Investigation on the Influence Caused by Shield Tunneling: WSN Monitoring and Numerical Simulation

Traditional monitoring techniques are faced with the problems of low acquisition frequency and easy to be aﬀected by the construction environment during the shield tunneling, which cannot meet the actual needs of timeliness monitoring of surrounding environmental impact on shield tunnel construction. Based on this actual demand, a wireless sensor network (WSN) system was used to monitor the response of shield tunnel segments and surrounding buildings during the shield tunneling in this study. According to the result of the signal transmission test, an optimization scheme of microelectromechanical system (MEMS) sensor layout is designed to improve the monitoring eﬃciency of the WSN system. Through the comparative analysis of WSN system monitoring data and traditional monitoring data, it is found that, with the increasing distance between the monitoring section and the tunnel face, the convergence value of tunnel lining clearance gradually tends to be stable, and the wireless monitoring results of transverse clearance convergence of the tunnel in this section are consistent with the overall deformation trend of the convergence gauge monitoring results. This study also simulated the shield tunneling adjacent buildings using a nonlinear ﬁnite element method. A parameter sensitivity analysis of the support pressure of the excavation face and the grouting pressure at the tail of the shield is carried out. The results show that the surface settlement can be reduced by properly increasing the grouting pressure and the support pressure of the excavation face. Moreover, increasing the support pressure of the excavation face has a better inhibition eﬀect on the settlement of the surface soil than increasing the grouting pressure.


