In the dual role of long-term natural environment and operation environment, leakage, unhomogeneous deformation, cracks and other structural disease will threaten the operation safety performance and service performance of the immersed tunnel. Furthermore, maintenance difficulty and financial costs increase dramatically with the increase in disease degree. Therefore, it is important to understand the development of tunnel disease and analyze the health status of the structure in real-time, so as to make timely maintenance strategies.
Since the 1990s, health monitoring systems have attracted the attention of academia and the engineering community. In civil engineering, the health monitoring system was applied to bridges, such as Shanghai Xupu Bridge[1], Flintshire Bridge[2], Akashi Kaikyo Bridge[3], and Great Belt East Bridge[4]. Subsequently, it was applied to buildings and shield tunnels, such as the building in Japan[5], the LBPSB building in California[6], Stafford Medical Building in Vermont University[7], the high-speed rail tunnel in Korea[8], and the bored tunnel in Singapore[9]. Immersed tunnels have also been built with health monitoring systems recently, such as the Yongjiang immersed tunnel[10] and Zhoutouzui immersed tunnel[11] in China.
The Nanchang Honggu Tunnel is the largest inland river immersed tunnel in China with complex geological and fracture conditions. The construction of the health monitoring and assessment system will provide strong support for its safe and economic operation and maintenance. Structure health monitoring and assessment cases of large immersed tunnels are rarely seen in the literature to date. There is no monitoring and structural health assessment criteria for immersed tunnels in China. The structural health monitoring and assessment system of this project has to be designed according to the hydrological, geological, structural characteristics and technical requirements of the operation, maintenance, monitoring and health assessment of the tunnel.
Based on the construction practice of the Nanchang Honggu Tunnel health monitoring and assessment system, this paper describes important contents of the health monitoring scheme, sensors layout, health assessment model and software functions.
1 Engineering Background
The total length of Nanchang Honggu Tunnel is about 2 650 m, including 1 305 m immersed tube section. Among 12 immersed tube segments, nine separate segments make up a total of 115 m in length, and the other three are 90 m. Construction clearance is 11.5 m×4.5 m, and the cross section size is 30 m×8.3 m. The runoff of the Ganjiang river is concentrated in April to June, accounting for 49.6% of the year, which causes a highly fluctuating water level from the rainy season to dry season. Fig.1 shows the tunnel crossing file and stratigraphic distribution, and Fig.2 shows the tunnel cross section.
①— Backfill; ③— Fine sand; ④— Medium sand; ⑤ —Coarse sand; ⑥ —Gravel sand
⑧—Strongly weathering powder sandstone; ⑧1 —Medium weathering powder sandstone
⑧2 —Slightly weathering powder sandstone; ⑧3 —Calcareous mudstone
Fig.1 Tunnel crossing profile and stratigraphic distribution
Fig.2 Tunnel cross section (unit:m)
2 Honggu Tunnel Health Monitoring and Assessment System Platform
The Honggu Tunnel health monitoring and assessment system consists of the hardware system and software system. The hardware system includes the sensor subsystem, data acquisition subsystem and data transmission subsystem. The software system includes the database subsystem, data processing and control subsystem, and the health assessment and pre-warning subsystem. The functions of the whole system are realized by the coordination of each subsystem. Fig.3 shows the platform architecture of the Honggu Tunnel health monitoring and assessment system.
Fig.3 Platform architecture of the health monitoring and assessment system
•Sensor subsystem As a basic part of the whole monitoring system, the automatic monitoring sensor subsystem is the sensing layer, which can provide authentic, real-time and reliable monitoring data for all monitoring items. It includes the pre-buried magnetic flux sensor, pressure box, crack gauge, buried strain gauge and steel bar meter, etc.
•Data acquisition subsystem The data acquisition subsystem’s role is to collect signals such as sound, light, electricity and magnetism, which are measured by the sensor subsystem, and process them into digital signals. Various acquisition equipment is used according to various output signals in this project. The magnetic flux collection instrument V1.0 corresponds to the magnetic flux sensors, the multi-channel vibration acquisition instrument corresponds to the pressure box, buried strain gauge and steel bar meter, and the signal data acquisition system V1.0 corresponds to the crack meter.
•Data transmission subsystem The digital signals processed by the data acquisition subsystem are required for sending to the database center for analysis and processing through the data transmission subsystem. In view of the fact that the wireless signal is weak in the Honggu Tunnel, the sensors are integrated through the wired mode, and then, data is transmitted to the monitoring center through the optical fibers, and then to the cloud base via the network. Data access is performed on computers of the mobile terminals.
