![Drive-by damage detection in bridges using the apparent profile
E. J. OBrien*,†
and J. Keenahan
School of Civil, Structural and Environmental Engineering, University College Dublin, Newstead, Belfield, Dublin 4, Ireland
SUMMARY
The concept of using sensors on a passing vehicle, rather than on the bridge, is known as ’Drive-by’ damage de-
tection. The newly developed traffic speed deflectometer (TSD) is a device used for pavement deflection measure-
ments and is investigated here in numerical simulations as a means of bridge damage detection. A TSD vehicle
model containing two displacement sensors is simulated crossing a simply supported finite element beam contain-
ing damage simulated as a loss in stiffness of one of the elements. An adapted Cross Entropy optimisation algo-
rithm procedure is proposed to determine the apparent profile, where the displacements recorded by the sensors
are used as the inputs. The time-shifted difference in the apparent profile is used as the damage indicator. Results
show that this can be reliably used as a damage indicator in the presence of noise and changes in the transverse
position of the vehicle on the bridge. Copyright © 2014 John Wiley & Sons, Ltd.
Received 8 May 2014; Revised 30 September 2014; Accepted 23 October 2014
KEY WORDS: bridge; damage; highway; speed; traffic speed deflectometer; TSD; RWD; optimisation; time-shifted
1. INTRODUCTION
Owing to the shortcomings in visual inspections, in recent years, there has been a move towards
sensor-based monitoring of bridge condition. More recently, a small number of authors have shifted
to the instrumentation of a passing vehicle, rather than the bridge, in order to assess bridge condition.
This approach is referred to as ’drive-by’ bridge inspection [21]. For short-medium span bridges, there
is a good probability that only one truck will be present at a time given the minimum gap between
trucks is 1 to 2 s (corresponding to 20–40 m for a typical speed of 20 msÀ1
). It is reasonable therefore
to believe that the instrumented passing vehicle will not be contaminated by other heavy traffic on the
bridge. Therefore, bridge monitoring of this nature does not require lane closures, thus reducing traffic
delays and traffic management costs and improving levels of safety. Also, the bridge itself does not
need to be instrumented, and so the concept has the potential to be far more cost effective than
traditional SHM.
The feasibility of using a moving vehicle fitted with sensors to extract the first natural frequency of a
bridge has been verified in numerical studies and field tests [24,30,41]. The use of drive-by methods for
bridge health monitoring, rather than bridge frequency detection has been developed by Kim and
Kawatani [21], McGetrick et al. [26] and Toshinami et al. [38]. Later, Yabe and Miyamoto [40] use
the mean displacement of the rear axle of a city bus passing over a bridge a large number of times
as a damage indicator.
McGetrick et al. [27] investigate the effect of a road profile on the feasibility of using an instru-
mented quarter-car in numerical simulations to detect bridge natural frequencies. Results indicate that
the approach works well in the absence of a road profile. However, when an ISO Class ’A’ [19] road
*Correspondence to: E. J. OBrien, School of Civil, Structural and Environmental Engineering, University College Dublin,
Newstead, Belfield, Dublin 4, Ireland
†
E-mail: eugene.obrien@ucd.ie
STRUCTURAL CONTROL AND HEALTH MONITORING
Struct. Control Health Monit. 2015; 22:813–825
Published online 14 November 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/stc.1721
Copyright © 2014 John Wiley & Sons, Ltd.](https://image.slidesharecdn.com/drive-bydamagedetectioninbridgesusingtheapparentprofile-200806133658/85/Drive-by-damage-detection-in-bridges-using-the-apparent-profile-1-320.jpg)
Drive by damage detection in bridges using the apparent profile
- 1. Drive-by damage detection in bridges using the apparent profile E. J. OBrien*,† and J. Keenahan School of Civil, Structural and Environmental Engineering, University College Dublin, Newstead, Belfield, Dublin 4, Ireland SUMMARY The concept of using sensors on a passing vehicle, rather than on the bridge, is known as ’Drive-by’ damage de- tection. The newly developed traffic speed deflectometer (TSD) is a device used for pavement deflection measure- ments and is investigated here in numerical simulations as a means of bridge damage detection. A TSD vehicle model containing two displacement sensors is simulated crossing a simply supported finite element beam contain- ing damage simulated as a loss in stiffness of one of the elements. An adapted Cross Entropy optimisation algo- rithm procedure is proposed to determine the apparent profile, where the displacements recorded by the sensors are used as the inputs. The time-shifted difference in the apparent profile is used as the damage indicator. Results show that this can be reliably used as a damage indicator in the presence of noise and changes in the transverse position of the vehicle on the bridge. Copyright © 2014 John Wiley & Sons, Ltd. Received 8 May 2014; Revised 30 September 2014; Accepted 23 October 2014 KEY WORDS: bridge; damage; highway; speed; traffic speed deflectometer; TSD; RWD; optimisation; time-shifted 1. INTRODUCTION Owing to the shortcomings in visual inspections, in recent years, there has been a move towards sensor-based monitoring of bridge condition. More recently, a small number of authors have shifted to the instrumentation of a passing vehicle, rather than the bridge, in order to assess bridge condition. This approach is referred to as ’drive-by’ bridge inspection [21]. For short-medium span bridges, there is a good probability that only one truck will be present at a time given the minimum gap between trucks is 1 to 2 s (corresponding to 20–40 m for a typical speed of 20 msÀ1 ). It is reasonable therefore to believe that the instrumented passing vehicle will not be contaminated by other heavy traffic on the bridge. Therefore, bridge monitoring of this nature does not require lane closures, thus reducing traffic delays and traffic management costs and improving levels of safety. Also, the bridge itself does not need to be instrumented, and so the concept has the potential to be far more cost effective than traditional SHM. The feasibility of using a moving vehicle fitted with sensors to extract the first natural frequency of a bridge has been verified in numerical studies and field tests [24,30,41]. The use of drive-by methods for bridge health monitoring, rather than bridge frequency detection has been developed by Kim and Kawatani [21], McGetrick et al. [26] and Toshinami et al. [38]. Later, Yabe and Miyamoto [40] use the mean displacement of the rear axle of a city bus passing over a bridge a large number of times as a damage indicator. McGetrick et al. [27] investigate the effect of a road profile on the feasibility of using an instru- mented quarter-car in numerical simulations to detect bridge natural frequencies. Results indicate that the approach works well in the absence of a road profile. However, when an ISO Class ’A’ [19] road *Correspondence to: E. J. OBrien, School of Civil, Structural and Environmental Engineering, University College Dublin, Newstead, Belfield, Dublin 4, Ireland † E-mail: eugene.obrien@ucd.ie STRUCTURAL CONTROL AND HEALTH MONITORING Struct. Control Health Monit. 2015; 22:813–825 Published online 14 November 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/stc.1721 Copyright © 2014 John Wiley & Sons, Ltd.