Introduction
e shield method is widely used in the construction of urban rail transit system. During the shield tunneling, the adjacent area is inevitably disturbed, ascribing to the excavation surface support pressure and ground volume loss [1,2]. It is of great significance and value to investigate the influence of shield construction on adjacent areas [3,4].
Huang and Wei [5] studied that a shield tunnel in a section of Shanghai subway was affected by a large number of overloads, which resulted in serious convergent deformation of the shield tunnel section and structural issues. e damaged section of the tunnel even showed bolt fracture, mass concrete falling off, and other issues. Sun et al. [6] analyzed the influence of shield construction on adjacent underground buildings, and the results showed that the shield construction would cause significant settlement of ground soil when it arrived. It is necessary to improve the advanced speed of shield construction and tail grouting and carry out secondary grouting in the overlying soil of shield construction to ensure construction safety. By considering the surface deformation and taking reasonable tunneling variables, Li et al. [7] explained the control measures of the ground deformation and settlement of buildings. Yoo and Kim et al. [8] introduced TURISK, a system that can predict ground deformation caused by tunnel excavation and evaluate the impact of excavation on adjacent buildings, and applied it to Daegu Metro Line 2. Based on the on-site monitoring data and the simulation results using finite element analysis (FEA), Mirhabibi and Soroush [9] studied the impact of parallel tunnel excavation on ground buildings by using subway Line 1 under construction in Shiraz and got some practical conclusions. In terms of the impact of shield construction on the structure, through three-dimensional numerical simulation of the tunnel and adjacent areas, Mroueh and Shahrour [10] obtained that if the dead weight of the building was ignored during modeling, the settlement value caused by tunnel excavation would be significantly reduced.
Total station instrument, level instrument, convergence gauge, and distributed optical fiber sensor are mostly used in the monitoring program of subway shield tunnel construction, as shown in Figure 1. e traditional monitoring measures are widely employed in the construction of shield tunnels. But the traditional way of tunnel construction monitoring is increasingly unable to meet the requirements in practice because of its high cost, nonautomation, nonrealtime property, low monitoring frequency, and limited monitoring area. It is difficult to realize full-range synchronous real-time monitoring, which makes the monitoring results difficult to reflect the health status of the structure in a timely and effective manner. erefore, a wireless sensor network (WSN) system was developed to monitor the response of shield tunnel segments and surrounding buildings during shield tunneling in recent years [11,12]. At present, multiple preliminary achievements have been made in the application of wireless intelligent monitoring in structural health monitoring engineering. Straser et al. [13] first proposed the application of wireless sensor network for structural health monitoring and developed a low-cost wireless sensing unit applied to the dynamic performance monitoring test of the Alamosa Canyon Bridge. Professor Rice and Spencer [14] made some achievements in the research on WSN used for structural health monitoring such as bridges. Ayyildiz et al. [15] introduced a system for monitoring the health of reinforced concrete structural members and applied it to reinforced concrete structural columns under simulated earthquake conditions. e system can successfully detect the cracks of structural members. Wang et al. [16] applied the wireless sensor network inclination monitoring system to the shield tunnel of Shanghai metro for real-time monitoring, put forward the conversion formula applicable to the calculation of horizontal convergence value of the tunnel slice inclination value of the shield tunnel, and verified the accuracy of the conversion relation through on-site monitoring data. Bennett et al. [17] set up a WSN monitoring system including a gateway and 26 wireless sensor fulcrums in the shield tunnel of the London Metro. e system monitors the rotation deformation and cracks width of shield tunnel segments for a long time, which showed the feasibility of the wireless sensor network monitoring method in subway shield tunnel. Bennett et al. [18] developed a set of wireless sensor network monitoring systems for tunnel structure health monitoring, which summarized the monitoring data into the gateway and transmitted it to the user visual interface through the GPRS network, and then applied the system to the monitoring projects of London Underground and Prague Underground. e above measurement point layout is mainly based on the relevant algorithms and criteria theory to obtain the optimal point layout, but the use of numerical simulation and wireless sensing technology characteristics combined with engineering practice of the measuring point layout is relatively rare; based on this, this paper will put forward a flexible and efficient wireless sensor fulcrum layout scheme.
To sum up, compared with the traditional monitoring method, WSN is easy to expand, intelligent, and of a high automatic degree. Sense of structure deformation based on wireless sensor network technology can achieve real-time dynamic monitoring and all-weather, automatic, remote security early warning, which can obtain the structure of the sensory information and timely grasp of the structure of the state of health. e WSN intelligent monitoring system and MEMS sensors including acquisition fulcrum, inclination sensor, and laser sensor are shown in Figure 2. is paper aims to design an optimization scheme of MEMS sensor layout. A three-dimensional finite element model was conducted to simulate the excavation process of the tunnel using ABAQUS, taking into account the effect of the complex environments involving adjacent building. Moreover, an extensive parametric study was carried out to investigate the impact of grouting pressure and support pressure on adjacent building and ground response.

Project Overview
e site condition consists of two parts: (i) the adjacent buildings and (ii) the right-bound tunneling. During the tunneling process, the shield machine passed the adjacent buildings which is a six-story frame structure (as illustrated in Figure 3). e minimum distance between the shield body and the adjacent building was 1 m. e shield tunnel and the adjacent building are shown in Figure 3. e building is 28 m in length and 15 m in width with a single-story height of 3 m. e foundation of the building is a strip foundation with a depth of 2 m, consisting of concrete square piles with a crosssectional dimension of 400 mm × 400 mm and an average depth of 20 m.
According to the site investigation, the geology of the right line tunnel contains miscellaneous fill, soft plastic silty clay, soft plastic clay, silty clay, plastic silty clay, diorite, and so on. e geotechnical layer related parameters are shown in Table 1. e interval shield tunneling of the right line with a depth of 19 m is in the form of circular reinforced concrete lining structure adopting the single-layer lining structure with staggered joints. e inner diameter of the lining is 5800 mm, the outer diameter is 6400 mm, and the lining thickness is 300 mm.

