A test device and method for simulating continuous cyclic hydraulic gradient erosion of a graded soil

Abstract

This invention discloses an experimental apparatus and method for simulating continuous circulating hydraulic gradient erosion of graded soil. The apparatus includes a pneumatically controlled circulating hydraulic gradient loading system, a data monitoring system, and a multi-stage sieving and graded collection system. A pneumatic controller adjusts the gas phase pressure at the top of a sealed water tank, driving high-pressure water to flow upwards through the sample cylinder, forming an upward seepage. A pore pressure sensor captures the internal water head dynamics in real time. The soil-water mixture enters a subsequent multi-stage series sieving device to achieve adaptive particle separation. Combined with a real-time conversion formula based on Archimedes' law of buoyancy, the loss mass of pure dry soil at each particle size range is directly output online in milliseconds without interrupting the experiment, achieving full-time, non-destructive, and digital monitoring of particle migration. The experimental apparatus and method of this invention can realize continuous circulating hydraulic gradient loading, pore pressure feedback closed-loop correction, stratified sieving of seeping particles, graded collection, and online mass conversion.

Description

Technical Field

[0001] This invention belongs to the field of physical simulation test technology for landslide geological hazards and geotechnical engineering, specifically relating to a test device and method for simulating continuous circulating hydraulic gradient erosion graded soil. Background Technology

[0002] Rainfall-induced landslides, reservoir bank slope deformation, and repeated fluctuations in groundwater levels are all closely related to the dynamic changes in the seepage field within the soil. Under conditions of alternating rainfall and cessation of rainfall, fluctuations in groundwater levels, or disturbances caused by engineering drainage, the hydraulic gradient within the slope often exhibits a continuous and smooth change, accompanied by periodic loading and unloading. Such dynamic changes in the hydraulic gradient can induce the initiation, migration, deposition, or loss of fine particles within the soil, leading to pore structure reconstruction, changes in permeability, and adjustments in effective stress, ultimately resulting in the disintegration of the slope framework or instability of the soil structure.

[0003] However, existing seepage erosion testing equipment suffers from the following serious technical defects when simulating the aforementioned complex mechanical processes:

[0004] First, the existing head loading methods are outdated and lack real-time error correction. For example, application number 202510006659.6, entitled "Experimental Device and Method for Simulating Erosion Inside Seawall Soil under Circulating Hydraulic Gradient," uses a screw-slider to raise and lower a water tank. Mechanical raising and lowering suffers from severe motion lag and inertial vibration disturbances, making it difficult to achieve the continuous, smooth, and mechanically undisturbed hydraulic gradient changes found in nature. Furthermore, as the experiment progresses, the water level in the tank decreases the gravitational head, and the dynamic reconstruction of soil permeability causes complex resistance fluctuations. Because existing devices lack a closed-loop correction mechanism linked to pore water pressure data, there is a systematic deviation between the actual applied hydraulic gradient and the target value. Therefore, there is an urgent need for a closed-loop correction mechanism that can achieve continuous circulating hydraulic gradient loading and perform high-frequency feedback fine-tuning based on pore pressure data.

[0005] Secondly, unreasonable water flow design leads to distortion of soil stress state and induces self-clogging. For example, application number 202310486580.9, entitled "A Triaxial Percolation Test Device and Method for Crushed Stone Soil Acoustic Emission," adopts a top-down downward flow design. The permeability of the downward-flowing water is in the same direction as the gravity of the soil particles. During the test, the soil becomes increasingly compacted, failing to accurately simulate the critical point of loosening and failure caused by groundwater backing during slope instability in nature. Furthermore, it easily leads to the accumulation of transported fine particles at the bottom supporting medium, causing severe self-clogging, resulting in obstructed water flow and abnormal test interruption. Moreover, landslides, foundation pit piping, and reservoir bank collapses in nature are largely caused by upward groundwater outflow (backing force). Therefore, there is an urgent need for devices that can truly simulate and study the critical failure characteristics of soil during the process of gravity being offset, and allow fine particles to overflow smoothly under hydrodynamic action, fundamentally avoiding the technical defects of traditional devices that cause test interruption or distortion due to internal self-clogging.

