Surface Settlement during Tunneling: Field Observation Analysis

Abstract

We address the effect of three groups of factors on supplementary ground surface displacements during tunnel construction. The first group of factors includes the engineering and geological properties of the massif in which the tunneling is conducted; the second group includes the structural features of the designed tunnels and surrounding buildings, and the third group includes the engineering parameters of the tunneling process. The research takes advantage of the geotechnical monitoring data obtained during the construction of underground facilities and the engineering parameters of shield tunneling during construction of single- and double-track Moscow underground lines by using EPB (earth pressure balance)–TBM (tunnel boring machines) in different soils. The dependence of additional displacements, occurring above the designed tunnel, on the TBM pressure, is addressed in detail. The presence of a close interdependence is evidenced by a correlation coefficient equal to 0.77. No dependence of the settlement on the diameter or depth of the designed tunnel, the distance from the tunnel axis to the monitored object, the loading that comes from a building in the affected area, or the boring rate was identified. The consideration of this parameter can be used to predict the soil displacement around the tunnel at construction facilities having similar geological profiles and boring parameters.

1. Introduction

At present, tunnel boring machines (TBMs) are widely used to make underground tunnels as part of the consolidated development of underground spaces, to minimize the adverse effect on the surrounding development and to ensure the safety of the works and the quality of the tunnels under construction within pre-developed schedules [1]. This technology, avoiding the stripping of the ground surface, allows building underground structures in various geological environments without using special methods of tunnel boring. Ground surface settlement, caused by tunneling, is a problem in urban development, since additional displacements may cause damage to surrounding buildings, structures, highways, and utility networks [2].

Open face shields may be applied to rocky water-free soils, where the rock itself can prevent collapse during work performance [3]. However, when soft ground is developed, it must be prevented from falling into the tunnel boring machine to control extensive supplementary displacements of the massif surrounding the tunnel boring site [4]. In addition, groundwater pressure also reduces the face stability. Tunnels can be excavated below the groundwater level in different soils using face support technology [5]. Pressure is produced in the front part of the shield, in the plenum, stabilizing the soil of the face and preventing water from flowing into the machine body and the tunnel that is being built [6].

According to the Industry-wide Standard 2.27.19-2011 “Tunnel construction using tunnel boring machines and high-precision lining techniques”, issued by the National Association of Builders [7], there are several types of TBM pressure building techniques, applied depending on the geological and hydrogeological conditions:

−

Slurry pressure building;

−

Earth pressure building;

−

Air pressure building;

−

Combined pressure building.

Currently, shields are mainly used in Moscow during the construction of new branches of the Moscow underground, and the earth pressure technique is applied to ensure the face stability in the process of construction.

A TBM with this type of pressure building uses excavated material to support the downhole soil, which enters the plenum in the plastic state after mixing with a special conditioning composition [8]. Excess slurry is removed by a screw conveyor, which is used so that the plenum maintains the design pressure. The pressure in the plenum must correlate with the rate of the shield advancement and be high enough to maintain the soil stability, which is controlled by a combination of the force on the cutterhead and the rate of the material removal from the plenum using the screw conveyor [9].

The ideal soil for tunneling purposes is clay soil with a soft plastic consistency, due to which the conditions are in place to better control the soil’s removal rate and pressure maintenance both in the plenum and the near-face space [10]. When using TBM in coarser-grained soils, the optimal conditions for construction are achieved by artificial conditioning additives.

Conditioners are injected through the holes at the side of the cutterhead. Machines, designated for injecting the conditioner into the plenum and even into the body of the screw conveyor, are also used [11]. However, the most effective injection point is the cutterhead, because it guarantees the direct mixing of the additive with the excavated soil. The material is easier to transport to the surface if it is in a stable plastic condition, so a conditioner is added to the incoherent rock. Conditioners prevent the scattering of the excavated material during its conveyor transportation [12]. Additives, injected to stiff clay, generally prevent the formation of clumps and facilitate the clay’s passage through the plenum without adhering to the screw [13].

In the last century, researchers had the opportunity to study the extensive data generated during the construction of the London Tube. In 1969, Peck [14] presented an empirical formula of the ground surface subsidence during tunneling works, depending on the tunnel depth and the angle of internal friction of the ground. Further, in 1977, Atkinson and Potts [15] described the surface subsidence of clay and sandy soils relative to the tunnel depth and width of the design area of effect. Later, Verruij and Booker [16] analyzed the surface settlement due to tunnel ovalization (deformation), depending on its radius and the Poisson ratio.

