1. Introduction
In the field of urban roads and bridges, compared to long-span bridges with their characteristics of long span, high cost, and long construction time, medium-span bridges within the range of 20–50 m are widely used in urban bridge design because of their flexible layout, economical application, and short construction time. As critical components of urban infrastructure systems, in-service middle-span crossroad bridges are vulnerable to inevitable deterioration due to long-term traffic demand, material aging, and harsh operational environmental effects. Continuous deterioration of middle-span crossroad bridge structures, if not repaired or reformed in a timely manner, can lead to cumulative damage, ultimately affecting structural performance to different degrees. In order to maintain the availability and integrity of existing crossroad bridges throughout their lifespan, it is essential to make sure that effective inspection and maintenance strategies are implemented in an optimal way within the limited available budgets [
1].
Structural health monitoring (SHM) has been recognized as a viable technology for ensuring the structural and operational safety of infrastructure systems by providing early warning of damage, preventing costly repairs or even catastrophic failure [
2]. The emergence of structural health monitoring (SHM) systems for bridges can be traced back to the 1980s. In England, a long-term monitoring instrument and automatic data acquisition system were installed on the Foyle Bridge, which primarily monitored the data of main girder deflection, temperature, strain, etc. It was the earliest monitoring system installed and was considered relatively complete [
3]. In the mid to late 1980s, the United States began to install monitoring sensors on many bridges. For example, more than 500 sensors were installed on the Sunshine Skyway in Florida to measure structural parameters such as temperature, strain, and displacement during and after the bridge’s construction [
4]. Muria developed a monitoring system to monitor the dynamic characteristics of a Tampico cable-stayed bridge with a total length of 1543 m [
5]. With the rapid development of signal processing, sensing, data collection, and computing technology in recent years, the application of structural health monitoring (SHM) in engineering structures has become increasingly popular [
6,
7,
8,
9,
10,
11,
12], and the SHM system has been increasingly implemented on different kinds of bridges, such as the Kap Shui Mun Bridge and Ting Kau Bridge [
13], Runyang Bridge [
14], Yonghe Bridge [
15], Tsing Ma Bridge [
16], and Yangtze River Bridge [
17] in China, the Oresund Bridge [
18], Henrique Bridge [
19], and Great Belt East Bridge [
20] in Europe, the I-39 Northbound Wisconsin River Bridge [
21] in the US, and so on. Although many SHM systems have been established and studied, the main focus has been long-span bridges. Regarding middle-span bridges, few studies have been carried out. Compared to the intricate SHM systems employed for long-span bridges, designing an SHM system for medium-span bridges is comparatively simpler. Although it still demands higher accuracy in monitoring parameters, it requires fewer parameters and is easier to operate. Given that medium-span bridges occupy a significant proportion of urban roads, designing SHM systems for such kinds of bridges has promising prospects in practical urban road bridge applications.
In this study, based on a mid-span crossroad bridge in Shenzhen, which underwent a series of inspections and corresponding reinforcements during 2011–2017, a structural health monitoring system was designed and installed on the bridge in order to assess the effectiveness of the prior reinforcement measures and prevent the recurrence of structural defects.
Section 2 mainly describes the target bridge and its existing defects. In
Section 3, a detailed SHM system design for the target bridge and the determination of monitoring and warning values are presented. In
Section 4, the monitoring results are mainly analyzed.
Section 5 summarizes and discusses the conclusion of SHM system design and its performance.
As one of the busiest crossroad bridges in Shenzhen, the health of the target bridge is crucial not only for ensuring traffic safety on the structure but also for maintaining the safety and normal operation of the entire traffic system beneath it. Long-term continuous monitoring enables the measurement of real-time structural responses and their long-term variations, thereby enabling authorities to identify significant changes in structural behavior and take timely action [
22]. In order to understand the real-time state of the bridge operation stage and prevent potential safety hazards to the traffic system on the bridge, a monitoring and warning system was established to closely monitor the bridge. As one of the most important structural response parameters, structural strain plays an important role in the SHM-based condition assessment of bridges [
22,
23], which allows for both the derivation of stress by structural components under in-service loadings, thereby facilitating the evaluation of the safety reserve or reliability of structural components, and the determination of other parameters like the inner forces of the monitored structural components, such as compressive or tensile forces, bending moments, shear forces, and torques [
7,
24,
25,
26,
27]. Based on long-term strain monitoring data, Cardini and DeWolf developed an envelope of maximum distribution factors, peak strains, and neural axis locations to determine the structural changes of a bridge [
28]. Therefore, we selected the strain of the main girder and a bridge pier as the monitoring parameters during the entire monitoring period. In addition, the displacement of the main girder and pier of the crossroad bridge as well as the crack width of the girder were also selected for long-term monitoring. The corresponding threshold was set simultaneously, which can give warnings when the indicators exceed the threshold. Based on this system, the real-time condition of the bridge can be monitored, and comparative analysis can be carried out through the analysis system, so as to grasp the running condition and ensure the safety of the bridge.