• Database subsystem The database subsystem is a data processing system. In this project, the tunnel manager is administrated to share the database with the software system maintenance company which is the database operator.
• Data processing and control subsystem Data transmitted from the data transmission subsystem requires further processing and analysis through data processing and the control subsystem. This subsystem implemented the data query, storage, extraction, processing and visualization, etc., and controlled data acquisition equipment installed in the tunnel.
• Health assessment and pre-warning subsystem The main function of the health assessment and pre-warning subsystem is to evaluate and analyze the health status of each monitoring section and the whole structure based on the established Honggu Tunnel health assessment model and sent out real-time warnings of the corresponding health status.
3 Health Monitoring and Assessment System
3.1 Honggu Tunnel health monitoring technical scheme
Monitoring indices are selected by considering the tunnel design scheme[12-14] and geological condition. Fig.4 shows the details of the monitoring sensors’ placement.
The implementation of monitoring contents is based on the combination of real-time monitoring and artificial regular monitoring. Monitoring content and corresponding sensors are listed in Tab.1.
The pipe joint section is very vulnerable. Once uneven settlement or tensile displacement between adjacent pipes exceeds the shear key or rubber gaskets’ capacity, shear key or rubber gaskets will fail, which can cause the tunnel to be out of operation. Therefore, in this project, sensors are mainly placed in the pipe joint section, the maximum height of water level section, the middle of the tunnel segment section and where the longitudinal gradient changes.
3.2 Honggu Tunnel health assessment model
3.2.1 Establishment of hierarchical assessment model
Comprehensive assessment of the immersed tunnel transforms the tunnel structure into a number of key indicators, and then the evaluation criteria of the indicators is established by theoretical research, engineering experience and existing specification. The mathematical model of comprehensive assessment methods is introduced to calculate the reference value F, which reflects tunnel health status[15-18].
The analytic hierarchy process(AHP) method[19] is adopted to establish the Honggu Tunnel health assessment model. Fig.5 shows the details of the model.
①—Joint displacement (longitudinal); ②—Joint displacement (transverse); ③—Shear key stress(vertical);
④—Rebar stress (transverse); ⑤—Rebar stress (longitudinal); ⑥—Tunnel settlement;⑦—Concrete stress (transverse); ⑧—Concrete stress (longitudinal); ⑨—PC cable tensile force;⑩—Tunnel tube inclination; 
—Tunnel structure crack; 
—Shear key stress(horizontal)
Fig.4 Sensors placement
Tab．1 Health monitoring scheme
MethodMonitoring indexSensorReal-time monitoringUneven settlementDifferential pressure gaugeJoint opening and displacementCrack meterRebar stressVibrating string steelbar meterConcrete stress and temperatureConcrete stress and thermo meterShear key stressPressure boxAnchor cable tensile forceMagnetic flux anchor cable stress meterTube inclinationTiltmeterRegular monitoringLocal settlementLevelingCracksCrack meter+Artificial surveyLeakageArtificial surveyWater levelWater level meterThickness of cover layerSound locater
Fig.5 Honggu Tunnel health assessment system hierarchical model
3.2.2 Determination of subjective weight of health assessment index
In comprehensive evaluating programs,the traditional AHP method uses 1-9 scale methods to construct the judgment matrix. When the order of the judgment matrix is large, it is difficult to satisfy the consistency requirements of the judgment matrix. The multiplication scale method[20] is an improved AHP method, 1-9 scales in the AHP method is replaced by 1-2 scales, by means of the same and slightly larger. When the importance of index A is the same as that of index B, the judgment scale is 1, and the weights of index A and index B are 0.5 and 0.5, respectively. Similarly, index A is slightly larger than index B; the judgment scale is 2; and the weights of index A and index B are 0.575 and 0.425, respectively. This method has better judgment transitivity and rational scale value compared with the traditional AHP method. In this project, the multiplication scale method was introduced to determine the subjective weight.
3.2.3 Establishment of index assessment criteria
The health assessment of the whole structure was based on the assessment of indices in the index layer. Based on the design data, relevant specifications and published literature, index assessment criteria were specified. Health status is divided into four grades, both in the index assessment and whole structure assessment. Index assessment criteria is shown in Tab.2.
The overall criteria of the tunnel health assessment and coping principle is described in Tab.3, and the procedure to determine the comprehensive assessment value F is interpreted in the following part.
3.2.4 Determination of objective weight of health assessment index
The entropy weight method is widely used in objective weight determination[21-23]. It is supposed that the index weight is determined according to the amount of information contained in each index. In this project, the entropy weight method is introduced to determine the objective weight. Specific steps are as follows:
1) Establish the judgment matrix of tunnel health state,
(1)
2) Normalize the judgment matrix,
(2)
where xmax, xmin are the most satisfied and least satisfied indices of each health level, respectively.
