A rapid prediction model for the seepage stability of wide-graded soil and its construction method

Abstract

This invention relates to a rapid prediction model for the permeability stability of wide-graded soil mixes and its construction method. The model and construction method of this invention comprehensively consider the influence of physical conditions (porosity, specific gravity), loading conditions (loading stress, loading frequency, loading time), and environmental conditions (hydraulic gradient) on the stability type of wide-graded soil mixes. The fine particle loss rate model for wide-graded soil mixes constructed in this invention has clear physical meaning and a simple structure, greatly reducing experimental time and difficulty. It provides significant engineering convenience for wide-graded soil mixes with various gradation combinations and for field units lacking experimental conditions, and has high market promotion value.

Description

Technical Field

[0001] This invention belongs to the field of road engineering technology and relates to a rapid prediction model for the permeability stability of wide-graded soil and its construction method. Background Technology

[0002] Wide-graded soil refers to soil materials with a particle size greater than 2 mm accounting for more than 50% and a uniformity coefficient (Cu) greater than 10. Because wide-graded soil is widely found in nature, it is readily available and possesses good strength and resistance to deformation after compaction. It is commonly used as engineering fill material in dam foundations, filter layers, and airport high embankments. However, due to its wide particle size distribution, fine particles in the soil skeleton are prone to migration when hydraulic conditions change or when subjected to load disturbances. This migration and loss of fine particles along the pore channels formed by coarse particles is called internal erosion. The migration and loss of fine particles within the soil often alters the soil's microstructure, leading to changes in soil strength, increased permeability, and increased soil deformation, ultimately resulting in structural failure.

[0003] Earth-rock dam foundation materials, often composed of wide-grade soil, are prone to seepage damage during reservoir impoundment, flood discharge, and flood season events, due to changes in hydraulic conditions. Studies by Zhu Xiangdong et al. have shown that after prolonged erosion, dams and slopes possess stable internal hydraulic drainage systems. However, when new external disturbances (such as earthquakes, cyclic loads, or changes in hydraulic conditions) occur, the original equilibrium of the soil slope is disrupted. Under the influence of water flow, fine particles easily migrate through the pores in the coarse-grained framework, leading to erosion and impacting structural stability.

[0004] To ensure the stability of geotechnical structures, Terzaghi proposed a design criterion for filter layers. Based on this, Kenney et al. and Kezdi et al. conducted unidirectional seepage tests using permeability instruments to explore the stability of soils with different gradations and proposed a widely used soil stability criterion. However, in practical engineering, earth-rock dams and slopes often face complex hydraulic conditions and external disturbances due to climate change. Studies by Gu et al. and Wang et al. have shown that water level fluctuations and flood passage can induce soil particle migration, affecting the stability of geotechnical structures. Zhang et al. found that railway subgrades with stable gradations in water-rich areas are prone to mudslides and other defects under cyclic loading, affecting normal railway operation. All of the above studies indicate that changes in hydraulic conditions and external disturbances have a significant impact on soil stability. The applicability of the experience summarized from studies based on unidirectional seepage needs further exploration under different hydraulic loading methods and external disturbances.

[0005] In summary, the problems of seepage failure of earth-rock dams, slope stability, and mud pumping in high-speed railway subgrades are all essentially the migration of fine particles in the pore structure formed by coarse particles. These are common problems of wide-grade soil mixes. A thorough understanding of the internal erosion development laws and failure mechanisms is of great theoretical and engineering significance for preventing and warning of the stability of slopes, earth-rock dams, and other livelihood projects. Summary of the Invention

[0006] To address the above problems, this invention provides a rapid prediction model for the permeability stability of wide-graded soil, as detailed below:

[0007]

[0008] Where: M fp M represents the cumulative mass of fine particles lost at time t; sfp σ is the total mass of fine particles in the sample; f is the loading frequency; f0 is 1 Hz; σ mean The mean value of the cyclic load; σ0 is 1 kPa; G s denoted as soil specific gravity; n as porosity; i as hydraulic gradient; i0 as critical hydraulic gradient; t as loading time; t0 as unit time, taking a value of 10 min; k1, k2, k3, k4, k5 as model parameters; e as a mathematical constant, 2.718.

[0009] Take the cumulative mass of fine particles lost at t = 120 min as M fp Calculate the fine particle loss rate: Fine particle loss rate = M fp / M sfp *100%

[0010] When the fine particle loss rate is less than or equal to 3%, the permeability stability of the wide-graded soil is classified as stable.

[0011] When the fine particle loss rate is greater than 3% and less than or equal to 20%, the permeability stability of wide-graded soil is classified as transitional.

[0012] When the fine particle loss rate is greater than 20%, the permeability stability of the wide-graded soil is classified as unstable.

[0013] This application also provides a method for constructing the prediction model of claim 1, comprising the following steps:

[0014] (1) Obtain the basic physical performance parameters of the test width graded soil sample through basic physical performance test; then, fill the sample model tube according to the subgrade layering in actual engineering. Before filling, each layer of soil sample is dyed with a different color by dyeing agent.

