axis represents the six time intervals. Although the
selected displacement ratios showed randomness, the overall trend of the vertical deformations between the
bridge deck and the main steel girders on both sides was consistent. For time internals 1–3, 4 and 6, it can be
observed that although the displacement ratio increased with the increase in traffic-induced vibrations, the
differences for bridge deck location with and without cracks was small. When the vibrations exceeded a certain
level, as shown in Sect. 5, the displacement ratio of the bridge deck that produced shots was significantly higher
than that of the bridge deck part without cracks. The unique characteristics shown in Fig. 12 indicate that the
damage evaluation method based on the displacement ratio between the bridge deck and the main steel girders
effectively detected the bridge deck deterioration in the specific vibration environment. In future, more in-depth
research on vibration levels will be conducted.

Calculation of logarithmic decay rate

The results of the abovementioned deformation and displacement ratios verified the possibility of identifying
the damage location of the bridge deck. As shown in Table 3, for layout CASE-1, a free vibration attenuation

Scientific Reports | (2024) 14:25247 | https://doi.org/10.1038/s41598-024-77179-y 9

www.nature.com/scientificreports/

waveform after vehicle passing was arbitrarily cut, and the logarithmic attenuation rates at nodes M2 and M5
were calculated at the vibration frequency of 4.10 Hz as follows:
1 Ai
δ = ln (2)
m Ai+10

where δ is the logarithmic decay rate, m is the number of peaks, taken as 10 in this experiment, Ai is the
amplitude at peak i, and Ai+10 is the amplitude after 10 peaks.
From Table 3, the logarithmic decay rate at node M2 on the bridge deck (at the location where crack occurred)
was 5.32%, while that at node M5 (no damage) was 7.32% corresponding to 4.1 ± 0.1 Hz band-pass filter range.
For 10.3 ± 0.1 Hz band-pass filter range, the logarithmic decay rate at cracked node M2 was 4.27%, while that
for node M5 was 3.93%. It is found that the logarithmic decay rate was smaller at the cracked position for
10.3 ± 0.1 Hz band-pass filter range. The reason for this is that the randomness of waveform selection and the
simplicity of calculation method.

Conclusions
Degradation with time and damage caused by loads, directly affects the safety and comfort of use of in-service
bridges. There are many aspects that can be improved about the assessment of the security of bridges and the
usability of their components, such as bridge decks. Bridge managers face challenges in selecting the repair and
reinforcement timing and locations for in-service bridges. To address the above issues, this study proposed to
use an inspection walkway under the bridge deck to conduct vibration testing on the bridge using a wireless
acceleration vibration testing system. An easy-to-implement method for identifying vibration characteristics
and damage locations was developed of bridges with maintenance channels. The results obtained from this study
are as follows:

(1) When using an inspection walkway under the bridge for vibration testing, traffic will be unobstructed.
Using a wireless vibration testing system can reduce inspection time and labor.
(2) Using the wireless vibration testing system developed in this study, bridge accelerations, and vibration fre-
quencies and mode shapes were accurately measured using vehicular excitation as the dynamic excitation
source. The accuracy of vibration measurements and analytical results was verified by comparing them with
the finite element analysis results. However, more sensors would be needed for measuring higher-order
vibration mode shapes.
(3) By using a high-pass filter for double integration, the displacements caused by vehicle passing over the deck
were calculated. These were very similar to those measured by a displacement meter. A series of 0.2-Hz
band-pass filters were applied to the vibration peaks to extract the vibration mode shapes. These were found
to correctly reflect the vibration characteristics of the bridge.
(4) It was challenging to identify bridge deck damage from changes in vibration frequencies. However, based
on the difference in displacements caused by passing vehicles, a relationship between accelerations and
displacements can be specific to the location of the damage.
(5) This study proposed a damage identification method based on the vibration difference between the bridge
deck and main steel girders. The displacement ratios demonstrated that when the vibrations exceeded a
certain level, this method effectively identified bridge deck damage. In future, it will be essential to accumu-
late data from more experimental studies and improve existing procedures to ensure the reliability of the
proposed evaluation method.
(6) Using the current and future test results, it is considered that damage and degradation progression and rates
of the bridge deck can be regularly monitored, and the overall maintenance and management of the bridge
condition can be planned.

The effectiveness and accuracy of the wireless testing system and the testing method using an inspection walkway
proposed in this study have been verified. However, it is necessary to conduct more experimental studies to
provide a more solid basis for quantitative evaluation methods.

Data availability
The data presented in this study are available on request from the corresponding author.

Received: 9 November 2023; Accepted: 21 October 2024

References
1. Vasavi, S., Sravanthi, G. L., Ram, B. S. & Gokhale, A. A. Predictive analytics of bridge safety for intelligent transportation system
using ensemble model. Mater. Today 45, 5608–5616 (2021).
2. Al-Salih, H., Juno, M., Collins, W., Bennett, C. & Li, J. Application of a digital image correlation bridge inspection methodology on
geometrically complex bifurcated distortion‐induced fatigue cracking. Fatigue Fract. Eng. Mater. Struct. 44(11), 3186–3201 (2021).
3. Erduran, E., Pettersen, F. M., Gonen, S. & Lau, A. Identification of vibration frequencies of railway bridges from train-mounted
sensors using wavelet transformation. Sensors 23(3), 1191 (2023).
4. de Oliveira, J. V. M. et al. Vibration monitoring of a small concrete bridge using wavelet transforms on GPS data. J. Civ. Struct.
Health Monit. 9(3), 397–409 (2019).
5. Zhang, Z. et al. A novel GB-SAR system based on TD-MIMO for high-precision bridge vibration monitoring. Remote Sens. 14(24),
6383 (2022).