Currently available research findings [15,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38] suggest that the settlement of the ground surface is affected by the following factors, which can be divided into three groups (

Table 1. The factors belonging to each group of potential influence on the supplementary displacement of the soil body in the process of tunneling.

Factor No.	Group of Influence	Factor
1	Geological engineering conditions	Physical–mechanical properties of soils [17,18,19,20]
2		Rheological properties of soils [21]
3		Class and type of soil * [15,17,22,23]
4		Thickness of soil layers above the tunnel [24,25]
5		Ground water level [20,23,24,25]
6	Structural features of a new construction facility and the number of stories in the surrounding buildings	Tunnel diameter * [18,23,26,27]
7		Tunnel slope [26,27,28]
8		Tunnel depth * [15,18,23,26,30,31]
9		Distance between the tunnel and the monitored object in the plan * [24,25]
10		Loading that comes from the building in the area of influence * [20,24,25,27]
11	Engineering parameters of the tunneling	Pressure balance * [21,24,25,27,28,29,32,33,34,35,36]
12		Injection rate and injected amount; modulus of elasticity of the grout [19,20,24,25,32,33,34,36,37,38]
13		Advance rate of the boring machine * [19,23,24,25,32]
14		Rotary table torque [19,24,25,34,35]

* Factors considered in this article.

):

• The geological engineering conditions;
• The structural features of new buildings and adjacent built-up areas;
• The engineering parameters of the tunneling.

We previously conducted a comprehensive study of the groups of factors on the sediment of buildings. Several articles have studied the influence of one or more factors. Numerous studies have considered different factors to solve particular problems using artificial intelligence. However, there is no comprehensive understanding of which parameters are more significant and how factors the correlate with field observations.

2. Materials and Methods

In previous work [39,40,41,42], we presented the results of the geotechnical monitoring of the settlement of buildings and structures in the area of influence of the tunneling works. Here, an additional study of the well-known data was conducted to identify the relationship between the supplementary displacements and the following factors: the depth and distance in the plan from the tunnel axis to the monitored object, the class and type of the face soil, the TBM diameter, and the loading that comes from the buildings and structures affected by the works.

The analysis was performed using vertical displacement data for about 30 construction facilities, including residential and public buildings, crosswalks, and ballast sections of public railroad tracks. The tunnels of the Big Circular Line and Nekrasovskaya Line of the Moscow underground were bored into the rock soil and soft ground using TBMs with cutting diameters of 6 and 10 m.

In addition, during the tunneling of the Rublevo–Arkhangelskaya line, we collected the monitoring data and engineering information on the tunneling parameters: the earth pressure and the shield advance considering the installation of the tunnel’s bearing structures.

During the construction of a new line of the Moscow underground that has two single-track tunnels with an external diameter Dex = 6.3 m, a Robbins EPB 2110/378 earth pressure tunnel boring machine was used; its cutting diameter was 6.59 m. During the construction of the lefthand tunnel, tunnel monitoring data were collected for the buildings on the daylight surface, the tunnel boring advance rate, and the tunnel bearing structure installation time, as well as the average earth pressure during the installation of each 1.2-m wide liner ring in the design position.

The investigated section of the tunnel’s longitudinal profile (Figure 1) was composed of sandy soils from the crown to the ground surface. However, the machine cutter contained predominantly clay soils. The distance from the tunnel top to the daylight surface varied from 23 to 29 m.

The determination of the technological parameter of the tunneling, the earth pressure, was calculated using the geological engineering benchmark data and design documents, according to the method presented in the Industry-wide Standard 2.27.19-2011 “Tunnel construction using tunnel boring machines and high-precision lining techniques”, issued by the National Association of Builders [7]. At the same time, as the TBM approached the monitoring object, large increments of deformations were recorded. It was decided to adjust the earth pressure considering the arching according to the Portodiakonov method [43], which turned out to be higher. The calculated earth pressure according to the Portodiakonov method (red line) and the actual earth pressure (green and blue line) are shown in Figure 2. As shown further in the work, a gradual increase in the earth pressure enabled the stabilization of the supplementary building settlement.

A single-floor addition to a residential building located at a minimum distance of 6 m from the axis of the tunnel was included in the area of influence of the tunnel construction works. The addition was made of precast reinforced concrete and had wall panels 0.25 m thick; it had a strip foundation made of wall panels resting on FL foundation slabs and a post footing under the columns. The configuration of the addition was close to the shape of a rectangle, with maximum axial dimensions of 28.1 × 18.2 m, a height from ground level of 4.8 m, and a basement that was 2.2 m high.