3) Determine the entropy of assessment index,
Tab．2 Index assessment criteria
Note:H is the design thickness of the backfill layer.
Tab．3 Quantitative grade classification of tunnel disease
GradeHealth statusComprehensive assessment value FCountermeasureANot damaged or slightly damaged3.5
(3)
(4)
where fij is the specific weight of the i-th assessment object.
4) Calculate the entropy weight,
(5)
3.2.5 Fusion weight
The commonly used methods of fusion weight calculation are the weighted arithmetic mean, weighted square root and weighted square sum, etc. In this paper, the weighted arithmetic mean[20] method is used, and the subjective weight is regarded as a little more important than the objective weight. The fusion weight calculation formula is
ωi=0.575ω1i+0.425ω2i
(6)
where ω1i is the subjective weight and ω2i is the objective weight.
3.2.6 Flowchart of the evaluation procedure
With the aforementioned work, the fuzzy-AHP evaluation method can be carried out through the following flowchart(see Fig.6). Membership grade vector R1 is the evaluation vector of the index level; R2 is the evaluation vector of the factor level; and R3 is the evaluation vector of the target level; Grade vector G={3.75,3,2,1.25} according to Tab.3.
Fig.6 Flowchart of the fuzzy-AHP evaluation procedure
3.3 Honggu Tunnel health monitoring and assessment software system
The software system of this project adopted the hierarchical B/S structure, .NET development platform, and ASP.NET as well as ADO.NET as the core technology. Fig.7 shows a diagram of the system network topology. Fig.8 shows a diagram of the system software structure. Fig.9 shows the software interface in PCs. The platform function is described in Tab.4.
Fig.7 Network topology diagram
Fig.8 Software structure diagram
Fig.9 Software interface in PCs
Tab．4 Platform function description
NameFunctional specificationHomepageGIS topology, health status of structures, warning of struc-tures, work reportsStructureHealth status of structure selected, Topo exhibition, warning of structure selected, work report of structure selected, moni-toring index chartMonitoring programDeformation theme, strain/stress theme, force themeData analysisData comparison, data association, report managementWarning managementWarning information, processed warning information, un-processed warning information, batch processing of warning informationSystem configurationUser log, text message push management
4 Conclusions
1) The subsystems of the health monitoring and assessment system of the Nanchang Honggu Tunnel are elaborated from two aspects, i.e., the hardware system and the software system. The whole system includes the sensor subsystem, data acquisition subsystem, data transmission subsystem, database subsystem, data processing and control subsystem, and health assessment and pre-warning subsystem.
2) The health monitoring scheme is designed for the Nanchang Honggu Tunnel, including monitoring indices, monitoring methods, sensor selection, sensor layout and so on.
3) The AHP model for the structural health assessment of the Nanchang Honggu Tunnel is established. The structural health assessment is evaluated hierarchically from three levels: index level, factor level and target level.
4) The entropy weight method and multiplication scale method are employed to determine the objective weight and subjective weight separately. Then, they are integrated into a fusion weight by the weighted arithmetic mean method. This method improves the inadequacy of the conventional subjective weight method, which cannot consider the field measured data information.
5) The flowchart of the Nanchang Honggu Tunnel health monitoring and assessment system is given, and the structure and compiling technology of the software system are described in detail. The established health monitoring and assessment system provides efficient technical support for the Honggu tunnel management and maintenance and it also provides a good reference for other immersed tunnels.
[1]Liu X L. Fundamental research on safety and durability of major structures in civil and hydraulic engineering[J]. China Civil Engineering Journal, 2001, 34(6): 1-7. DOI:10.3321/j.issn:1000-131X.2001.06.001.(in Chinese)
[2]Curran P, Tilly G. Design and monitoring of the flintshire bridge, UK[J]. Structural Engineering International, 1999, 9(3): 225-228. DOI:10.2749/101686699780481970.
[3]Sumitoro S, Matsui Y , Kono M , et al. Long span bridge health monitoring system in Japan[C]// Proceedings of SPIE—The 6th Annual International Symposium on NDE for Health Monitoring and Diagnostics. Newport Beach, CA, USA, 2001, 4337:517-524. DOI: 10.1117/12.435628.
[4]Andersen E Y, Pedersen L. Structural monitoring of the Great Belt East Bridge[J]. Strait Crossings, 1994, 94: 189-95.
[5]Iwaki H, Shiba K, Takeda N. Structural health monitoring system using FBG-based sensors for a damage-tolerant building[C]//Proceedings of SPIE: Smart Structures and Materials. San Diego, CA, USA, 2003:392-399.