Scientific Reports | (2024) 14:25247 | https://doi.org/10.1038/s41598-024-77179-y 10

www.nature.com/scientificreports/

6. Yan, Z., Jin, Z., Teng, S., Chen, G. & Bassir, D. Measurement of bridge vibration by UAVs combined with CNN and KLT optical-
flow method. Appl. Sci. 12(10), 5181 (2022).
7. Ercolessi, S., Fabbrocino, G. & Rainieri, C. Indirect measurements of bridge vibrations as an experimental tool supporting periodic
inspections. Infrastructures 6(3), 39 (2021).
8. Zhu, Y., Xu, B., Li, Z., Hou, J. & Wang, Q. Monitoring bridge vibrations based on GBSAR and validation by high-rate GPS
measurements. IEEE J. Sel. Top. Appl. Earth Observ. Remote Sens. 14, 5572–5580 (2021).
9. Sun, S., Liang, L., Li, M. & Li, X. Vibration-based damage detection in bridges via machine learning. KSCE J. Civ. Eng. 22(12),
5123–5132 (2018).
10. Wang, H. & Nagayama, T. Response spectrum model of vehicle dynamic load for the prediction of bridge vibration level due to
single vehicle-passage. Eng. Struct. 260, 114180 (2022).
11. Gu, Y. W. et al. Experimental and numerical study on vibration and structure-borne noise of high-speed railway composite bridge.
Appl. Acoust. 192, 108757 (2022).
12. Fujino, Y., Siringoringo, D. M., Ikeda, Y., Nagayama, T. & Mizutani, T. Research and implementations of structural monitoring for
bridges and buildings in Japan. Engineering 5(6), 1093–1119 (2019).
13. Fujino, Y. & Siringoringo, D. M. Recent research and development programs for infrastructures maintenance, renovation and
management in Japan. Struct. Infrastruct. Eng. 16(1), 3–25 (2020).
14. Akgul, F. Inspection and evaluation of a network of concrete bridges based on multiple NDT techniques. Struct. Infrastruct. Eng.
17(8), 1076–1095 (2021).
15. Yu, E. et al. Bridge vibration under complex wind field and corresponding measurements: a review. J. Traffic Transp. Eng. 9(3),
339–362 (2022).
16. Janas, L. Experimental study on vibration and noise characteristics of steel-concrete railway bridge. Sensors 21(23), 7964 (2021).
17. Xia, Z., Li, A., Shi, H. & Li, J. Model updating of a bridge structure using vibration test data based on GMPSO and BPNN: case
study. Earthq. Eng. Eng. Vib. 20(1), 213–221 (2021).
18. Brownjohn, J., Au, S. K., Li, B. & Bassitt, J. Optimised ambient vibration testing of long span bridges. Proc. Eng. 199, 38–47 (2017).
19. Ma, K. C., Yi, T. H., Yang, D. H., Li, H. N. & Liu, H. Nonlinear uncertainty modeling between bridge frequencies and multiple
environmental factors based on monitoring data. J. Perform. Constr. Fac. 35(5), 04021056 (2021).
20. Nie, Z. et al. Output-only complete mode shape identification of bridges using a limited number of sensors. Mech. Syst. Signal.
Process. 178, 109246 (2022).
21. Li, J. & Zhang, H. Moving load spectrum for analyzing the extreme response of bridge free vibration. Shock Vib. 2020, 1–13 (2020).
22. Ouchenane, M. Study of the vibratory behavior of the bridge under the passage of mobile loads convoys: comparative study
between uni-dimensional and three-dimensional modeling. Civ. Eng. J. 4(5), 926 (2018).
23. Ren, Y. et al. An anomaly pattern detection for bridge structural response considering time-varying temperature coefficients.
Structures 46, 285–298 (2022).
24. Ma, K. C., Yi, T. H., Yang, D. H., Li, H. N. & Liu, H. Multiorder detection of bridge modal-frequency anomalies considering
multiple environmental factors. J. Perform. Constr. Fac. 36(6), 04022046 (2022).
25. Iwabuki, H., Fukada, S., Osafune, T., Shimura, M. & Sasaki, E. Contribution of large-vehicle vibration and bridge vibration to low-
frequency noise generation characteristics. Appl. Acoust. 155, 150–166 (2019).
26. Mazzeo, M., De Domenico, D., Quaranta, G. & Santoro, R. Automatic modal identification of bridges based on free vibration
response and variational mode decomposition technique. Eng. Struct. 280, 115665 (2023).
27. Li, J., Zhu, X. & Guo, J. Bridge modal identification based on successive variational mode decomposition using a moving test
vehicle. Adv. Struct. Eng. 25(11), 2284–2300 (2022).

Author contributions
Conceptualization, Liang Li; methodology, Takaaki Ohkubo; software, Liang Li; validation, Liang Li and Takaaki
Ohkubo; formal analysis, Liang Li; investigation, Takaaki Ohkubo; resources, Liang Li; data curation, Liang Li;
writing—original draft preparation, Liang Li; writing—review and editing, Liang Li; visualization, Liang Li. All
authors have read and agreed to the published version of the manuscript.

Funding
This research was supported by Tianjin Research Innovation Project for Postgraduate Students (2022KJ044).

Declarations

Competing interests
The authors declare no competing interests.

Additional information
Correspondence and requests for materials should be addressed to L.L.
Reprints and permissions information is available at www.nature.com/reprints.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives
4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in
any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide
a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have
permission under this licence to share adapted material derived from this article or parts of it. The images or
other third party material in this article are included in the article’s Creative Commons licence, unless indicated
otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and
your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain
permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/
licenses/by-nc-nd/4.0/.

© The Author(s) 2024

Scientific Reports | (2024) 14:25247 | https://doi.org/10.1038/s41598-024-77179-y 11