[0015] (2) Saturate the soil sample;

[0016] (3) After saturation, the test soil sample is loaded using a static dynamic loading system according to the pre-set static dynamic loading sequence, loading waveform, loading stress and loading frequency; during the static dynamic loading process, the hydraulic gradient is adjusted by the hydraulic control system to obtain the effect of the hydraulic gradient on the soil sample permeability.

[0017] (4) Source quantification analysis of particle loss

[0018] During the experiment, a particle collection system was used to collect the lost fine particles. After the experiment, the overall particle loss results were analyzed, and then particles of different colors were separated to trace the source of the lost particles in each layer of the sample. The soil samples before and after the fine particle loss were compared to qualitatively and quantitatively analyze the particle loss caused by the suction effect of pore water pressure under the action of water-mechanical coupling. A model for predicting the mass of lost fine particles was established that comprehensively considers the physical state, loading state and environmental state.

[0019] Based on the above scheme, the sample model cylinder in step (1) includes a cylinder body, and the cylinder body consists of, from bottom to top, a water inlet layer, a water inlet layer partition, several layers of sample soil, a coarse gravel layer partition, and a coarse gravel layer; the water inlet layer is a cavity formed by the water inlet layer partition and the wall of the cylinder body; the water inlet partition and the coarse gravel layer partition are both porous plates.

[0020] The lower part of the cylinder is provided with a cylinder inlet for conveying water into the water inlet layer, and the upper part of the cylinder is provided with a cylinder outlet for discharging water from the cylinder.

[0021] Based on the above scheme, the static dynamic loading system described in step (3) includes a loading plate set above the coarse gravel layer, an electric loading rod connected to the loading plate, and a servo control switch for controlling the electric loading rod.

[0022] Based on the above scheme, the hydraulic control system in step (3) includes a water tank, an inlet pipe for inputting water from the water tank into the inlet layer of the sample model cylinder, and a return pipe for transporting the outflow water from the sample model cylinder back to the water tank; the inlet pipe is equipped with an inlet pipe valve; the return pipe is equipped with a return pipe valve and a water pump; the hydraulic control system also includes a height adjustment track for adjusting the relative head difference during the test, and the water tank can be adjusted up and down along the track to obtain different relative head differences.

[0023] Based on the above scheme, the particle collection system in step (4) includes a three-way connecting pipe set on the return water pipe and a collection pot connected to the three-way connecting pipe for collecting soil particles lost from the test soil sample during the test; the inlet and outlet of the three-way connecting pipe are connected to the return water pipe; the connection between the collection pot and the three-way connecting pipe is detachable; the three-way connecting pipe forms a bend between the inlet and outlet for changing the direction of water flow; the middle part of the bend protrudes towards one side of the collection pot.

[0024] Based on the above scheme, the collecting vessel has an enlarged section on the side away from the three-way connecting pipe; the enlarged section is provided with several brushes for suppressing the dispersion of fine particles.

[0025] Based on the above scheme, the particle collection system includes at least two of the above-mentioned three-way connecting pipes and at least two collection containers.

[0026] Based on the above scheme, when the soil sample is saturated in step (2), the water in the water tank is input into the water inlet layer of the sample model tube through the water inlet pipe. The water passes through the water inlet layer partition, several layers of sample soil, coarse gravel layer partition and coarse gravel layer from bottom to top, and finally is transported to the return water pipe from the water outlet on the side wall corresponding to the coarse gravel layer. Finally, under the action of the water pump, the water flows back to the water tank.

[0027] Based on the above scheme, when analyzing the lost particles in step (4), an industrial camera and PCAS software are used to perform color recognition and trace the lost particles.

[0028] This application also provides a model for predicting the permeability stability of a wide-graded soil with fixed physical properties, the specific physical properties being: coarse aggregate with a particle size of 2-10 mm crushed stone, and fine filler material with a particle size of 0.075-1 mm.

[0029] The formula for the cumulative mass of fine particles lost at time t in the model is as follows:

[0030]

[0031] Where: M fp M represents the cumulative mass of fine particles lost at time t; sfp σ is the total mass of fine particles in the sample; f is the loading frequency; f0 is 1 Hz; σ mean The mean value of the cyclic load; σ0 is 1 kPa; G s ρ is the soil specific gravity; n is the porosity; i is the hydraulic gradient; i0 is the critical hydraulic gradient; t is the loading time; t0 is the unit time, taken as 10 min; k1 = 2 * 10 -3 k2 = 1.43 × 10 -3k3 = 0.153, k4 = 3.5 × 10 -5 k5 = 4.3; e is a mathematical constant, 2.718.

[0032] The beneficial effects of this invention are:

[0033] (1) The model and model construction method of this invention comprehensively consider the influence of physical state (porosity, specific gravity), loading state (loading stress, loading frequency, loading time) and environmental state (hydraulic gradient) on the stability type of wide-graded soil. The fine particle loss rate model for wide-graded soil constructed by this invention has clear physical meaning and simple structure, which greatly reduces the test time and the test difficulty. It provides significant engineering convenience for wide-graded soil with various gradation combinations and for field units lacking test conditions, and has high market promotion value.