[0006] Third, the methods for measuring leachate particles are outdated and severely disturb the effluent boundary. Current experiments typically rely on manual, periodic sampling, drying, sieving, and weighing to determine the particle size distribution and mass of leachate particles. The resulting data are discrete, making it impossible to obtain a continuous sequence of leachate particle mass changes over time. If a fine, high-resistance filter or multi-stage sieving device is directly installed at the effluent outlet, it easily creates back pressure at the outlet, artificially altering the effluent boundary conditions of the sample cylinder and interfering with the internal flow field. Furthermore, existing multi-stage sieving designs suffer from problems such as unreasonable screen arrangement, mis-discharge of fine particles in the preceding stages, and inaccurate mechanical decoupling between the material retained by each stage of the screen and the collection unit, making high-precision online continuous measurement impossible. Therefore, there is an urgent need for a measurement method that can achieve sequential sieving, graded collection, and online mass conversion of leachate particles without significantly disturbing the effluent boundary of the sample (i.e., normal pressure and normal flow).

[0007] In summary, the development of an experimental device and method that integrates upflow erosion mechanism, continuous circulating hydraulic gradient air pressure loading, pore pressure closed-loop correction, zero effluent disturbance, sequential particle sieving, graded collection, and online continuous mass conversion is a technical bottleneck that urgently needs to be solved in the field of seepage erosion testing in geotechnical engineering. Summary of the Invention

[0008] To address the aforementioned technical problems, this invention provides a test apparatus and method for simulating continuous circulating hydraulic gradient erosion of graded soil. The apparatus achieves programmable continuous circulating hydraulic gradient loading through air-to-water pressure conversion and closed-loop correction through feedback from a pore water pressure sensor. The test employs an upward flow, with the water effluent from the top of the sample undergoing normal pressure seepage. The effluent water-soil mixture is progressively sieved using a three-layer series sieve device (upper, middle, and lower), allowing particles of different sizes to enter their corresponding collection units via clear connecting pipes. Online mass conversion is achieved by combining a full-water constant-volume collection structure and a weighing sensor. The series sieve device and the sample cylinder are separate components, preventing disturbance to the sample's effluent boundary.

[0009] To achieve the above objectives, the technical solution of the present invention is as follows:

[0010] A test apparatus for simulating continuous circulating hydraulic gradient erosion graded soil includes:

[0011] A pneumatic pressure controlled circulating hydraulic gradient loading system includes a computer control terminal, a pneumatic pressure controller electrically connected to the computer control terminal, and a sealed water tank with an air inlet at the top and a water outlet at the bottom; the air inlet is connected to the pneumatic pressure controller via an air pipe.

[0012] The data monitoring system includes a sample tube with a pebble layer, a gravel layer, a metal filter, and the test soil layer laid out sequentially from bottom to top inside; the outlet of the sealed water tank is connected to the inlet at the bottom of the sample tube through a pipeline to form an upward seepage flow; several pore water pressure sensors are buried at different heights of the test soil, and each of the pore water pressure sensors is electrically connected to the computer control terminal.

[0013] A multi-stage sieving and grading collection system includes a series sieving device independently installed behind the water outlet of the sample cylinder and several grading collection units. The top inlet of the series sieving device is connected to the top outlet of the sample cylinder. Inside the device, multiple layers of screens with decreasing apertures are arranged vertically from top to bottom. Each layer of screen has a corresponding screen retention outlet on its side wall. Each screen retention outlet is connected to the corresponding grading collection unit through an independent connecting pipe.

[0014] The system utilizes a pneumatic pressure controller to regulate the gas phase pressure at the top of the sealed water tank. Pressure is transmitted through the gas-water interface, driving high-pressure water to flow upwards through the sample, forming an upward seepage. A pore pressure sensor captures the dynamic water head inside the sample in real time. The water-soil mixture at the outlet enters a separate, downstream sieving unit under gravity and mechanical action. This achieves fully automated linkage of the entire process from loading to monitoring to sieving to collection. The upward seepage avoids interference from gravity on particle initiation, realistically replicating the mechanical boundaries of confined groundwater upwelling or sand inrush at the dam foundation. The pneumatic pressure is easier to precisely control with a computer than the water tank height. The computer controls the pneumatic pressure controller to gradually increase the pneumatic pressure, increasing the input water head and continuously raising the hydraulic gradient; conversely, the computer controls the pneumatic pressure controller to gradually decrease the pneumatic pressure, decreasing the input water head and continuously lowering the hydraulic gradient.

[0015] Specifically, each of the graded collection units adopts a mechanically decoupled, full-water constant-volume collection structure, and each graded collection unit specifically includes:

[0016] External mounting bracket and base; the mounting bracket is mounted on the base;

[0017] A flow guide funnel is rigidly fixed to the top of the external fixing frame, and the top of the flow guide funnel is connected to the corresponding screen.

