Transmissibility performance assessment for drive-by bridge inspection
Introduction
It is acknowledged that more than 11% of bridges are structurally deficient in the United States [1]. In Europe, majority of the bridges age between 60 and 80 years old [2]. In Australia, 72% of bridge transportation network was constructed before 1976 [3]. In Japan, the boom of bridge infrastructure development occurred between 1955 and 1975, and many of these bridges are expected to have serious deterioration defects within the next decade [4]. In addition to the aging problem, other factors such as construction defects, structural deterioration, material degradation, harsh environmental conditions, changing and increasing loading, as well as extreme events such as natural disasters may contribute to the non-optimal performance or even total failure of a bridge infrastructure [4].
On the other hand, a large proportion of bridge network is comprised of short to medium span bridges, with span length ranging between and . Those bridges are usually maintained by local governments. However, due to increasing budget constraints of local governments with consequent decrease in maintenance expenditure, large portions of those bridges may not be well-maintained [5]. Given the major role of bridge network as a key component of the transport infrastructure, its great economic impacts and potential consequences of any incident on public safety, it is, therefore, critically important to develop rapid and cost-effective tool for bridge health monitoring (BHM) [6].
Conventional BHM techniques require a large number of sensors to be directly mounted on the bridge to monitor the dynamic response [7], and consequently to identify the dynamic characteristics of the bridge, which may provide useful information about the bridge’s current health condition [8]. Although, this approach has shown promise in the past [9], its application is not widespread. This is mainly because of the large capital cost associated with instrumentation and maintenance of the sensory system as well as transmission, storage and processing of massive amount of data being generated in real-time [10].
To achieve a better performance in feasibility and cost efficiency, the idea of indirect monitoring of bridges or so called drive-by bridge inspection has been increasingly investigated. This approach takes advantage of coupling effects between a moving vehicle and a vibrating bridge, referred to as Vehicle-Bridge Interaction (VBI), to identify the characteristics of a bridge solely from the response of a passing vehicle over the bridge [11], [12]. As a result, no vibration sensors are required to be mounted on the bridge, and the approach only relies on a few vibration sensors to be installed on the test vehicle. Contrary to the direct monitoring methods, this approach is a rapidly expanding realm of research. A lot of research works are being conducted to identify bridge modal parameters from vehicle response, owing to the fact that various types of bridge deterioration are indicated by a change in modal parameters [13].
Successful identification of a bridge fundamental frequency from the response of a moving vehicle has been reported in the past research works based on the analytical, numerical, experimental or even field test investigations [10], [14], [12], [15], [16]. According to a recent field-test experiment conducted by Yang et al., accurate determination of the bridge frequency is feasible for low velocities and sufficiently high dynamic excitation of the bridge [10]. Promising extraction of the bridge mode shapes based on instantaneous amplitude by adopting different signal processing tools such as short time frequency domain decomposition or Hilbert transformation has also been reported [17], [18], [19]. Identification of damping using moving vehicle response has also been investigated by several researchers [20], [21], [22]. It is demonstrated that the intensity of spectral response obtained from both the bridge and the vehicle decreases as the bridge damping increases [23]. A few research works have also been focused on development of hybrid sensing to further improve the efficacy of VBI framework in BHM [24], [25]. Those researchers have adopted synchronized measurements from a reference point on the bridge and an instrumented vehicle to identify the bridge characteristics.
Although, the past research works have demonstrated the potential of a moving vehicle response for bridge characterization, its performance in characterizing a bridge depends on several key factors. This includes, the quality of the road surface profile [26], the speed of the moving vehicle, the presence of the ongoing excitation on the bridge, and the dynamic characteristics of the moving vehicle. The adverse effects of the road roughness on bridge characterization is a well-known phenomenon, and many researchers have made efforts to minimize or even eliminate its negative impact by specifically designed test vehicles such as a tractor-dual-trailer system or two connected vehicles [27], [28]. In addition, it has been consistently reported that for an optimum performance of drive-by bridge inspection, the vehicle should pass over the bridge at low speed to reduce the blurring effects of road roughness [29]. In a separate study conducted by Y.B. Yang et al., a hand-drawn cart was designed for measuring the bridge frequencies. It was demonstrated that the dynamic characteristics of a test cart were crucial to successful identification of the frequencies of the bridge. In particular, the elastic properties of the cart wheels were studied, and it was shown that the Polyurethane (PU) wheels were most suitable for use in the indirect test for extracting the bridge frequencies, since they showed no particular self-frequencies in the range of frequencies of interest [30].