[0034] (2) In the process of model construction, the present invention uses a special device for permeability stability test. This device enables the method of the present invention to carry out model test research on the phenomenon of particle loss in the model subgrade under various combinations of static and dynamic loads, soil properties and moisture conditions. It can simulate different static and dynamic load sizes, frequencies and forms, and test soil samples with various particle sizes, gradations and densities and saturations. Attached Figure Description

[0035] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0036] Figure 1 This is a schematic diagram of the overall structure of the device in Embodiment 1 of this application;

[0037] Figure 2 This is a schematic diagram of the sample model cylinder in the device of this application;

[0038] Figure 3 This is a schematic diagram of the structure of the collecting vessel in the device of this application;

[0039] Figure 4 This is one embodiment of the particle collection system in the device of this application;

[0040] Figure 5 This is a schematic diagram of the cyclic load in step (3) of Embodiment 2 of this application;

[0041] Figure 6 This is a schematic diagram of the experimental hydraulic gradient in step (3) of Embodiment 2 of this application;

[0042] Figure 7 This is a graph showing the cumulative mass loss of particles in the sample at different times in step (4) of Example 2;

[0043] Figure 8 The composition of particles lost from the graded sample at time discontinuity i=1 under cyclic loading is given.

[0044] Figure 9 The composition of particles lost from the graded sample under cyclic loading with hydraulic gradient i = 1.5 at a time interval is given.

[0045] Figure 10 Particle composition lost from graded samples under cyclic loading with hydraulic gradient i = 2.5 over a time interval;

[0046] Figure 11 Particle composition lost from graded samples with hydraulic gradient i=3 under cyclic loading;

[0047] Figure 12 This is a robustness verification diagram from Example 4;

[0048] Figure 13 The results of the literature experiment gradation discrimination are based on common discrimination criteria. Detailed Implementation

[0049] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0050] Example 1

[0051] This embodiment provides an apparatus for a permeability stability test, comprising:

[0052] The sample model cylinder 1 is used to fill multiple layers of sample material, each layer of sample material being dyed a different color.

[0053] Hydraulic control system 2 is used to provide circulating water flow and test water pressure for the specimen inside the specimen model cylinder 1;

[0054] A static dynamic loading system 3 used to provide pressure to the specimen inside the specimen model cylinder 1;

[0055] The data detection and acquisition system 4 is used to collect data on the pressure and pore water pressure of the sample inside the sample model cylinder 1.

[0056] A particle collection system 5 is used to collect particles lost from the sample model cylinder 1 during the test.

[0057] It also includes a computer 7 for monitoring and storing data collected by the data detection and acquisition system 4, and the computer 7 is also used to control the static dynamic loading system 3.

[0058] Using the aforementioned apparatus, permeability stability tests can be conducted on materials according to pre-set test items. For the different colored loss of sample material particles collected by the particle collection system 5, they can be distinguished by color, thereby obtaining the loss status of different sample layers.

[0059] Distinguishing between different colored sample particles using traditional manual methods would be extremely labor-intensive. Therefore, this application provides a solution that significantly reduces the workload of distinguishing different colored particles. Specifically, the apparatus for the permeability stability test of this application further includes an image acquisition and analysis system 6 for color identification of sample particles lost during the test. More specifically, the image acquisition and analysis system 6 includes a camera for acquiring images (specifically, an industrial camera can be used) and analysis software (such as PCAS software) for color analysis and identification of the images.

[0060] As a specific implementation scheme, the sample model cylinder 1 includes a cylinder body 1-1, and inside the cylinder body 1-1, from bottom to top, there are a water inlet layer 1-2, a water inlet layer partition 1-3, and several layers of sample soil 1-4. Figure 2 The diagram shows the state of four layers of samples, including coarse gravel layer partition 1-5 and coarse gravel layer 1-5; the water inlet layer 1-2 is the cavity formed by water inlet layer partition 1-3 and the wall of cylinder 1-1; both water inlet partition 1-3 and coarse gravel layer partition 1-5 are porous plates. The material of cylinder 1-1 can be glass or acrylic.

[0061] The wall of the cylinder 1-1 is provided with a number of sensor channels 1-11. Figure 1 Due to space limitations, only one row of sensor channels is shown in the image.

[0062] The lower part of the cylinder 1-1 is provided with a cylinder inlet 1-12 for conveying water into the water inlet layer 1-2, and the upper part of the cylinder 1-1 is provided with a cylinder outlet 1-13 for discharging water from the cylinder 1-1, specifically located on the side wall corresponding to the coarse gravel layer 1-5.

[0063] The hydraulic control system 2 includes a water tank 2-1, an inlet pipe 2-2 for inputting water from the water tank 2-1 into the water inlet layer 1-2 of the sample model cylinder 1, and a return pipe 2-3 for transporting the water from the sample model cylinder 1 back to the water tank 2-1; an inlet pipe valve 2-4 is provided on the inlet pipe 2-2; and a return pipe valve 2-5 and a water pump 2-6 are provided on the return pipe 2-3.

[0064] As a preferred embodiment, the hydraulic control system 2 further includes a height adjustment track 2-7 for adjusting the relative head difference during the test. The water tank 2-1 can be adjusted up and down along the track 2-7 to obtain different relative head differences.