Monitoring Program
e shield tunneling has a certain impact on the adjacent buildings, ground, and other structures. It is necessary to monitor buildings and surface ground in the adjacent area periodically. e MEMS sensor including laser sensors and inclination sensors was applied in this monitoring program, and the WSN system was used to transmit monitoring data. During the shield tunneling, the laser sensors and convergence gauges were applied to monitoring the transverse clearance convergence of the lining. Moreover, the inclination sensors and total station were applied to monitoring the inclination of the adjacent building. e arrangement of monitoring points shall be followed in the following principles: (i) e layout of wireless monitoring points is designed first, followed by its details. Combined with the results of numerical simulation and engineering practice, the location of measuring points is comprehensively considered. (ii) According to the actual project, the location and quantity of wireless measuring points should be adjusted appropriately. (iii) e monitoring target should be adjusted according to the actual change of the project.
According to the experience of measuring point layout in shield construction and the actual engineering, the wireless measuring point layout of shield construction tunnel and adjacent buildings is carried out. Generally, a monitoring section is employed every 20 rings during shield construction. However, when there is an obvious deformation area, it is necessary to conduct one monitoring section of the transverse clearance convergence of the lining, and it is arranged every 8 rings of the lining in this area. e laser Advances in Civil Engineering sensing fulcrum is arranged in the tunnel for the real-time monitoring of the lateral clearance convergence of the lining. e distance between the laser sensing fulcrum in the tunnel is about 90 m, and the nearest is about 20 m. e signal transmission distance between the first layer relay jump fulcrum and the intelligent terminal is about 100 m. e layout of field measuring points is shown in Figure 4. Four inclination sensors are arranged at the four corners of the roof. e site layout is shown in Figure 4, and the plane layout of measuring points in the tunnel lining and roof is shown in Figure 5.
e detailed layout information of monitoring points of the tunnel and adjacent buildings passing through the section is shown in Table 2. Figure 6 shows the comparison between the results of wireless monitoring and convergence meter monitoring of tunnel lateral clearance convergence. It should be pointed that the positive value of headroom convergence means that the tunnel transverse spacing increases, while the negative value means that the tunnel transverse spacing decreases. e results show that the convergence value of tunnel clearance tends to be stable with the increasing distance between the monitoring section and the working face. e overall deformation trend of the wireless monitoring results is consistent with that of the convergence gauges. However, the data collected by convergence gauges fluctuate more than that those collected by wireless monitoring. is is because the wireless monitoring is not manually operated and automatically collects data, while the convergence meter will be interfered with by relevant factors every time it collects data. From the data, the maximum change value is less than 2 mm, which is much less than the safety control value of 10 m.

Incline of Adjacent Buildings.
Based on the monitoring results of the building inclination and settlement of the adjacent buildings, the influence of the shield tunnel construction on the six-story frame structure of the section is analyzed. Figure 7 shows the variation of the building tilt during shield tunnel construction with the difference of the distance between the excavation face and the monitoring section where the wireless measuring points A and B are located. It should be pointed out that a positive value indicates a tilt towards the side near the tunnel, while a negative value indicates the opposite. As can be seen from the above picture, the whole building leans towards the side of the tunnel. As the shield passed through the monitoring section where the measuring points A and B are located, the inclination of the house slows down, which is due to the squeezing of the surrounding soil when the shield moves forward. As the shield continues to move forward, the overall tilt of the adjacent buildings continues to increase. Until the slurry solidifies, the overall inclination increases slowly and gradually becomes stable. e maximum tilt rate is not more than 0.00008, which is within the safety control value of 0.004 and belongs to the tilt change within the safety range.

Geometry and Mesh.
A three-dimensional basic FE model with a dimension of 72 m (length) × 60 m (width) × 60 m (depth) was generated considering the adjacent building and the karst cavern filled with water. e dimension of the model was large enough to reduce the influence of the boundary conditions on the primary zone of stress and strain induced by the deep excavation. e bottom of the model is restrained from the displacements in all directions, while the nodes of the four vertical boundaries were set to zero displacements in a horizontal direction. A uniformly distributed gravity is applied to the whole model.
As illustrated in Figure 6, the soil mass is simulated by 92,304 elements of type C3D8R (8-node trilinear displacement and reduced integration modes). e six-story frame structure, lining, and grouting layer are simulated by 13,812 elements of C3D8I (8-node trilinear displacement and incompatible modes). In addition, 2,304 elements of type S4 (a 4-node doubly curved general-purpose shell) are employed for the struts and pillars. Uniform distribution of the initial void ratio of the soil is configured prior to the analysis.