[0018] An overflow cylinder is suspended below a load cell. The load cell is fixed to an external mounting bracket and electrically connected to a computer control terminal. The load cell does not contact the flow guide funnel. The lower outlet of the flow guide funnel is inserted into the overflow cylinder, and a physical gap is maintained between the flow guide funnel and the inner and outer walls of the overflow cylinder to achieve mechanical decoupling.

[0019] An outer cylinder is placed on the base and located around the overflow cylinder, and the outer cylinder does not contact the overflow cylinder; an overflow port is provided on the side wall of the overflow cylinder, and the overflow port communicates with the inside of the outer cylinder.

[0020] "Mechanical decoupling" cuts off the transmission path of the mechanical impact force of the upstream water flow; "full water constant volume" uses Archimedes' law of buoyancy and the principle of drainage to make the volume of drainage caused by the injected water flow equal to the volume of injection, thus maintaining a constant total mass of liquid in the cylinder.

[0021] The interference of dynamic water flow impact and dynamic head fluctuation on the weighing sensor is eliminated; the reading increment of the weighing sensor is purely contributed by the settled solid soil particles, realizing static precision weighing under flowing water conditions.

[0022] Preferably, each layer of screen in the series screening device is inclined in the same direction; the series screening device is also equipped with a vibration module, the series screening device is supported by a fixed bracket, and a silicone pad is provided at the bottom of the fixed bracket.

[0023] The tilt angle provides the gravitational component of the downward force, while the vibration module provides high-frequency disturbance force, breaking the frictional locking between the particles and the screen. The silicone pad uses a damping effect to intercept the outward diffusion of vibration waves. This ensures that the leached particles do not clog the screen mesh or accumulate and remain there; at the same time, it isolates the vibration from interference with nearby high-precision weighing sensors.

[0024] Preferably, the tandem screening device is provided with an upper screen with a 2 mm aperture, a middle screen with a 0.25 mm aperture, and a lower screen with a 0.075 mm aperture in a vertical direction from top to bottom; the graded collection unit includes a first collection unit corresponding to the upper screen, a second collection unit corresponding to the middle screen, and a third collection unit corresponding to the lower screen.

[0025] According to geotechnical testing standards, these three apertures are the key geometric boundaries for classifying coarse-grained (gravel, coarse sand), medium-fine sand, and fine-grained soil (silt, clay). They enable the precise separation of leached material of different orders of magnitude, providing a refined particle size classification basis for the study of selective erosion in graded soils.

[0026] Preferably, the sample tube is a transparent or cross-sectionable cylindrical container, and its upper and lower ends are sealed by flanges or caps.

[0027] Optically transparent materials provide an intuitive visual window, allowing researchers to observe the macroscopic physical evolution of piping channels, their initiation, development, and skeletal collapse within the soil using the naked eye or in conjunction with a high-speed camera. High-rigidity end seals ensure the system remains leak-proof and pressure-free even under high seepage pressure.

[0028] A test method for simulating continuously circulating hydraulic gradient erosion graded soil includes the following steps:

[0029] Step 1: Test preparation; Design the soil sample gradation, dry density, and moisture content according to the test objective, and calculate the theoretical loading parameters of the target hydraulic gradient; Fill the test soil into the sample tube, and bury several pore water pressure sensors at different heights in the test soil;

[0030] Step 2: Device connection and initialization; complete the physical and signal connections of each device component in the experiment; fill the sealed water tank with water and remove air from the pipeline; add pure water to the overflow cylinder of the graded collection unit used to collect the seepage particles until it just overflows from the overflow port, and after the liquid level has settled and stabilized, record the initial total mass M0 of each graded collection unit filled with water using a weighing sensor.

[0031] Step 3: Start the circulating hydraulic gradient loading; input and configure the theoretical loading parameters obtained in Step 1 into the computer control terminal to generate a digital target loading curve; start the air pressure controller to make the air pressure in the sealed water tank change continuously and smoothly according to the target loading curve, thereby forming a continuous circulating hydraulic gradient at the bottom of the sample tube, driving the seepage from bottom to top through the test soil.

[0032] Step 4: Gradually screen the leached particles; start the vibration module on the computer control terminal and the series screening device. The water-soil mixture discharged from the top outlet of the sample tube enters the series screening device. Under the action of vibration, it is screened step by step through the internal screens with decreasing aperture from top to bottom. The intercepted material and entrained water of each screen are introduced into the corresponding overflow cylinder through the corresponding independent connection channel and guide funnel. Fine particles smaller than the aperture of the bottom screen and wastewater are discharged through the drain outlet at the bottom of the series screening device.