[0065] As a preferred embodiment, the static dynamic loading system 3 includes a loading plate 3-1 disposed above the coarse gravel layer 1-5, an electric loading rod 3-2 connected to the loading plate 3-1, and a servo control switch 3-3 for controlling the electric loading rod 3-2.

[0066] As a preferred embodiment, the data detection and acquisition system 4 includes a soil pressure cell sensor 4-1 and a pore water pressure sensor 4-2 inserted through sensor channels 1-11 between the water inlet layer partition 1-3 and the sample soil layer 1-4 and / or between adjacent sample soil layers 1-4, and a data acquisition box 4-3 for acquiring information from the soil pressure cell sensor 4-1 and the pore water pressure sensor 4-2; both sensors are commercially available conventional products, and will not be described in detail here.

[0067] Traditional particle collection systems typically use filter funnels for particle collection. Because particles are very small (less than 1 mm), some small particles enter the filter media (commonly filter screens and filter cotton) during collection. These particles are difficult to remove during statistical analysis, affecting the final experimental results. Furthermore, particles clog the filter media, causing the water flow to slow down as the experiment progresses. This leads to a buildup of unfiltered water in the funnel, and some of this accumulated water may even overflow, carrying particles and causing the experiment to fail. Even using a larger funnel before the experiment does not solve the slow filtration problem. To address this issue, this application provides a particle collection system with improved filtration efficiency, such as... Figure 3 As shown, the particle collection system 5 includes a three-way connecting pipe 5-1 installed on the return water pipe 2-3 and a collection container 5-2 connected to the three-way connecting pipe 5-1 for collecting soil particles lost from the test soil sample during the test; more specifically, the inlet 5-11 and outlet 5-12 of the three-way connecting pipe 5-1 are connected to the return water pipe 2-3; the collection container 5-2 is connected to the connection port 5-13 of the three-way connecting pipe 5-1 (specifically, a threaded connection);

[0068] In a typical three-way connector 5-1, the section between the inlet 5-11 and the outlet 5-12 is a straight pipe. With this type of straight pipe, water carrying fine particles enters directly from the inlet 5-11, and the fine particles are carried away by the water and flow out from the outlet 5-12 before they have a chance to settle, thus affecting the collection effect of fine particles. To solve this technical problem, the three-way connector 5-1 of this application forms a bent section 5-14 between the inlet 5-11 and the outlet 5-12 to change the direction of water flow; Figure 3 As shown, specifically, the middle part of the curved pipe section (5-14) protrudes towards one side of the collecting vessel 5-2. Therefore, when the water flows into the inlet 5-11, it will change direction at the curved pipe section 5-14 and flow towards the collecting vessel 5-2, which is conducive to the settling of fine particles in the collecting vessel 5-2.

[0069] The collecting vessel 5-2 has an enlarged section 5-21 on the side away from the three-way connecting pipe 5-1; the enlarged section 5-21 is provided with a number of brushes 5-22 for suppressing the dispersion of fine particles; the brushes 5-22 can intercept the fine particles carried in the water and make them settle in the enlarged section 5-21. At the same time, the brushes 5-22 can also effectively reduce the disturbance of the water flow to the water in the enlarged section 5-21, which is conducive to the settling and stabilization of fine particles.

[0070] Although most of the fine particles carried in the water will settle in the collection vessel 5-2 due to gravity and the action of the brush 5-22, it is unavoidable that a small number of fine particles will still be carried in the water flowing out of the outlet 5-12, which will affect the accuracy of the final test. Therefore, based on the above technical solution, the particle collection system 5 of this application also includes a fine particle filter 5-3 installed on the outlet 5-12. The mesh size of the fine particle filter 5-3 is customizable, so that only water can pass through and fine particles cannot. As a specific implementation, the mesh size of the fine particle filter 5-3 in this embodiment is 1000 mesh.

[0071] When conducting permeability stability tests, it is necessary to collect fine particles lost from soil samples at different time points. To complete this test, the flowing water must be shut off during the experiment to clear away the fine particles collected in the previous time point before restarting the test. This method not only significantly prolongs the test time but also affects the continuity of the simulation. Therefore, to continuously test the particle loss of soil samples at different time points during the experiment, such as... Figure 4As shown, this application provides a specific implementation of a particle collection system 5. Based on the above-described particle collection system 5, the new particle collection system 5 includes at least two of the aforementioned three-way connecting pipes 5-1 and at least two collection containers 5-2 (the three-way connecting pipes 5-1 and the collection containers 5-2 correspond one-to-one). In use, according to the experimental design, when a collection container 5-2 has finished collecting, the valve on the three-way connecting pipe 5-1 corresponding to that collection container 5-2 is closed. Figure 4 (The number will not be repeated in the original text). At the same time, open the valve on the three-way connecting pipe 5-1 corresponding to the other empty collection container 5-2. Then, remove the collection container 5-2 that has already collected the particles, take out the collected particles, clean the collection container 5-2, and finally reinstall it.