Material Properties.
e sectional view of the station and bridge is sketched in Figure 7. As shown in Figure 7, the ground mainly consists of miscellaneous fill, soft plastic silty clay, soft plastic clay, silty clay, plastic silty clay, diorite, and so on. e detailed material parameters used in the FEA are listed in Table 1. e main features of the Mohr-Coulomb constitutive model adopted for modeling are shown as follows: (1) e material is initially isotropic.
(2) Yield behavior is influenced by the second principal stress (σ 2 ).   e construction procedures are simulated by the "element death" technique. e speed of the construction is configured based on the field record. As shown in Figure 8, the details of the simulation process are described as follows: Step 1. Initialization is conducted. e initial stress field of the model caused by gravity and surcharge load is configured in this step, while the initial displacement field of both is zero before the commencement of excavation.
Step 2. To simulate the tunneling of the shield machine, the 1st ring shield machine element is deactivated and the 8th ring shield machine element is activated. Remove the support pressure in step 1 while applying the same support pressure on the excavation face. Activate the first ring lining and grouting layer, and apply uniform grouting pressure on the inner surface of the soil corresponding to the grouting layer.
Step 8. e shield unit advances 7 rings. In this process, 7 ring lining and grouting layer units are "activated." e grouting pressure is applied simultaneously to the surrounding soil elements within this range.
Step 15. e shield unit continues to advance 7 rings; at this time, the grouting pressure in step 8 is all canceled, and the corresponding grouting layer is replaced by hardened material.

Influence of Adjacent Buildings during Tunneling.
As shown in Figure 9, measuring points A and B are two points at different positions on adjacent buildings. It is stipulated that the positive vertical displacement represents the uplift and the negative vertical displacement represents the settlement in the monitoring results. e surface settlement decreases gradually from the central axis of the tunnel to both sides, and the settlement curve is approximately symmetrical parabola shape. It is indicated that the settlement of measuring points A and point B has always occurred during tunneling. As the nearest distance between the adjacent buildings and the tunnel is 1 m, the settlement value of the adjacent buildings is close to the surface settlement value at the axis. e settlement of adjacent buildings is within 3 mm and is less than 12 mm of the settlement control value of adjacent buildings specified in the specification [19].

Influence of Surface Ground during Tunneling.
As shown in Figure 10, total station monitoring points E and F are two surface settlement monitoring points located above the longitudinal axis of the tunnel. e vertical displacement in the numerical simulation results and the total station monitoring results is specified in this study. e positive value of vertical displacement indicates the uplift, and the negative value represents the settlement. As shown in Figure 10, through a comparative analysis of the total station results and numerical simulation, the change trend of measuring points E and F is consistent. Due to the compression of the soil, the stress state of the soil in front of the shield tunnel changes, which leads to the settlement of the soil. When the excavation face is about 5 m away from the measuring point E, the supporting force of the excavation face increases, causing the soil mass to squeeze towards the direction of shield tunneling and thus causing a small surface uplift. en, the supporting force of the excavation face decreases, and the soil mass settles. When the shield arrived at the section where the two measuring points were located, the surrounding soil mass changed its stress state by nearly     1 mm due to excavation unloading. When the shield passes through the two measuring points, the friction between the shield shell and the surrounding soil produces a shear force, resulting in a settlement value of nearly 0.4 mm. As the shield tail comes out, a "tail void" between the pipe and the soil is created, resulting in a settlement value of 0.6 mm on the surface. Finally, the shield is far away from the two measuring points, and the soil is subject to long-term consolidation settlement.