[0033] Step 5: Online continuous mass conversion and dynamic closed-loop correction; During the experiment, the computer control terminal synchronously and frequently collects data from each pore water pressure sensor and real-time mass data from each weighing sensor. ;

[0034] The computer control terminal automatically calculates the actual real-time hydraulic gradient inside the soil based on the real-time water pressure data fed back from each pore water pressure sensor. and will The target hydraulic gradient at this moment A comparison is performed, and if a deviation is found, a control command is issued in real time to adjust the output air pressure of the air pressure controller to achieve dynamic closed-loop correction. At the same time, the computer control terminal calculates the mass of dry soil particles lost in each particle size range online in real time based on the density difference formula. And a continuous curve of the change in dry soil loss mass over time was plotted.

[0035] By synchronizing time, pressure, quality, and control commands across multiple dimensions in space and time, soil erosion is mapped into a high-frequency data stream in real time, and the input air pressure is adjusted using a feedback mechanism. This has resulted in an online, non-destructive quantitative testing method that does not disrupt the continuity of the experiment, solving the problem that traditional methods require mid-test shutdown, disassembly, and drying before measurements can be taken.

[0036] Specifically, the target hydraulic gradient curve mentioned in step one is: i(t) = i0 + A·sin(2πft); where i(t) is the target hydraulic gradient at time t, i0 is the average hydraulic gradient, A is the amplitude, and f is the loading frequency.

[0037] Simulating the periodic oscillation and dynamic excitation characteristics of groundwater in actual engineering projects such as tidal rise and fall, frequent start and stop of pumped storage power stations, and periodic dewatering of foundation pits, it can accurately control the frequency, amplitude, and reference value of circulating water, providing a standard and controllable power source for studying soil erosion under periodic dynamic loads.

[0038] Specifically, in step five, the density difference formula is:

[0039] in: Density of soil particles; The density of water; The difference between the real-time measurement value of the weighing sensor and the initial total mass of the water-filled area.

[0040] This breakthrough enables the direct extraction of dry soil mass while the soil is wet, without the need for offline drying. The data refresh rate can reach the millisecond level, providing metadata for plotting continuous time curves.

[0041] The beneficial effects of this invention are:

[0042] (1) The continuous cyclic hydraulic gradient loading is achieved by using air pressure to water pressure conversion. The target loading curve can be programmed and executed smoothly, which is closer to the hydraulic gradient change process under actual working conditions such as rainfall and groundwater level fluctuation. It can also perform real-time pore pressure feedback closed-loop correction based on the actual water pressure fed back by the sensor inside the sample tube, eliminating the hydraulic gradient distortion caused by soil erosion evolution and sudden changes in porosity, thus improving loading accuracy and test repeatability.

[0043] (2) The present invention absorbs kinetic energy through a rigid and independent guide funnel and adopts a suspended full-water constant-volume overflow cylinder. By utilizing the principle of equal volume drainage, the force on the weighing sensor is not affected by the dynamic water force. This mechanical decoupling design enables the system to maintain ultra-high weighing accuracy even in dynamic environments with continuous water flow impact and overflow.

[0044] (3) The present invention cleverly sets up a vibrating multi-stage series screening device at the back end to achieve adaptive separation of particles, and with the real-time conversion formula based on Archimedes' buoyancy law, the loss mass of pure dry soil in each particle size range can be directly output online in milliseconds without interrupting the test, thus realizing "all-time, non-destructive, and digital" monitoring of particle migration. Attached Figure Description

[0045] Figure 1 This is a schematic diagram of the overall structure of the experimental device of the present invention;

[0046] Reference numerals: 1. Air pressure controller; 2. Sealed water tank; 3. Computer control terminal; 4. Sample cylinder; 5. Pore water pressure sensor; 6. Test soil; 7. Gravel layer; 8. Pebble layer; 9. Metal filter screen; 10. Three-layer series sieving device; 11. Upper sieve; 12. Middle sieve; 13. Lower sieve; 14. First collection unit; 15. Second collection unit; 16. Third collection unit; 17. Vibration module; 18. Fixed bracket; 19. Weighing sensor; 20. Silicone pad. Detailed Implementation

[0047] The specific embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the following embodiments are only for illustrating the present invention and are not intended to limit the scope of protection of the present invention; without departing from the concept of the present invention, those skilled in the art can make equivalent substitutions or improvements to the structural form, screen aperture, number of sensors, loading curve, and collection unit form.