[0072] The test apparatus in this embodiment can be used for soil permeability stability tests, such as permeability stability tests of wide-graded soil, tests to determine the cumulative loss of fine particles in wide-graded soil, and to determine relevant parameters when constructing a model for predicting the permeability stability of wide-graded soil.

[0073] Example 2

[0074] Based on the apparatus used for the permeability stability test in Example 1, this application provides a method for testing the permeability stability of wide-graded soil, specifically including the following steps:

[0075] (1) The basic physical performance parameters of the wide-graded soil sample were obtained through a series of basic physical performance tests, such as compaction tests and sieve tests. (Specifically, in this embodiment, wide-graded soil samples from Changsha were selected. After testing, the basic physical performance parameters of the soil sample were as follows: coarse aggregate with a particle size of 2-10 mm crushed stone, fine filler material with a particle size of 0.075-1 mm, and a maximum dry density of 2.11 g / cm³.) 3 The optimal moisture content is 7.4%. Then, soil samples from several layers of soil (1-4) are dyed with different colors. The dyed soil samples are filled into cylinder 1-1, and soil pressure sensor 4-1 and pore water pressure sensor 4-2 are installed between the water inlet partition 1-3 and soil sample layer 1-4, as well as between adjacent soil sample layers 1-4, through sensor channels 1-11. Specifically, the filling is carried out under the conditions of optimal moisture content and a relative density of 0.9.

[0076] In actual testing, the subgrade was constructed using a layered filling method similar to that used in engineering projects. This embodiment employed four soil sample layers, and German SINO-401 dye was used to color and mark the different soil particles, from bottom to top: red, blue, yellow, and green. Figure 2 As shown.

[0077] During the experiment, soil pressure cell sensor 4-1 and pore water pressure sensor 4-2 were set up to measure the changes in internal stress and pore water pressure of the sample during loading, in order to explain the mechanism of the change in stability type.

[0078] (2) Saturate the soil sample

[0079] The saturation process is as follows: Water in water tank 2-1 is input into the water inlet layer 1-2 of the sample model cylinder 1 through the water inlet pipe 2-2. The water passes through the water inlet layer partition 1-3, several layers of sample soil 1-4, coarse gravel layer partition 1-5 and coarse gravel layer 1-5 from bottom to top. Finally, it is transported to the return water pipe 2-3 from the water outlet 1-13 on the side wall corresponding to the coarse gravel layer 1-5. Finally, under the action of water pump 2-6, the water flows back to water tank 2-1.

[0080] Specifically, the saturation time is 24 hours. At saturation, the water level in the tank should be slightly higher than the sample (i.e., the head difference should not be too large).

[0081] (3) After saturation is completed, the static and dynamic loading sequence, loading waveform, loading stress and loading frequency are set by computer 7, and the electric loading rod 3-2 and loading plate 3-1 are controlled by servo control switch 3-3 to perform loading;

[0082] Specifically, the loading diagram for axial load and loading frequency is as follows: Figure 5 As shown;

[0083] Specifically, during the static-dynamic loading process, the height of the water tank 2-1 is adjusted via the height adjustment track 2-7 to account for the influence of the hydraulic gradient on the permeability (internal erosion) of the test soil sample. Figure 6 As shown;

[0084] Table 1. Particle Migration Test Protocol

[0085]

[0086] (4) Source quantification analysis of particle loss

[0087] After the experiment, the particles collected in collection vessel 5-1 were taken out for overall particle loss analysis. The results are as follows: Figure 7 As shown. Then, by separating the particles of different colors, the source of any lost particles in each sample layer can be traced.

[0088] Considering the small size and large number of soil particles, manual sorting for source tracing and quantitative analysis of particle loss from different soil sample layers presents certain difficulties. As a preferred implementation, this invention uses an industrial camera and PCAS software to identify the color of lost particles. The specific process is as follows: The captured image is binarized, and the area of ​​particles of different colors in the image is statistically analyzed. It is approximated that the ratio of the area of ​​each color particle is equal to the ratio of their masses. The mass of each color particle is calculated based on the total mass, allowing for source tracing (determining which soil sample layer the lost particles originated from) and quantitative analysis. Soil samples before and after fine particle loss are compared to qualitatively and quantitatively analyze the particle loss caused by the suction effect of pore water pressure under water-mechanical coupling. The results are as follows: Figure 8-11 As shown.

[0089] The experimental data obtained above can be used to establish a model for predicting the mass of lost fine particles that comprehensively considers physical conditions (porosity, specific gravity), loading conditions (average cyclic load, loading frequency, loading time) and environmental conditions (hydraulic gradient).

[0090] Example 3

[0091] Based on the apparatus of Embodiment 1 and the method of Embodiment 2, this application provides an implementation method for constructing a permeability stability prediction model for wide-graded soil, specifically including the following steps:

[0092] (1) The basic physical performance parameters of the test wide-grade soil sample were obtained through basic physical performance tests; then, the soil sample was filled in the sample model tube 1 according to the roadbed layering in actual engineering. Before filling, the soil sample of each layer was dyed with a different color by dyeing agent.