Given the findings of the examination, the Category 2 technical condition was assigned to the addition, with a maximum settlement limit of 30 mm and a maximum relative difference in settlement limits of 0.001.

The geotechnical monitoring of the building was initiated when the shield was located 20 m from the projection of the nearest wall (Figure 3). The enclosing structure of the addition had a deformation control benchmark (m1); it showed the maximum additional vertical displacement, which was successfully reduced as the pressure balance rose. The benchmark monitoring data, the pressure balance, and the advance rate are provided in

Table 2. Monitoring data and tunneling parameters.

Ring No.	P, Bar	Monitoring Cycle, No.	Advance Rate, m/hour	Settlement per Monitoring Cycle, mm	Absolute Settlement of the Relative Zero Cycle, mm
441	1.8	2	0.30	−0.9	−2.9
453	1.7	3	0.30	−8.8	−11.7
457	1.7	4	0.36	−13.3	−25.0
459	2.2	5	0.40	−2.6	−26.6
462	2.5	6	0.37	−0.7	−27.3
464	2.8	7	0.33	−1.1	−28.4
465	2.8	8	0.36	1.2	−27.2
466	2.8	9	0.35	0.8	−26.4

. The average face development time needed to install one ring was 3 h and 30 min, during which time, the TBM removed about 60 m³ of rock, and it took about one hour to install a precast ring.

3. Results

Given the above data, we analyzed the influence of the following factors on the surface settlement: depth (Figure 4a), the distance in the plan from the tunnel axis to the monitored facility (Figure 4b), the soil class and type in the tunnel face (Figure 4e), the TBM diameter (Figure 4d), and the loads associated with the buildings and structures affected by the tunneling works (Figure 4c). The authors identified a dependence only on the soil class in the face (soft ground/rock soil). There was little or no dependence between the other parameters and the settlement.

For all the graphs presented, the correlation coefficient between the settlement and the studied parameter was up to 0.4, which indicated a moderate relationship between the parameters.

The evaluation of the effect of the pressure balance values, presented in Table 1, on the additional vertical displacements was made calculated to identify regularities. The dependence between the settlement and the pressure balance is shown in Figure 5. The Pearson’s correlation coefficient [44] was equal to 0.77 between the earth balance and the benchmark settlement during one monitoring cycle. This means there was a close linear relationship interpreted as a strong correlation. Hence, the coefficient of determination [45] was 0.60. This parameter shows the percentage of sampling explainable by the influence of the studied factors. According to the Chaddock scale [46], the coefficient value indicated an evident relationship between the parameters. Using the Student’s t-test [47], the value of the correlation coefficient was recognized as significant in 95% of cases.

The analysis of the data from Table 2 shows that there was no correlation between the rate of tunneling and the building settlement. Consequently, the rate of the face excavation and lining installation did not affect the additional displacement of the surrounding structure.

The settlements’ calculation results were found to be similar to the settlements of our colleagues, whose articles we reviewed.

4. Discussion

The following conclusions can be drawn based on the analysis of the data obtained at the tunnel construction site:

(1)

The dependence of the additional displacements of the facilities above the TBM pressure balance was the most explicit (the correlation was 0.77). The dependence of the displacements on the type of soil was less evident, but it is also necessary to take this parameter into account when designing. There was no dependence on the other factors (see Table 1).

(2)

In the case of insufficient earth pressure (less than 2.0 bar), the building settlement increased in the considered conditions of the construction.

(3)

The rate of tunneling and installation of rings in the design position (excluding the cases of halting and restarting the shield) did not significantly affect additional displacements of the monitored facility.

(4)

Based on the case of this construction facility, there is a need to adjust the pressure balance at a distance of at least two diameters of the designed tunnel from the facility, based on the data of the geodetic observations of the ground surface settlement, taking account of the tunnel depth, as well as the geological and engineering conditions.

(5)

When the pressure exceeded 2.7 bar, a rise in the deformation benchmark was observed, which indicates the need to limit the pressure in each case.

It should also be mentioned that a well-studied tunnel enables controlling the pressure balance effectively and significantly reduces the settlement of the ground surface. Therefore, it is recommended to apply a system of advance drilling to identify the geological structure of the face in order to refine and correct the findings of the geological surveys at the initial stage of design.

Further, the authors plan to improve the method of identifying the pressure balance to further improve the TBM performance and optimize the construction of tunnels.