2. Target Bridge
The target crossroad bridge is located in Shenzhen, China, with a total length of 1270 m along the center line of the road; it was constructed in 1999. This crossroad bridge is designed as a two-way six-lane bridge, with a superstructure prestressed continuous beam that has a common height of 2.0 m. The whole crossroad bridge is divided into two sections, the east and the west, with the east bridge comprising 14 spans, and the west bridge comprising 16 spans. The cross-section of the east bridge is single-box with a double-chamber structure (see
Figure 1a) and a deck width of 14.74 m. As for the west bridge, it has the same cross-section and deck width as the east bridge, except the W13# bridge, which has a cross-section of a single box with three rooms and a deck width of 17.74 m. The piers of the crossroad bridge mainly consist of single-column piers with a diameter of 1.5 m; the top of some piers are enlarged to 2.0 m in order to accommodate bearings. Double-column piers have a diameter of 1.3 m and adopt the concealed cap beam. A picture of the target crossroad bridge is shown in
Figure 1.
From 2011 to 2013, a number of bridge inspection agencies conducted inspections of the crossroad bridge and found various structural issues. The detailed results of the bridge inspection are listed as follows.
- (1)
In August 2011, the phenomenon of relative vertical dislocation was observed between the decks of the middle spans E16#, E17#, E18#, E21#, E23#, and E24#, as well as the corresponding parts of the west bridge, with a vertical height difference of approximately 6–9 cm.
- (2)
In November 2012, bridges 2#, 3#, and 4# were found to be displaced to the outside of the curve to varying degrees, with bridge 3# being the most serious, with an average value of 8.0 cm and a maximum value of 10.0 cm. At the same time, varying degrees of cracking were found in the webs and bottom plates of prestressed concrete box girders in the east and west bridges 2#, 3#, and 4#.
- (3)
In June 2013, it was observed that bridge 3# in the east had a tendency of creeping to the outside again.
- (4)
In August 2013, an emergency inspection of the bridge was conducted, and the inspection results showed that the bridge had the following main issues:
- (a)
The results of the beam inspection indicated that vertical cracks were present in the webs of all bridges, except for the extension section of the target bridge. Furthermore, horizontal cracks were observed in the bottom plate, particularly in the curved bridge section.
- (b)
Circumferential cracking was detected on the inner side of the root curve or the outer side of the upper curve of some curved-beam split piers.
- (c)
The longitudinal movable support mainly exhibited the phenomenon of the upper steel plate sliding longitudinally relative to the lower steel basin.
According to the Technical Specification for Urban Bridge Maintenance (CJJ99-2003 [
29]), the maintenance grade of the crossroad bridge is classified as Class I, and based on the inspection results, the bridge was evaluated as an unqualified class.
- (5)
In July 2014, the corresponding repair and reinforcement measures for the detected bridge faults were completed. Minor cracks measuring less than 0.2 mm were sealed with grouting, while larger cracks exceeding 0.2 mm were reinforced by wrapping steel plates. As for the lateral slip issue of the curved beam, a combination of pier beam consolidation and installation of steel plate anchor bolt limit devices was employed for reinforcement treatment.
- (6)
In December 2016 to January 2017, upon conducting several subsequent inspections of the bridge, it was discovered that the lateral slip problem of the curved beam, which had been a concern prior to the reinforcement work, had been effectively resolved. However it was also identified that several other problems still persisted:
- (a)
The cracks in the corbel of the beam body and the circumferential cracking of the piers at the split pier of W10 and W11 spans still existed.
- (b)
The newly discovered oblique cracks in the outer web of E11 and E12 span were of significant concern due to their severity.
- (c)
The common pier column of the E27–W31 east-west bridge exhibited the most serious cracking, with oblique cracks present at both the root and upper portions of the column. In addition, circumferential cracks were observed to be expanding at the top of the No.1 pier of W40 in the straight section. Furthermore, 20, 12, and 8 circumferential cracks were identified in piers W34#, W36-2#, and W40-2#, respectively, with the maximum width of these cracks measuring 0.15 mm.
- (d)
The cracks in the inner and outer webs of the E12# pier, which are about 4 m away from the E12# pier, had developed significantly, with the maximum width of the crack in the middle web being 1.1 mm and the maximum width of the crack in the oblique bottom slab being 0.32 mm.
In July 2014, the relevant agencies completed the reinforcement construction of the crossroad bridge, and several bridge inspections were conducted from 2016 to 2017. Based on the inspection results, it was found that the reinforcement measures for the bridge were effective, but new structural issues had emerged that were yet to be solved.
Figure 2 illustrates some of the typical issues.
In order to obtain real-time information on the condition of crossroad bridges and ensure their safety, it is necessary to conduct long-term automatic monitoring and observation of the structural health status of crossroad bridges to mitigate the potential hazards associated with driving on and under crossroad bridges. Therefore, the construction of an efficient and practical monitoring and early warning system is of utmost importance in ensuring the safe operation of crossroad bridges and the highway traffic systems beneath them. Such a system must be designed to detect any potential structural issues or deterioration of crossroad bridges, thereby enabling timely maintenance and repairs in order to eliminate potential hazards associated with driving on and under crossroad bridges. By implementing such a monitoring system, the safety of both crossroad bridges and the highway traffic system can be effectively ensured.