[0093] (2) Saturate the soil sample;

[0094] (3) After saturation is completed, the static dynamic loading system 3 is used to load the test soil sample according to the pre-set static dynamic loading sequence, loading waveform, loading stress and loading frequency; during the static dynamic loading process, the hydraulic control system 2 is used to adjust different relative head differences to obtain the influence of hydraulic gradient on soil sample permeability;

[0095] (4) Source quantification analysis of particle loss

[0096] During the experiment, the lost fine particles were collected using particle collection system 5. After the experiment, the overall particle loss results were analyzed, and particles of different colors were separated to trace the source of lost particles in each layer of the sample. The soil samples before and after the fine particle loss were compared to qualitatively and quantitatively analyze the particle loss caused by the suction effect of pore water pressure under the action of water-mechanical coupling. This analysis was then used to establish a model for predicting the mass of lost fine particles that comprehensively considers physical conditions (porosity, specific gravity), loading conditions (average cyclic load, loading frequency, loading time), and environmental conditions (hydraulic gradient).

[0097] Example 4

[0098] Based on the apparatus of Example 1 and the methods of Examples 2 and 3, this application provides a rapid prediction method for the permeability stability of wide-graded soil, as detailed below:

[0099] 1) Establish a predictive model for the mass of lost fine particles.

[0100] based on Figure 7 Based on the experimental data patterns shown, a model for predicting the mass of lost fine particles was established, comprehensively considering physical conditions (porosity, specific gravity), loading conditions (average cyclic load, loading frequency, loading time), and environmental conditions (hydraulic gradient), as shown in equations (1)-(3). Simultaneously, the model parameters k1, k2, k3, k4, and k5 were obtained by fitting the experimental data using the stepwise regression analysis method in SAS software, as shown in Table 2. This fitting method is a conventional method and will not be elaborated further here.

[0101] Among them, f0, σ0, and t0 can be dimensionalized for loading frequency, average cyclic load, and loading time, respectively.

[0102] Among them, the critical hydraulic gradient i0 is related to the physical state of the sample itself (porosity, specific gravity), and the relationship is shown in equation (2).

[0103]

[0104] i0=(G s -1)(1-n) (2)

[0105]

[0106] Where: M fp M represents the cumulative mass of fine particles lost at time t; sfp σ is the total mass of fine particles in the sample; f is the loading frequency; f0 is 1 Hz; σ mean The mean value of the cyclic load; σ0 is 1 kPa; G sdenoted as soil specific gravity; n as porosity; i as hydraulic gradient; i0 as critical hydraulic gradient; t as loading time; t0 as unit time, with a value of 10 min; k1, k2, k3, k4, and k5 are model parameters.

[0107] Table 2. Model parameter fitting results

[0108]

[0109] As can be seen from the above formula, formula (3) can simulate the cumulative loss of fine particles at any time. In the actual prediction of the permeability stability of wide-graded soil, t = 120 min (120 min is the time when the test ends).

[0110] To determine the applicability of the loss fine particle mass prediction model established in this invention, the robustness of the proposed model was verified using wide-graded soil as shown in Table 3. A scatter plot of robustness verification was plotted with the measured loss fine particle mass as the x-axis and the predicted mass as the y-axis. The results are as follows: Figure X As shown in the figure, it can be seen that most of the scattered points are concentrated around the line y = x, R 2 =0.95, indicating excellent fitting effect. Therefore, the model established in this invention has high accuracy and applicability, and can be extended to other similar wide-graded soils, as well as the prediction of the loss of fine particles in the same wide-graded soils under other working conditions.

[0111] Table 3. Corresponding operating conditions for robustness verification

[0112]

[0113] 2) Establish criteria for determining stability types

[0114] Existing methods for determining soil stability types often include: First, observing experimental phenomena, i.e., when a sample exhibits phenomena such as piping or soil flow, it can be determined to be in an unstable state; Second, based on Terzaghi's filtration criterion and combined with a large amount of experimental data, stability evaluation indicators are established through particle size distribution curves.

[0115] However, soil stability is not only affected by particle size distribution, but also significantly influenced by loading conditions (mean cyclic load, loading frequency, loading time) and environmental conditions (hydraulic gradient). Therefore, this application proposes a stability type discrimination criterion (i.e., determining the stability type by collecting lost fine particles through internal erosion tests and determining the ratio of the cumulative lost mass of fine particles to the total mass of fine particles), and conducts a survey of 21 groups (from...). Figure 13 (5 articles) Experimental data on fine particle loss quality and existing 7 discrimination criteria (Istomina) [6] Guidelines, Kezid [7]Guidelines, Kenney [8] Guidelines, Burenkova [9] Guidelines, Fell

[10] Guidelines, Li

[11] Guidelines, Chang

[12] The criteria are used to support this, and the specific results are as follows: Figure 13 As shown.

[0116] Specifically, this application defines particles smaller than 1 mm as fine particles, and M... fp (Cumulative loss of fine particles) and M sfp The ratio of (total mass of fine particles in the sample) is defined as the percentage of lost fine particles.