[0048] Example 1

[0049] An experimental apparatus for simulating continuous circulating hydraulic gradient erosion of graded soil includes a pneumatically controlled circulating hydraulic gradient loading system, a multi-stage sieving and grading collection system, and a data monitoring system. These three systems work together to achieve circulating hydraulic gradient loading, pore pressure feedback correction, particle sieving and collection, and online quality monitoring.

[0050] like Figure 1 As shown, the pneumatically controlled circulating hydraulic gradient loading system includes a pneumatic controller 1, a sealed water tank 2, and a computer control terminal 3. The sealed water tank 2 is a pressure-bearing container with an air inlet at the top connected to the pneumatic controller 1 via an air pipe. The water outlet at the bottom of the sealed water tank 2 is connected to the water inlet at the bottom of the sample cylinder 4 via a pipeline. Valves, vents, or pressure stabilizing elements can be installed on the pipeline for initial water filling, venting, shutdown maintenance, and adjustment of loading stability.

[0051] The data monitoring system includes a sample cylinder 4, which is a transparent or cross-sectional cylindrical container sealed at both ends by flanges or glands. Inside the sample cylinder 4, from bottom to top, are a pebble layer 8 and a gravel layer 7, with a metal filter 9 laid on top of the gravel layer 7. The pore size of the metal filter can be determined based on the particle size of the soil sample and the particle sizes of the pebble layer 8 and gravel layer 7, which serve as buffer layers; for example, a metal filter with a pore size of approximately 5 mm can be used.

[0052] The test soil 6 was layered and compacted above the metal filter screen 9, according to the target dry density and moisture content. Pore water pressure sensors 5 were installed in the upper, middle and lower parts of the test soil 6, and the sensor leads were connected to the computer control terminal 3.

[0053] The specific implementation principle of continuous circulating hydraulic gradient air pressure loading and pore pressure closed-loop correction is as follows:

[0054] The pressure controller 1 injects pressurized gas into the top of the sealed water tank 2. According to Pascal's law, the gas pressure is transmitted to the surface of the liquid inside the tank. This supplies the total input pressure to the sample cylinder 4. for:

[0055] Total input pressure =Water tank's own gravity head + air pressure applied by the air pressure controller;

[0056] As long as the air pressure output by the air pressure controller 1 increases or decreases continuously and smoothly, the water head pressure at the input end will increase or decrease synchronously and continuously.

[0057] Before the experiment began, the experimenters set the target loading curve in the computer control terminal 3. The target loading curve could be a time-pressure curve or a time-target hydraulic gradient. The curve was then calibrated and converted into a pressure output curve.

[0058] Preferably, the target hydraulic gradient curve can be in the form of a sine-like function: i(t) = i0 + A·sin(2πft), where i(t) is the target hydraulic gradient at time t, i0 is the average hydraulic gradient, A is the amplitude, and f is the loading frequency. Depending on the experimental requirements, the target loading curve can also be a trigonometric function, trapezoidal function, piecewise function, or user-defined function. The computer calculates the initial air pressure value to be applied based on the initial state and sends a command to the air pressure controller 1. The air pressure controller 1 adjusts the high-pressure air source according to the command, changing the air pressure at the top of the sealed water tank 2. The change in pressure inside the water tank instantly pushes the water flow into the sample cylinder.

[0059] The average hydraulic gradient i0 is calculated based on the ratio of the upstream and downstream water level difference to the seepage path length under normal operating conditions in the actual project. The hydraulic gradient fluctuation amplitude A is calculated based on the maximum transient head increment when the actual project encounters extreme dynamic loads (such as tidal fluctuations or sudden drops in reservoir water level), combined with the seepage path length. The cyclic loading frequency f is determined based on the reciprocal of the physical period of the hydraulic fluctuation source in the actual project (such as the tidal period, the daily regulation period of the pumped storage power station, or the frequency of dynamic disturbance). Thus, a theoretical loading parameter set is jointly established, with the average hydraulic gradient i0, amplitude A, and frequency f as its core.

[0060] Multiple pore water pressure gauges 5 are installed at different heights within the sample cylinder. These gauges measure the actual water pressure (P1, P2…) at different locations inside the soil in real time. The sensors transmit this water pressure data back to the computer in real time (e.g., 100 times per second). Upon receiving the real-time water pressure, the computer immediately converts the water pressure into head and calculates the actual real-time hydraulic gradient inside the soil.