[6]Nayeri R D, Masri S F, Chassiakos A G. Application of structural health monitoring techniques to track structural changes in a retrofitted building based on ambient vibration[J]. Journal of Engineering Mechanics, 2007, 133(12): 1311-1325. DOI:10.1061/(asce)0733-9399(2007)133:12(1311).
[7]Fuhr P L, Huston D R, Kajenski P J, et al. Performance and health monitoring of the Stafford Medical Building using embedded sensors[J]. Smart Materials and Structures, 1992, 1(1): 63-68. DOI:10.1088/0964-1726/1/1/009.
[8]Lee J S, Choi I-Y, Lee H-U, et al. Tunnel measurement system and its application to Korea high-speed rail tunnels[J]. China Rail Science, 2004, 25(3):21-26.
[9]Mohamad H, Soga K, Bennett P J, et al. Monitoring twin tunnel interaction using distributed optical fiber strain measurements[J]. Journal of Geotechnical and Geoenvironmental Engineering, 2012, 138(8): 957-967. DOI:10.1061/(asce)gt.1943-5606.0000656.
[10]Liu Z G, Huang H W, Zhao Y H, et al. Immersed tube tunnel real-time health monitoring system[J]. Chinese Journal of Underground Space and Engineering, 2008, 4(6): 1110-1115. DOI:10.3969/j.issn.1673-0836.2008.06.025.(in Chinese)
[11]Gong H. Health monitoring on immersed tunnel [D]. Guangzhou: Jinan University, 2006. (in Chinese)
[12]Grantz W C. Immersed tunnel settlements[J]. Tunnelling and Underground Space Technology, 2001, 16(3): 203-210. DOI:10.1016/s0886-7798(01)00040-2.
[13]Huang M H. Running performance analysis and health monitoring system design of Yongjiang underwater tunnel[D]. Harbin: Harbin Institute of Technology, 2008. (in Chinese)
[14]Su J, Zhang D L, Niu X K, et al. Research on design of subsea tunnel structural health monitoring[J]. Chinese Journal of Rock Mechanics and Engineering, 2007, 26(S2): 3785-3792.(in Chinese)
[15]Fera M, MacChiaroli R. Use of analytic hierarchy process and fire dynamics simulator to assess the fire protection systems in a tunnel on fire[J]. International Journal of Risk Assessment and Management, 2010, 14(6): 504. DOI:10.1504/ijram.2010.037087.
[16]Saaty T L. Decision making with the analytic hierarchy process[J]. International Journal of Services Sciences, 2008, 1(1):83-98. DOI:10.1504/ijssci.2008.017590.
[17]Yazdani-Chamzini A, Yakhchali S H. Tunnelboring machine (TBM) selection using fuzzy multicriteria decision making methods[J]. Tunnelling and Underground Space Technology, 2012, 30: 194-204. DOI:10.1016/j.tust.2012.02.021.
[18]Aliahmadi A, Sadjadi S J, Jafari-Eskandari M. Design a new intelligence expert decision making using game theory and fuzzy AHP to risk management in design, construction, and operation of tunnel projects (case studies: Resalat tunnel)[J]. The International Journal of Advanced Manufacturing Technology, 2011, 53(5/6/7/8): 789-798. DOI:10.1007/s00170-010-2852-7.
[19]Hyun K C, Min S, Choi H, et al. Risk analysis using fault-tree analysis (FTA) and analytic hierarchy process (AHP) applicable to shield TBM tunnels[J]. Tunnelling and Underground Space Technology, 2015, 49: 121-129. DOI:10.1016/j.tust.2015.04.007.
[20]Sun W, Zhang D M, Jiang X H. Theory and practice on influence and protection of open excavation on existing large diameter shield tunnel[M]. Shanghai: Tongji University Press, 2014. (in Chinese)
[21]Luo X. Study on the diagnosis method and system of highway tunnel health status[D]. Shanghai: Tong Ji University, 2007. (in Chinese)
[22]Wang Y Q, Zhou S W, Sun T J, et al. A diagnosis method for lining structure conditions of operated tunnels based on asymmetric closeness degree[J]. Modern Tunnelling Technology, 2015, 52(2): 52-58. DOI:10.13807/j.cnki.mtt.2015.02.008.(in Chinese)
[23]Kuang L H, Xu L R, Liu B C, et al. A combination weighting method for determining the index weight in geological hazard risk assessment[J]. Chinese Journal of Underground Space and Engineering, 2006, 2(6): 1063-1067, 1075. DOI:10.3969/j.issn.1673-0836.2006.06.039.(in Chinese)