[0117] like Figure 13 As shown, in 7 groups of samples, the proportion of fine particle loss exceeded 20% of the total fine particle mass, with a total of 49 discrimination tests conducted and black stability observed 13 times. In 9 groups of samples, the proportion of fine particle loss was between 3% and 20%, with a total of 63 discrimination tests conducted and black stability observed 29 times. In 5 groups of samples, the proportion of fine particle loss was less than 3%, with a total of 35 discrimination tests conducted and black stability observed 30 times. When the fine particle loss in the soil exceeds 20%, the soil structure has been damaged, and classifying it as stable is unreasonable. When the proportion of fine particle loss is between 3% and 20%, the Kezid, Fell, and Burenkova discrimination criteria differ significantly from other criteria, and the results of different discrimination methods for the same gradation show considerable differences. When the proportion of fine particle loss is between 0% and 3%, the results of each discrimination method are highly consistent. Therefore, based on the proportion of fine particle loss, such as... Figure 4 As shown, the soil mass is classified into three parts: the proportion of fine particle loss is less than or equal to 3% for the stable type, the proportion of fine particle loss is greater than 3% but less than or equal to 20% for the transitional type, and the proportion of fine particle loss is greater than 20% for the unstable type.

[0118] Table 4 Stability Judgment Criteria

[0119] type Stablize transition Unstable Fine particle loss rate / % 0-3％ 3％-20％ More than 20%

[0120] In summary, by combining the loss fine particle mass prediction model established in 1) and the stability type discrimination criterion established in 2), the permeability stability of wide-graded soil under different conditions can be quickly predicted.

[0121] Figure 13 The references involved are as follows (of which, references 1-5 are...). Figure 13 The validation literature (references 6-12) provides the sources for the seven discrimination criteria.

[0122] [1] Liang Li, Tian Dalang, Ning Yue, et al. Internal erosion test and stability judgment of wide-graded coarse-grained soil [J]. Journal of Underground Space and Engineering, 2020, 16(3):835-843.

[0123] [2]KE L,TAKAHASHI A.Experimental investigations on suffusioncharacteristics and its mechanical consequences on saturated cohesionlesssoil[J].Soils and Foundations,2014,54(4):713-730.

[0124] [3] Dai Shaoheng, Zhang Sheng, Tong Chenxi, et al. Study on internal erosion test of dyed calibrated sand under cyclic loading [J]. Chinese Journal of Rock Mechanics and Engineering, 2022, 41(2):423-432.

[0125] [4]ISRAR J,INDRARATNA B.Internal stability of granular filters understatic and cyclic loading[J].Journal of Geotechnical and GeoenvironmentalEngineering,2017,143(6):04017012.

[0126] [5]INDRARATNA B,ISRAR J,RUJIKIATKAMJORN C.Geometrical method forevaluating the internal instability of granular filters based on constrictionsize distribution[J].Journal of Geotechnical and GeoenvironmentalEngineering,2015,141(10):04015045.

[0127] The relevant literature for the discrimination criteria is as follows:

[0128] [6]ISTOMINA V S.Filtration stability of soils[M].Moscow:Gostroizdat,1957:155-172.

[0129] [7]KEZDI A.Soil physics:selected topics[M].New York,Amsterdam:Elsevier Science,1979:126-133.

[0130] [8]KENNEY T C,LAU D.Internal stability of granular filters[J].Canadian Geotechnical Journal,1985,22(2):215-225.

[0131] [9]BURENKOVA V V.Assessment of suffusion in non-cohesive and gradedsoils[J].Filters in Geotechnical and Hydraulic Engineering.Balkema,Rotterdam,1993:357-360.

[0132]

[10] FELL R,WAN C F,CYGANIEWICZ J,et al.Discussion of time fordevelopment ofinternal erosion and piping in embankment dams by robin fell,chi fai wan,johncyganiewicz,and mark foster[J].Journal of Geotechnical andGeoenvironmentalEngineering,2004,130(9):980-981.

[0133]

[11] LI M,FANNIN R J.Comparison of two criteria for internal stabilityof granular soil[J].

[0134] Canadian Geotechnical Journal,2008,45(9):1303-1309.

[0135]

[12] CHANG D,ZHANG L.Extended internal stability criteria for soilsunder seepage[J].Soils and Foundations,2013,53(4):569-583.

[0136] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on its differences from other embodiments. The above descriptions are merely preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention are included within the scope of protection of the present invention.

Claims

1. A method for determining the seepage stability of wide-graded soil using a predictive model, characterized in that, The model formula is as follows: In the formula: Let be the cumulative mass of fine particles lost at time t; This represents the total mass of fine particles in the sample. Loading frequency; 1Hz; This is the average value of the cyclic load; 1 kPa; This refers to the specific gravity of the soil. Porosity; For hydraulic gradient; The critical hydraulic gradient; Loading time; The unit time is 10 min; k1, k2, k3, k4, and k5 are model parameters; e is a mathematical constant, 2.

718. Take the cumulative mass of fine particles lost at t=120min Calculate the fine particle loss rate, fine particle loss rate = / *100% When the fine particle loss rate is less than or equal to 3%, the permeability stability of the wide-graded soil is classified as stable. When the fine particle loss rate is greater than 3% and less than or equal to 20%, the permeability stability of wide-graded soil is classified as transitional. When the fine particle loss rate is greater than 20%, the permeability stability of the wide-graded soil is classified as unstable.