[0061]

[0062] in, The total head difference between the two ends of the sample is (i.e., the total head at the input end minus the atmospheric pressure head at the output outlet). L is the height of the sample (a fixed value).

[0063] Computer will Compared with the target value set at this moment Comparison:

[0064] if < This indicates that the soil has become denser or there is a lag in the pipeline, resulting in increased resistance and causing the hydraulic gradient to not reach the expected level. The computer immediately issues an instruction to slightly increase the air pressure in the air pressure controller.

[0065] if > The computer immediately issued an instruction to the air pressure controller to slightly reduce the air pressure.

[0066] This closed-loop design has the following beneficial effects:

[0067] The effect of a drop in water level in the tank is eliminated: as the experiment progresses, the water in the tank decreases, and the tank's own gravitational head decreases. When the closed-loop system detects a downward trend in the hydraulic gradient, it automatically increases the air pressure to compensate for the lost water level, maintaining the gradient accuracy.

[0068] The system hysteresis is overcome: gas is compressible, and there is a slight delay in the transmission of gas pressure to water pressure. Through high-frequency feedback and adjustment, this delay can be corrected to the microsecond level, achieving truly continuous and smooth changes.

[0069] Adapting to dynamic reconstruction within the soil: When fine soil particles are washed away and pores enlarge, water flow accelerates, and the gradient changes. The closed-loop system can sense these internal changes in real time and actively adapt, ensuring that experimental conditions are not distorted.

[0070] During the test, the air pressure controller 1 changes the air pressure inside the sealed water tank 2 according to the instructions of the computer control terminal 3, so that the water pressure inside the tank acts on the bottom of the sample tube 4, and the seepage water flows in from the bottom of the sample tube 4 and passes through the sample from bottom to top.

[0071] The integrated process of sequential particle screening, graded collection, and online continuous quality conversion is specifically implemented as follows:

[0072] The multi-stage sieving and grading collection system includes a three-layer sieving device 10 connected in series. The inlet of the three-layer sieving device is located at the top and is connected to the outlet at the top of the sample cylinder 4. The three-layer sieving device 10 is set behind the outlet of the sample cylinder by a fixed bracket 18. The three-layer sieving device 10 has a cylindrical structure. Inside, from top to bottom, there are three layers of sieves: an upper sieve 11 with a 2 mm aperture, a middle sieve 12 with a 0.25 mm aperture, and a lower sieve 13 with a 0.075 mm aperture. The side walls of the three-layer sieving device are provided with outlets for the sieve residue, which are connected to the corresponding grading collection unit through independent connecting pipes. Specifically, the outlet of the upper screen 11 is connected to a first collection unit 14 with a diameter greater than 2 mm; the outlet of the middle screen 12 with a diameter of 0.25 mm is connected to a second collection unit 15 with a diameter of 0.25 to 2 mm; and the outlet of the lower screen 13 with a diameter of 0.075 mm is connected to a third collection unit 16 with a diameter of 0.075 to 0.25 mm.

[0073] After the water-soil mixture discharged from the outlet of sample cylinder 4 enters the three-layer series screening device 10, the mixture first flows through the upper screen 11. Particles larger than 2 mm are retained by the upper screen and enter the first collection unit 14 through the first connecting pipe. Particles smaller than 2 mm and water continue to fall to the middle screen 12. Subsequently, the middle screen 12 retains particles from 0.25 to 2 mm and introduces them into the second collection unit 15 through the second connecting pipe; particles smaller than 0.25 mm and water continue to fall to the lower screen 13. Finally, the lower screen 13 retains particles from 0.075 to 0.25 mm and introduces them into the third collection unit 16 through the third connecting pipe; fine particles smaller than 0.075 mm and water are discharged through the drain outlet at the bottom of the three-layer series screening device 10 to the filtrate collection, waste liquid treatment, or water recycling unit.

[0074] To facilitate the automatic falling of the trapped material on the screens into each collection unit, each screen is tilted in the same direction. A vibration module 17 is installed on the three-layer series screening device 10, and a silicone pad 20 is installed at the bottom of the fixed support 18. The silicone pad can reduce the vibration transmitted to the bottom test platform, thereby affecting the seepage field of the soil in the weighing sensor and the sample cylinder 6.

