A Novel Approach to Use Soil-Cement Piles for Steel Sheet Pile Walls in Deep Excavations

In recent years, owing to advances in technology, excavation pits have shown increased improvements. Taking advantage of advanced solutions combined with traditional ones has brought about considerable advantages for construction contractors and saves on expenses to carry out construction projects. Owing to their ability to analyze geotechnical problems, several calculation and simulation software, such as Plaxis, Bentley, along with many others, have grown in popularity. Among them, Midas is one such software, which is a set of solutions developed by the MIDAS IT company and is widely applied in many constructions. The authors evaluated the ability to use Midas software to calculate the stability of a wall in a deep excavation pit for the Ho Chi Minh City Water Environment Improvement Project. The results of these researches reveal that combining soil-cement piles and steel sheet piles decreases the internal forces in sheet steel pile walls. At the same time, this solution not only reduces horizontal displacement but also keeps the settlement of the soil around the excavation pit within the permissible range, which helps to ensure that the adjacent pavements are stable and will not crack. The results of this study can be applied to similar geological constructions.


Introduction
Tunneling construction has become an indispensable trend for traffic systems by exploring underground space in modern cities. It is the case, however, that each project has different conditions of geology, hydrography, tunnel structure, etc [11]. So profound research is constantly needed to assess the influences of tunneling on adjacent buildings.
Deep cement mixing columns (DCMCs) comprise the soil at the construction site and cement, which is grouted to the ground using the injection grouting pump. The drill bit is drilled down to loosen the soil until it reaches the depth of the soil layer that needs to be reinforced. After that, the drill bit then comes back and moves up. In the process of moving up, cement is grouted into the ground. This is a new technology applied in flooded areas where other types of columns do not meet the requirements. [1,16,21] In Ho Chi Minh City, the rapid pace of urban development has narrowed the construction area. The technical infrastructure systems are being upgraded and improved due to the rapid development of urban areas. As a result, the deep excavation of pits is frequent and inevitable. [2] Owing to their versatility and usefulness, sheet steel pile combined with a bracing support to stabilize the pit is often applied as a solution; however, among other problems, at the joints of steel sheet piles, water can often leak and spill into the excavation pit, thus causing difficulties during construction. [4] The solution of soil-cement pile walls is also frequently used in the construction of basements; however, due to the limited horizontal shearing resistance, they are only applicable to pits that are not too deep. [13,19] Currently, nobody has used soil-cement piles with steel sheet piles in a project; therefore, the authors proposed a simulation combining soil-cement piles with steel sheet piles to find a better solution.
Many authors have studied the factors affecting the horizontal displacement of the diaphragm wall in a deep excavation pit, including pre-stressing in the strut system, factors related to construction issues, and factors related to design issues. [14,17] Chang Yu Ou et al. (1993) [5] studied the relationship between the depth of a pit and the horizontal displacement of the diaphragm wall in the deep excavation pit. According to the results of this study, the largest horizontal displacement of the diaphragm wall in the deep excavation pit varies between 0.2 and 0.5% of the depth of the pit, as shown in Fig.1.  [4] Further, Chang Yu Ou (2006) mentioned the relationship between the depth of the diaphragm wall (H p ) and the displacement of the diaphragm wall, as shown in Fig. 2 [5]. The authors analyzed the horizontal displacement of the diaphragm wall in a 20m deep excavation pit using FEM (Fig.3).  . Correlation between fixed wall depth and horizontal displacement of a wall [5] When the struts are not filled, the wall will move like a cantilever. When the struts are installed, the stiffness of the struts is large enough to allow the wall to move around the junction between the wall and the struts and the largest horizontal displacement of the wall, located near the bottom of the pit. The displacement type of a wall is shown in Fig. 4 and Fig. 5.  Clough and O'Rourke (1990) discussed safety in a study of its effect on the horizontal displacement of the diaphragm wall in a deep excavation pit [6]. The researchers provided a correlation between the factor of safety against basal heave, the stiffness of the diaphragm wall, and the support system with the maximum horizontal displacement of the wall through Fig. 6.
In which F b is the factor of safety against the basal heave.
In order for the model to be able to give reliable deformation and stress distributions, the modeling area limits also need to be rational. The rationality here is understood that the model area must be large enough to cover the interaction between the deep excavation pit and the surrounding ground. K.J. Bakker (2005) proposed limiting the model area when analyzing deep excavation pits through the Plaxis software [3]. According to Bakker, modeling area limits depend on the width of the pit, the depth of the pit, and the length of the diaphragm wall. Helmut F. Schweiger (2002) studied the effect of modeling area limits on the analysis results of the horizontal displacement of diaphragm walls. Helmut F. Schweiger analyzed a deep excavation pit with different modeling area limits. From the research results, Helmut F. Schweiger commented that once the limit of the modeling area is reached, the expansion of the modeling area limit does not significantly affect the results of horizontal displacement analysis of the diaphragm wall in the excavation pit. B. Gebreselassie and H.G. Kempfert (2006) [8] conducted a sensitivity analysis of the parameters of the Hardening Soil model to the results of the behavior of a deep excavation pit in a normally consolidated clay foundation. The parameters that B. Gebreselassie and H.G. Kempfert [15] studied the correlation between the module E parameter in the Hardening Soil model of Plaxis and the SPT-N index by back analysis of some deep excavation works on the sedimentary foundation of Kenny Hill in Kuala Lumpur, Malaysia, and sea mud in western Malaysia.
According to the above studies, hardness parameters affect the analysis results of the basement wall horizontal displacement the most. However, determining the parameters for the ground models according to the model's theory is an impossible task because in reality the geological data as well as the results of laboratory and field experiments are not always complete and correct. Therefore, it is essential to determine the range of fluctuations for these parameters for each soil type or their correlation with other physical and mechanical parameters. This range of fluctuation and correlation was studied through the back analysis of deep excavation and the comparison of the monitoring results of some national and international authors. This paper presents the application of steel sheet piles combined with soil-cement piles to stabilize a deep excavation pit's wall for the Ho Chi Minh City Water Environment Improvement Project.