4. Monitoring Results
The monitoring process, which lasted from February to August 2018, produced data that were analyzed. They are illustrated in
Figure 13, with the results summarized in
Table 6. The detailed observations are listed as follows.
- (1)
Strain of the main girder
During the six-month monitoring period, all strain sensors of the main girder of the crossroad bridge were in normal working condition. However, the data of the measuring points at six positions exceeded the limit (listed in
Table 6), which indicates that these positions require continuous attention.
- (2)
Strain of the bridge piers
Similarly, all of the strain-measuring sensors of the piers worked normally during the monitoring. Because some sensors were located in the construction area, the data suddenly changed obviously, but the long-term trend of the data was stable, which means that no structural defects occurred. Among the pier strain-measuring points, the data of two positions exceeded the limit (listed in
Table 6), and the data of other measuring points were in a normal state. As the out-of-limit points were near the construction area under the bridge, the influence of construction on the strain of a pier should be considered, and these two positions should be paid continuous attention.
- (3)
Creep of the main girder
During the whole monitoring process, all displacement sensors of the main girder were in normal working condition, and all data fluctuated within a small range, with no data exceeding the limit; this indicates that the previous structural reinforcement measures were effective for controlling the main girder creep.
- (4)
Crack width
For the monitoring of crack width, all sensors also worked normally. However, the data of only one position were in a normal state, and the three other positions had data that exceeded the limit (listed in
Table 6); this means that the bridge crack width should be the focus of continuous monitoring. Considering the influence of on-site construction, the data for changes in crack width are within the normal acceptable range. However, if the crack width continues to develop, appropriate measures should be taken to repair and restrain the development of these cracks.
- (5)
Settlement of the bridge piers
Due to the on-site construction under the bridge, the monitoring of pier settlement was delayed until 25 June 2018. From 25 June to 6 August 2018, the sensors at the settlement points of the pier were in normal working order. Based on the monitoring data, it was found that there were settlement development problems in two areas, but the limit value had not yet been exceeded. Considering that they may be affected by site construction, these two areas require continuous attention in the future. Fortunately, the data in other areas were stable, without any fluctuation or development trends.
Overall, based on the issues identified during the bridge monitoring process, the following recommendations are proposed for follow-up monitoring:
- (1)
As the strain levels of the W32# girder as well as W32-2# piers exceeded the limit and tended to increase gradually, as demonstrated in
Figure 13a,c, respectively, it is recommended that the pier of W32-2# undergoes an appearance quality inspection, and that the damaged and exposed sections of the girder be repaired and the pier reinforced to ensure the structural integrity of the bridge.
- (2)
Due to the aging of the bridge and the occurrence of numerous subtle issues, it is suggested that continuous monitoring of the bridge be implemented to prevent the escalation of these problems. Particular attention should be paid to areas that slightly exceeded their limits and to the two piers (E32# and W37#) that are expected to exceed their limits soon.
5. Conclusions
In this work, the operational status of the targeted crossroad bridge was closely monitored by establishing a bridge monitoring system. Over the past few years, several inspections have been conducted on the crossroad bridge to identify its major issues. Based on the locations of the troublesome areas identified during the inspections, monitoring sensors were installed on the main girders and piers of the bridge to monitor the real-time changes in the structure. Meanwhile, the finite element model of the crossroad bridge was developed to perform stress analysis of the bridge during its operational stage. The strain levels of the main girders and piers, the deflection of the girders, and the settlement of piers were all calculated, and the resulting values were used as warning thresholds in the monitoring system. In the event that any of these values exceeded their respective warning thresholds, the corresponding areas were identified, and appropriate measures were taken promptly to ensure the structural safety of the crossroad bridge.
The monitoring process for the crossroad bridge lasted for 6 months, during which data on the strain and creep of the main beam, the strain and settlement of the piers, and the crack width of the bridge body were collected. Based on the monitoring data from February to August 2018, the real-time working state of the crossroad bridge was analyzed and evaluated. No abnormalities were found in the bridge displacement and pier settlement, and data changes were within the acceptable range due to the on-site construction environment near the structure. However, the strain-monitoring data of some girders (W13#, W32#, W35#, E26#, E28a, E29#) as well as some piers (W10#, W32-2#) exceeded the limits. In addition, the crack widths of piers W10# and W36-2# had noticeably expanded. Based on the monitoring data and the actual situation on site, it was determined that the appearance quality of pier W32-2# needed to be inspected and that the damaged as well as the exposed parts of the beam need to be repaired, with reinforcement of the pier at the same time, to ensure the integrity of the beam structure. For the bridge structures in the main girder areas that slightly exceeded the limit, or the two piers (E32#, W37#) that were about to exceed the limit, it is recommended that continuous attention be paid to them in the future.
Through the practical application of SHM in the repaired and reinforced crossroad bridge structure, this study determined the actual operational status of the reinforced crossroad bridge through the analysis of the long-term monitoring data of Shenzhen crossroad bridge and put forward the corresponding countermeasures for some existing problems. These results can provide a valuable reference for future research and the practical application of SHM in crossroad bridges with similar bridge structures.