From the literature, little attention has been made to the optimum choice of vehicle parameters to quantify the transmission performance of the test vehicle and, consequently, to optimize the effectiveness of drive-by-bridge inspection. Through this study, it is demonstrated that successful characterization of the bridge by the vehicle response significantly depends on dynamic characteristic of the vehicle. As a result, an index reflecting the transmissibility performance between the vehicle and the bridge is proposed. Through numerical and experimental investigations, it is demonstrated that, under sufficient excitation on the bridge, vehicle with maximum transmission performance is capable of identifying bridge-related information, consequently distinguishing different bridge states from one another. The results from this study will be useful for tuning the test vehicle, so as to reduce the number of unsuccessful trials during the field test. The contributions of the work are fourfold: First, this paper presents one of the early attempts of experimentally validating the feasibility of drive-by bridge inspection to identify an induced change in the structure of a bridge with less than 2% variation in the fundamental frequency. Second, the performance of transmission between the bridge and the vehicle is quantified through a novel index, and it is verified that a vehicle with a higher sensitivity index captures more bridge-related information while moving over the bridge. Third, through experiments, successful identification of the first three bending modes of the bridge up to frequency of 32 Hz, is verified while the vehicle moves at a constant and slow speed and while there is ongoing excitation on the bridge. Finally, the indices and criteria presented herein can open up opportunities for drive-by bridge inspection of a network of bridges with similar structural design by having an optimal test vehicle regularly passing over those bridges.
The structure of this paper is as follows; in Section 2, the concept of vibration transmissibility is first demonstrated through a simplified vehicle model and further, a novel index for quantifying the transmissibility performance is introduced. In Section 3, formulations of coupling effects between a moving vehicle and a vibrating bridge is derived, in which the vehicle is represented by a half-car model, followed by numerical investigations demonstrating the effectiveness of the proposed index. Finally, in Section 4, the experimental investigations of the work are presented and discussed. Section 5 concludes the work and highlights the limitations of the method with some suggestions for potential future works.
Section snippets
Transmissibility: proof of concept
In drive-by bridge inspection, the vehicle is acting as a moving sensor, thereby it is vital to evaluate vibration transmissibility between the source of vibration, (a vibrating bridge) and the receiver (a vehicle) to ensure acceptable transmission performance. To elaborate the concept of transmissibility, a simplified vehicle model with a single degree of freedom (SDOF) comprised of a mass () with a spring () and a damper () as shown in Fig. 1 is considered, where and ,
Vehicle-bridge interaction
In this section, the dynamic coupling equations between a moving vehicle and a bridge is developed. The capability of the proposed index in quantifying the transmission performance is then verified through numerical investigations.
Laboratory experiment
In this section, a scaled moving vehicle in the laboratory is used to investigate the feasibility of drive-by bridge inspection, while vehicles with different dynamic characteristics are employed. To this aim, the first three bending modes of a simply-supported beam specimen are targeted. Through experiments, it is demonstrated that a vehicle with a higher transmissibility index, has potential to identify the first three bending modes of the bridge, and subsequently separate different states of
Conclusions
This paper presented results from comprehensive analytical, numerical and experimental investigations on the feasibility of drive-by bridge inspection. A novel metric based on the transmission performance between a stationary vehicle and a bridge was proposed. It was demonstrated that a vehicle with a higher transmissibility index has a better chance to outperform when it is moving over the bridge. It was, further, shown that for successful extraction of bridge-related information from the
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgment
The authors wish to thank Japan Society for Promotion of Science (JSPS) for providing support for conducting this research. The authors would also thank Australian Research Council (ARC) for provision of support under Discovery Early Career Researcher Award (DECRA) scheme with Grant No. DE210101625. Thanks are also extended to the following students for their help in conducting the experiments: Mr. JIANG, Wenjie, Mr. TOSHI, Naoya, and Mr. HAN, Zhuoran.
References (52)
- et al.
A spectral-based clustering for structural health monitoring of the Sydney harbour bridge
Mech Syst Sig Process
(2017) - et al.
Measuring bridge frequencies by a test vehicle in non-moving and moving states
Eng Struct
(2020) - et al.
Extracting bridge frequencies from the dynamic response of a passing vehicle
J Sound Vib
(2004) - et al.
Estimation of dynamic tire force by measurement of vehicle body responses with numerical and experimental validation
Mech Syst Sig Process
(2019) - et al.
Use of a passing vehicle to scan the fundamental bridge frequencies: An experimental verification
Eng Struct
(2005) - et al.
Extracting the bridge frequencies indirectly from a passing vehicle: Parametric study
Eng Struct
(2009) - et al.
Extraction of bridge frequencies from the dynamic response of a passing vehicle enhanced by the emd technique
J Sound Vib
(2009) - et al.
Identification of bridge mode shapes using short time frequency domain decomposition of the responses measured in a passing vehicle
Eng Struct
(2014) - et al.
Identification of damping in a bridge using a moving instrumented vehicle
J Sound Vib
(2012) - et al.
Bridge damping identification by vehicle scanning method
Eng Struct
(2019)