Simple Method
The simple method is based on past cases that had produced graphs of the relationship between different factors and the horizontal displacement of the diaphragm wall. Ou et al. (1993) [5] established the relationship between the largest horizontal displacement and the depth of the excavation pit for clay and sandy soils. Clough and O'Rourke (1990) [6] also studied deep excavation in the Taipei area as a basis for developing a correlation diagram of the largest horizontal displacement of the diaphragm wall with the factor of safety against basal heave of the diaphragm wall and support system. The above charts can be used to predict an estimated displacement of the diaphragm wall in similar conditions.

Method of Dependent Pressures and FEM
The method of dependent pressures and FEM are two common methods for analyzing the horizontal displacement of diaphragm walls in deep excavation pits. The advantage of either methods is the simulation of nearly all factors affecting the horizontal displacement of the diaphragm wall in any deep excavation pits. On the other hand, these two methods can be applied in computer software to reduce the volume and calculation time with more accurate results. However, the basic theory of these two methods is not really simple, especially the FEM, so that the analyst must not only have a strong basic knowledge but also practical experience. Application of the method of dependent pressures and FEM in analyzing the horizontal displacement of diaphragm wall in the deep excavation pit has been studied by some authors. M. Mitew (2006) [12] used the two methods to conduct horizontal displacement analysis of the diaphragm wall in the deep excavation pit (Fig.7).
Using Geo-FEM software, Masew calculated K h following the methods of Terzaghi (1955), Menard and Bourdon (1964), and Monnet (1994). For FEM, Mitew used the Mohr Coulomb model and Plaxis 2D software. Mitew divided soil hardness into four cases: FEM 1, soil hardness based on Polish standards; FEM 2, soil hardness based on previous studies; FEM 3, soil hardness based on geological survey results; and FEM 4, soil hardness based on on-site stress measurements. The results are given in Table 1. The results of the analysis were compared with those collected in the field. Mitew reported that the dependent pressure method gave extremely variable results because it is highly dependent on how the coefficient of K h is determined. Meanwhile, the results of FEM had little variation and were close to the monitoring results. However, Mitew also emphasized that the chosen background model and its parameters are crucial when using FEM. Krasinski and Urban (2011) conducted a deep excavation analysis using two methods [9]. The first method simulated the interaction between the foundation and the wall with a system of elastic springs, and the authors used Obudowy Głebokich Wykopów software (OGW) to improve on Winkler model. The second method was a FEM that used Plaxis 2D with the Hardening Soil model. Krasinski and Urban found significant differences between methods. For the first method, the authors discovered shortcomings when simulating the behavior of the wall and the foundation with a beam on elastic springs because these elastic springs did not fully describe the complex physical phenomena in the interaction between the wall and the soil. The authors also noted that both methods gave different results. To determine which method is more reliable, we must verify their results with actual monitoring data.

Engineering Geological Parameters
We analyzed the pits belonging to package G, located in SIP1. Our study was a part of the project on improving the water environment in Ho Chi Minh City. For the jacking pit SIP1-15, the physical and mechanical indicators of the soil layer in the study area are given in Table 2. Geological surveying was conducted by the Viet Nam Water, Sanitation, and Environment Joint Stock Company (VIWASE). The physical and mechanical characteristics of the soil layer and the cement-soil piles are collected in Tables 2-4.