2. A method for constructing a rapid prediction model for the permeability stability of wide-graded soil, characterized in that, Includes the following steps: (1) Obtain the basic physical performance parameters of the test wide-grade soil sample through basic physical performance test; then, fill the sample model tube according to the subgrade layering in actual engineering. Before filling, each layer of soil sample is dyed with a different color. (2) Saturate the soil sample; (3) After saturation, the test soil sample is loaded using a static dynamic loading system according to the pre-set static dynamic loading sequence, loading waveform, loading stress and loading frequency; during the static dynamic loading process, the hydraulic gradient is adjusted by the hydraulic control system to obtain the effect of the hydraulic gradient on the soil sample permeability. (4) Quantitative analysis of the source of particle loss During the experiment, a particle collection system was used to collect the lost fine particles. After the experiment, the overall particle loss results were analyzed, and then particles of different colors were separated to trace the source of the lost particles in each layer of the sample. The soil samples before and after the fine particle loss were compared to qualitatively and quantitatively analyze the particle loss caused by the suction effect of pore water pressure under the action of water-mechanical coupling. A model for predicting the mass of lost fine particles was established that comprehensively considers the physical state, loading state and environmental state. The wide-graded soil mix consists of coarse aggregate with a particle size of 2-10mm crushed stone and fine aggregate with a particle size of 0.075-1mm. The formula for the cumulative mass of fine particles lost at time t in the model is as follows: In the formula: Let be the cumulative mass of fine particles lost at time t; This represents the total mass of fine particles in the sample. Loading frequency; 1Hz; This is the average value of the cyclic load; 1 kPa; This refers to the specific gravity of the soil. Porosity; For hydraulic gradient; The critical hydraulic gradient; Loading time; The unit of time is 10 minutes; k1 = 2 * 10 -3 k2 = 1.43 × 10 -3 k3 = 0.153, k4 = 3.5 × 10 -5 k5 = 4.3; e is a mathematical constant, 2.

718.

3. The method for constructing a rapid prediction model for the permeability stability of wide-graded soil according to claim 2, characterized in that, The sample model cylinder mentioned in step (1) includes a cylinder body, and the cylinder body consists of, from bottom to top, a water inlet layer, a water inlet layer partition, several layers of sample soil, a coarse gravel layer partition, and a coarse gravel layer; the water inlet layer is a cavity formed by the water inlet layer partition and the wall of the cylinder body; the water inlet layer partition and the coarse gravel layer partition are both porous plates. The lower part of the cylinder is provided with a cylinder inlet for conveying water into the water inlet layer, and the upper part of the cylinder is provided with a cylinder outlet for discharging water from the cylinder.

4. The method for constructing a rapid prediction model for the permeability stability of wide-graded soil according to claim 3, characterized in that, The static dynamic loading system described in step (3) includes a loading plate set above the coarse gravel layer, an electric loading rod connected to the loading plate, and a servo control switch for controlling the electric loading rod.

5. The method for constructing a rapid prediction model for the permeability stability of wide-graded soil according to claim 3, characterized in that, The hydraulic control system described in step (3) includes a water tank, an inlet pipe for inputting water from the water tank into the inlet layer of the sample model cylinder, and a return pipe for transporting the outflow water from the sample model cylinder back to the water tank; the inlet pipe is equipped with an inlet pipe valve; the return pipe is equipped with a return pipe valve and a water pump; the hydraulic control system also includes a height adjustment track for adjusting the relative head difference during the test, and the water tank can be adjusted up and down along the track to obtain different relative head differences.

6. The method for constructing a rapid prediction model for the permeability stability of wide-graded soil according to claim 3, characterized in that, The particle collection system described in step (4) includes a three-way connecting pipe installed on the return water pipe and a collection container connected to the three-way connecting pipe for collecting soil particles lost from the test soil sample during the test; the inlet and outlet of the three-way connecting pipe are connected to the return water pipe; the connection between the collection container and the three-way connecting pipe is detachable. The three-way connecting pipe forms a curved section between the inlet and outlet to change the direction of water flow; the middle part of the curved section protrudes towards one side of the collecting container.

7. The method for constructing a rapid prediction model for the permeability stability of wide-graded soil according to claim 6, characterized in that, The collecting vessel has an enlarged section on the side away from the three-way connecting pipe; the enlarged section is provided with several brushes for suppressing the dispersion of fine particles.

8. The method for constructing a rapid prediction model for the permeability stability of wide-graded soil according to claim 6, characterized in that, The particle collection system includes at least two of the aforementioned three-way connecting pipes and at least two collection containers.

9. The method for constructing a rapid prediction model for the permeability stability of wide-graded soil according to claim 2, characterized in that, In step (2), when the soil sample is saturated, the water in the tank is introduced into the water inlet layer of the sample model tube through the water inlet pipe. The water passes through the water inlet layer partition, several layers of sample soil, coarse gravel layer partition and coarse gravel layer from bottom to top, and is finally transported to the return water pipe from the water outlet on the side wall corresponding to the coarse gravel layer. Finally, under the action of the water pump, the water flows back to the tank.

Images