[0075] The collection unit includes a support frame, a base, a flow-guiding funnel, an outer cylinder, and an overflow cylinder. The support frame is fixed to the base, the flow-guiding funnel is located at the top of the support frame, and the load cells are fixed to the support frame. The load cells do not contact the flow-guiding funnel, and each load cell has a hook at its lower end. The overflow cylinder is suspended below the load cells by a rod / rope. The thin lower tube of the flow-guiding funnel is inserted into the overflow cylinder, but there are physical gaps between the funnel and the inner and outer walls of the overflow cylinder. The two do not touch each other in either the horizontal or vertical direction. The only vertical force on the overflow cylinder is its own weight (container + water + sediment). The outer cylinder is placed on the base, and the overflow cylinder is located inside the outer cylinder, without contacting the outer cylinder. An overflow port is provided on the wall of the overflow cylinder.

[0076] Figure 1 The arrows pointing from the screen to the collection unit shown are only to indicate the corresponding collection direction of the intercepted material. The intercepted material can enter the corresponding collection unit through convenient and unobstructed channels such as chutes and pipes.

[0077] The trapped material and entrained water falling from the screen first impact the guide funnel. Because the funnel is rigidly fixed to the external support, the impact force is directly absorbed by the support and is not transmitted to the sensor at all. After the water and soil flow through the lower pipe of the funnel, they slide into the overflow cylinder below in a pressureless and extremely gentle free fall, ensuring absolute stability in the weighing process. Because the water is pre-filled to the overflow ports on both sides of the overflow cylinder, when soil particles settle and occupy space, an equal volume of water will automatically overflow the overflow ports and fall into the outermost cylinder. Since the outer cylinder is directly fixed to the base, the gravity of the overflowing wastewater will not interfere with the weighing sensor suspended in mid-air.

[0078] The upflow erosion mechanism uses the seepage force in the opposite direction to gravity to simulate the loose and suspended state of soil. It can accurately simulate the critical point of loosening and failure caused by groundwater backing when a slope becomes unstable in nature, where the effective stress decreases to zero. It also effectively avoids internal self-clogging through overflow.

[0079] The three-layer tandem sieving device 10 of this application is spatially independent of the sample cylinder; it is a large-diameter vertical tower-like structure. The diameter and effective flow area of ​​the screens inside the sieving tower are much larger than the outlet diameter of the sample cylinder, resulting in near-zero resistance to fluid flow. Simultaneously, the unidirectional inclined configuration of each stage of the screens and the excitation effect of the high-frequency vibration module 17 ensure that intercepted solid particles are promptly discharged from the side wall outlet, fundamentally preventing secondary self-clogging and backpressure caused by particle aggregation in the mesh. This design ensures that the sample cylinder outlet maintains a constant atmospheric pressure free physical boundary, making the dynamic evolution of the seepage field inside the sample entirely dependent on the soil's own permeability characteristics and the input control gradient, completely eliminating interference from boundary conditions and guaranteeing the accuracy of the test data.

[0080] Online quality conversion principle:

[0081] Before the experiment began, the overflow cylinder of each collection unit was filled with pure water until it just overflowed from the overflow port. After standing to remove air bubbles, the weighing sensor at the top measured an initial total mass, which is the initial full water mass M0 of each collection unit. M0 = weight of the overflow cylinder + weight of the constant volume of pure water.

[0082] During the experiment, when soil particles of a specific size and entrained water flowed through the funnel into the overflow cylinder, the particles settled at the bottom of the cylinder due to gravity. Since the cylinder was already full of water, the amount of soil particles entering each volume Vs was equal to the volume of water. The entrained water will inevitably result in an equal volume of water being squeezed out from the overflow port and flowing into the outer cylinder for discharge. Once the liquid level fluctuations subside, the weighing sensor reads the total mass at this point, denoted as Mt.

[0083] Assume that within a short period of time, the total amount of matter falling through the funnel into the overflow cylinder includes:

[0084] The mass of the soil particles entering is Its volume is ;

[0085] The mass of water entrained is Its volume is ;

[0086] The total increase in mass entering the overflow cylinder is:

[0087]

[0088] Because the overflow pipe is initially full of water, any volume of material that falls in must be replaced by an equal volume of water.

[0089] The total volume of the material that fell in is:

[0090] Therefore, the volume of water squeezed out of the overflow outlet and discharged into the outer cylinder must be absolutely equal to this total inflow volume:

[0091] Therefore, the total mass of the discharged water is:

[0092]

[0093] because equal to the mass of the water entering ,so:

[0094] The change measured by the sensor is the total mass entering minus the total mass exiting:

[0095]

[0096] Substitute the preceding terms into:

[0097]

[0098]

[0099] Through derivation, it can be found that the real cause of the change in the weighing sensor readings is purely due to soil particles.