Construction sequence of an excavation pit
The construction sequence of an excavation pit is shown in Table 5. Construction of the first strut floor (+ 0.0m).

5
Lowering the underground water level, construction of the first excavation (3m of soil) to the elevation of -3.0m 6 Construction of the second strut floor (-3.0m).

7
Lowering the underground water level, construction of the second excavation (3m of soil) to the elevation of -6.0m 8 Construction of the third strut floor (-6.0m).

9
Lowering the underground water level, construction of the third excavation (3m of soil) to the elevation of -9.0m 10 Construction of the fourth strut floor (-9.0m).

11
Lowering the underground water level, construction of the fourth excavation (3m of soil) to the elevation of -12.0m 12 Construction of the fifth strut floor (-12.0m).

13
Lowering the underground water level, construction of the fifth excavation (3m of soil) to the elevation of -15.0m 14 Construction of the sixth strut floor (-15.0m).

15
Lowering the underground water level, construction of the sixth excavation (3m of soil) to the elevation of -18.0m 16 Construction of the seventh strut floor (-18.0m).

17
Lowering the underground water level, construction of the seventh excavation (3m of soil) to the elevation of -20.0m

Simulations using Midas GTS NX software
The Larsen sheet piles diaphragm wall was pressed down to a depth of 30m. Cross and longitudinal sections of the strut in the excavation pit are shown in Figs. [8][9][10][11]. Details on the steel sheet piles reinforcing the excavation pit wall and the parameters of the struts are displayed in Tables 6-7.

Results
The results of using the Equivalent Material Simulation method (EMS) method on the Larssen sheet piles wall are illustrated in Figs. 12-14. The results of using the Real Allocation Simulation method (RAS) to determine the internal forces acting on a Larssen sheet piles wall are shown in Figs 15-17.         Based on these results, we determined that the results of the finite element model using Midas GTS NX software are equivalent those of analytical methods.
We also tested models on the excavation pit stabilized by soil-cement and steel piles. We modeled the problem of combining soil-cement and steel sheet piles to stabilize the excavation pit using the EMS method. When the excavation pit's wall has not been reinforced by soil-cement piles through comparison between the two models of MC, HS and monitoring results, it was found that the HS model gives results closer to the monitoring results (the most displacement position above the distance from the bottom of the pit of 2-3 m). When the excavation pit's wall has been reinforced by soil-cement piles, the displacement of steel sheet piles will decrease.

Ground surface displacement
When the bottom of the excavation pit has been reinforced, but the excavation pit's wall has not been reinforced, the settlement of the outer edge and the inside edges of the reinforced soil-cement piles' walls have a large and uneven difference. When combining soil-cement piles and steel sheet piles, the settlement is reduced and uniformly distributed, while the settlement within the zone of reinforced soil-cement piles increases because the soil layer has been reinforced with cement; thus the self-load is larger, with γ = 21 kN/m 3 , compared to the existing layer 2 with γ = 14.8 kN/m 3 and layer 4 with γ = 20.5 kN/m 3 .

Conclusions
Through this research and analysis, the author gave a general conclusion about the research content, evaluation of the application of sheet piles combined with soil-cement piles to stabilize a deep excavation pit's wall in the Ho Chi Minh City water environment improvement project as below. 2) A combination of soil-cement piles and steel sheet piles not only reduces horizontal displacement but also keeps the settlement of the soil around the excavation pit within the permissible range. This ensures the stability of the surrounding pavement and therefore prevents cracks.  By EMS method: In the MC model, the maximum surface settlement of T z = 0.019 m decreases by 47.36% to T z = 0.010 m (located far from the outer edge of the soil-cement piles).
The HS model has the largest surface settlement of T z = 0.004 m, which decreases by 47.5% to T y = 0.0021m (located at the outer edge of the soil-cement piles).  By RAS method: In the MC model, the maximum surface settlement of T z = 0.022 m decreases by 34.09% to T z = 0.0145 m (located far from the outer edge of the soil-cement piles).
The HS model has the largest surface settlement of T z = 0.005 m, which decreases by 36% to T y = 0.0032 m (located at the outer edge of the soil-cement piles).
3) A combination of soil-cement piles and steel sheet piles reduces the internal force of the steel sheet pile walls. In addition, the wall of soil-cement piles has a horizontal waterproofing effect on the excavation pit (preventing water from flowing into the pit), with a permeability coefficient of k = 0.0864 m/day. Therefore, it is used extensively in irrigation, infrastructure, and construction works for waterproofing of dike banks and diaphragm walls.