1. Introduction
The seepage field generated by precipitation is one of the most important environmental influences in sand and gravel formations and an important influence on the evolution of a geoid [
1]. As part of the crustal stress, the groundwater condition in the seepage field is directly linked to the distribution of crustal stress inside the geological body. The excessive extraction of groundwater increases the effective stress between the internal structures of the geological body, causing the compression of the structural skeleton, which is macroscopically manifested as deformation of the geological body, causing the subsidence of the ground; meanwhile, seepage causes particles in the strata to flow out with the water, resulting in a ground cavity and leading to the collapse of the ground [
2,
3,
4,
5]. A series of problems unfavorable to the project are closely related to the distribution of the seepage field around it. In order to prevent the deformation and damage of the surrounding engineering and geological stability, the pumping of groundwater should be minimized. Therefore, generally, during precipitation, a circular water-stop curtain can be set up. For a falling-bottom water-stop curtain, the construction cost is high and the construction is difficult; for a suspended water-stop curtain, the advantages of both water-stop curtain methods are fully utilized, which not only reduces the cost of the project but also minimizes the pumping of groundwater and the ground settlement around the pit [
6,
7,
8].
Sand and gravel geological layers have a very widespread distribution in China; this is a complex stratum, especially in a section adjacent to a river. Sand and gravel layers have complex characteristics, such as large voids, high porosity, and strong permeability, which brings a series of unfavorable problems to engineering construction projects, such as potential collapse, water permeability, pipe surges, and so on. The excavation of foundation pits in a sand and gravel layer with a high water level may cause a series of problems, such as seepage damage or quicksand, which directly affect the safety of the project. In order to prevent the occurrence of engineering accidents in foundation pits and to reduce the phenomena of quicksand and the rumbling of surging sand in the basement, it is necessary to lower the groundwater level to below the bottom of the foundation pit [
9]. This prevents the occurrence of accidents in foundation pit engineering and reduces the phenomenon of quicksand [
10].
The research on water-stop curtains mainly includes two aspects: analytical methods and numerical simulations. Regarding research using analytical methods, Banerjee S et al. [
11,
12] and Bereslavskii [
13] deduced the analytical solution for a steady seepage field, based on the theory of fluid dynamics, which was used to study the distribution characteristics of a seepage field. Shen et al. [
14] derived a series of simple formulas to calculate the head difference between two sides of the stop curtain. Wu et al. [
15] proposed a semi-analytical method, combined with pumping test data, to determine the aquifer parameters. Chen Zheng [
16] proposed a semi-analytical solution for the unsteady flow of groundwater in semi-confined pressurized aquifers in strip and circular pits, for drop-bottom stopping curtains and suspended stopping curtains, respectively, and established a finite difference numerical model. Many scholars have also conducted research on analytical methods for water-stop curtains [
17,
18,
19,
20,
21,
22]. The above analytical solutions for seepage are based on ideal conditions and generally assume more conditions, e.g., soils are generally assumed to be homogeneous and isotropic. With the rapid development of computer technology, some complex ground seepage problems that are difficult to address with analytical methods can be better solved using numerical methods.
Regarding numerical simulation studies, Madanayaka et al. [
23] derived simple expressions to estimate the flow rate and hydraulic gradient values, based on hundreds of finite element simulations. Miyake N et al. [
24] investigated the optimal length of a water-stop curtain, based on an example of a large-scale pit excavation project in Tokyo, Japan, using finite element analyses. Pujades E et al. [
25] used numerical analysis to derive semi-empirical equations to analyze the water isolation effect of suspended and drop-bottom water-stop curtains. Xu et al. [
26] carried out numerical analysis based on a groundwater flow model to evaluate the seepage interception effect of water-stop curtains. Wu et al. [
27,
28,
29] studied the seepage characteristics of a seepage field with a water-stop curtain by combining pumping test data and numerical simulation data, which was based on a subway pit project in Hangzhou City. Wang et al. [
30,
31] carried out a numerical simulation by using a three-dimensional finite difference method to study the comprehensive effect of water pumping, the curtain stopping efficiency, and groundwater recharge. Luo et al. [
32] used a three-dimensional finite element simulation method to design a precipitation optimization scheme for the pit precipitation method and set up a water-stop curtain for water isolation. Liu Shengli et al. [
33] used a numerical simulation to compare and analyze various groundwater control schemes, such as the water isolation method, precipitation method, and combination method of water isolation and precipitation, and they formulated a reasonable precipitation scheme. You Yang et al. [
34] simulated the effect of the insertion depth of the water-stop curtain on the depth of the water level drop outside the pit based on the finite difference method. Ma, Changhui [
35] and Jiang, Xinliang [
36] directly used numerical simulation software to analyze the seepage field under different working conditions, which they used to solve actual engineering seepage problems. Many scholars have also conducted research on the numerical simulation method for water-stop curtains [
37,
38,
39,
40]. However, there is relatively little research on the influence of factors such as the precipitation depth and aquifer thickness on the characteristics of the seepage field [
41,
42].
From the current research on water-stop curtain pit precipitation, there are many assumptions about the theoretical analytical solution of the distribution characteristics of the seepage field and few numerical analytical solutions for the distribution characteristics of a seepage field. Therefore, in this study, numerical simulation is used to investigate the effects of the precipitation depth and aquifer thickness on the seepage field distribution characteristics of the pit. Finally, the calculated values of the numerical simulation are compared with the measured values in engineering tests to verify the reliability of the numerical simulation, which provides a theoretical basis for the impact of precipitation on the ground and the setting of the precipitation control index, and also provides information for the decision making and construction of similar foundation pit engineering projects.
4. Model Validation
The dewatering wells for the foundation pit operation of Yangwan Station include J1~J7 and J73~79, and the specific distribution is shown in
Figure 14. The spacing between the dewatering wells is 15 m, and the actual depth is about 30 m. The water pump is lowered to a depth of 24 m, and the water level at the end of the foundation pit is stable at around 105.9 m, which is 8.38 m lower than the initial groundwater level of 114.28. An ultrasonic flow meter is used to monitor the water output of the operating dewatering well and measure the water level inside the dewatering well.
We bring the measured water volume from the dewatering well into the calculation model for calculations and compare the measurement and calculation results of the water level drop in the foundation pit and dewatering well. The comparison results are shown in
Table 6.
Upon comparison of the calculated and measured results, it can be seen that the calculated water level drop in the pit at the west end of Yangwan Station is 7.63% lower than the measured value. The absolute error of the calculated water level drop in the dewatering well ranges from 5.09% to 26.62%, and the average error of the water level drop in the pit and dewatering well is 11.7%. This is mainly attributed to the non-uniformity of the strata and the error of the dewatering well position, indicating that the calculation model can meet the calculation and analysis needs. By studying the distribution characteristics of seepage field in a water-rich ultra-thick sand and gravel layer under precipitation conditions, the characteristics of its pore water pressure, flow direction, flow rate, and water inflow are investigated. This provides a theoretical basis for the design and construction of pit descending in similar projects.