Parametric Study.
e shield method is widely used in the construction of urban rail transit system. During the shield tunneling, the adjacent area is inevitably disturbed. e disturbance is mainly caused by the excavation surface support pressure and formation loss. A parameter sensitivity analysis of the support pressure of the excavation face and the grouting pressure at the tail of the shield is carried out. e simulation results of different excavation face support pressure and grouting pressure construction parameters are further analyzed. e support pressure parameters of the excavation face can be obtained by formula (1), and the support pressure parameters of the excavation face can be obtained by formula (2).
where K s is the adjustment coefficient of support pressure of excavation face. F SO is the initial support pressure of the excavation face. F S is the face support pressure adjusted according to different coefficients.
where K G is the adjustment coefficient of grouting pressure at the tail. F GO is the initial grouting pressure at the tail. F G is   Table 3.

Effect of the Different Parameters on the Surface
Settlement. Figure 8 and Figure 11 show the variation curves of surface settlement at different positions above the tunnel axis under the nine different working conditions. In the process of shield tunneling, the surface subsidence increases gradually, and the increased amplitude of settlement increases with the increase of the distance monitoring section. It should be pointed out that there is a small uplift on the surface at a distance of about -10 m from the excavation face to the measuring point E. is is because the shield increases the support pressure in the process of tunneling in this section. e maximum surface settlement of the two measuring points within the range of 15 m from the excavation face is close to 3.5 mm. e effect of changing the support pressure of the excavation face is more obvious than that of changing the grouting pressure. is is because the internal friction angle of the soil in this section is relatively small, the cohesion is relatively large, the self-reliance of the surrounding rock is relatively good, and the stability is relatively high. In conclusion, it can be seen that the excavation face support force can be reasonably increased in the construction to reduce the surface settlement.

Effect of the Different Parameters on the Adjacent
Buildings. Figures 12 and 13 show the change curves of settlement at different positions of adjacent buildings under the above nine different working conditions. When the shield passed the monitoring points A and B, the settlement of adjacent buildings increases gradually, and the increased amplitude of settlement increases with the increase of the distance monitoring section. e maximum settlement within 15 m from the excavation face to the monitoring point A and the monitoring point B is close to 3.5 mm and 3 mm, respectively. When the support pressure and grouting pressure increase gradually, the settlement value of adjacent buildings decreases gradually. For the settlement reduction of adjacent buildings, increasing the support pressure of the excavation face is more effective than increasing the grouting pressure (Figures 14-16. ).

Conclusions
In this paper, through the numerical simulation experiment, it is found that increasing the support force of the shield excavation face can effectively reduce the settlement of the surface and adjacent buildings and guide the construction of a subway section. rough numerical simulation experiment, it is found that increasing the supporting force of shield excavation face can effectively reduce the settlement of the ground and adjacent buildings. Numerical simulation of different construction parameters of grouting pressure and support force of excavation face is carried out. e following conclusions can be drawn: (1) According to the layout principle and matters needing attention in the engineering site, the layout scheme of the wireless measuring point is made. e results show that the monitoring data of the WSN is consistent with the one of traditional monitoring methods. Compared with traditional monitoring methods, wireless sensing technology shows the advantage of real-time dynamic monitoring in the application of practical engineering. (2) Shield tunneling has an obvious influence on surface settlement. It mainly makes an impact at the excavation face near the stage and the shield-tail crossing stage. ere is a slight uplift on the surface before the shield reaches the measuring point. en, a large settlement occurred caused by the shield machine overbreak.
(3) Increasing the grouting pressure and the support pressure of the excavation face can reduce the settlement of the surface and adjacent buildings. Besides, increasing the support force of the shield face can effectively control the settlement of surface and adjacent buildings.
Data Availability e data supporting the findings of this study are available within the article.

Conflicts of Interest
e authors declare that they have no conflicts of interest.