[0100] Bundle Substitute into the above formula:

[0101]

[0102] Final mass conversion formula:

[0103]

[0104] in: : Density of soil particles (a known constant, measured and entered into the computer before the test).

[0105] : The density of water (a known constant, 1.0).

[0106] The difference between the real-time measurement value of the weighing sensor and the initial total mass of the water-filled area.

[0107] The computer reads the data from the weighing sensor 19 every second (or more). Using the formula above, a continuous curve showing the change in the mass of dry soil loss of that particle size over time can be drawn instantly on the screen.

[0108] This design cleverly utilizes the density difference between soil and water. As long as the soil particles settle down and the water overflows, the weight difference can be amplified by a fixed coefficient to calculate the dry weight. There is no need to stop the machine to dry and weigh the product, turning "offline drying" into "online real-time drying".

[0109] The upper guide funnel and the suspended overflow cylinder are mechanically decoupled (do not contact each other), and the particles settle naturally, which eliminates the dynamic impact interference of high-pressure water flow in the pipeline on the sensor to the greatest extent and ensures weighing accuracy.

[0110] Example 2:

[0111] The test method of the present invention includes the following steps:

[0112] The first step is to determine the soil sample gradation, dry density, moisture content, target hydraulic gradient curve and cyclic loading parameters according to the experimental design, prepare the test soil 6, and bury the pore water pressure sensor 5 in the upper, middle and lower parts of the test soil respectively.

[0113] The second step is to complete the connection of the sealed water tank 2, sample cylinder 4, three-layer series sieve device 10, grading collection unit, weighing sensor 19, vibration module 17 and computer control terminal; fill the sealed water tank 2 with water and remove air from the pipeline; add water to each grading collection unit to a fixed volume and record the initial full water mass M0.

[0114] The third step is to set the target loading curve on the computer control terminal 3, start the air pressure controller 1, so that the air pressure in the sealed water tank 2 changes according to the target curve, and a continuous circulating hydraulic gradient is formed at the bottom of the sample tube 4.

[0115] The fourth step is to start the computer control terminal and vibration module 17 to synchronously collect pore pressure data and mass data of each collection unit; the water-soil mixture at the outlet of the sample cylinder 4 enters the three-layer series screening device 10 for step-by-step screening.

[0116] The fifth step is to introduce the material trapped in the upper screen 11, the middle screen 12 and the lower screen 13 into their respective collection units.

[0117] The sixth step involves the computer control terminal calculating the mass change curve of particles in each size range over time based on the mass sequence and density difference formula of the weighing sensor 19; judging the loading error based on the pore pressure monitoring results; and adjusting the air pressure output in real time through the computer control terminal 3.

[0118] Step 7: After the test, depressurize, disconnect power and clean the device according to the safety procedure. Export the target loading curve, pore pressure response curve, particle mass-time curve and particle migration rate curve for each particle size range. These are used to analyze the particle initiation threshold, migration hysteresis effect and cumulative erosion law under the circulating hydraulic gradient.

[0119] Optionally, the number of pore water pressure sensors 5 can be set to three or more according to the sample height and test accuracy requirements; the mesh size of the three-layer series sieving device 10 can be adjusted according to the research object, and the number of mesh layers can be increased or decreased, as long as each mesh is set in series in descending order of mesh size along the flow direction of the water-soil mixture, and the material intercepted by each mesh enters the corresponding collection unit respectively; the vibration module 17 can be an eccentric motor vibrator, an electromagnetic vibrator or a pneumatic vibrator, and the vibration frequency and amplitude are adjustable; a pressure stabilizing chamber, a throttling element or a flow monitoring element can be set between the sealed water tank 2 and the sample cylinder 4 to improve the stability of water pressure transmission under a specific loading frequency; the structure of the graded collection unit can be a built-in collection cylinder, an external overflow cylinder, a detachable collection cup or other equivalent full-water constant-volume structure.

[0120] After connecting all the experimental instruments, the only human operation required for the entire experiment is to preset and change the hydraulic gradient through the computer control terminal, eliminating human intervention, significantly improving the accuracy of the experiment, and the experimental data can be collected in real time and saved in folders in the form of Excel, with a high degree of automation, greatly reducing manual time and the time required for the experiment.

[0121] After the experiment, the target loading curve, pore pressure response curve, particle mass-time curve and particle migration rate curve for each particle size range were exported from the computer control terminal 3. These were used to analyze the particle initiation threshold, migration hysteresis effect and cumulative erosion law under the circulating hydraulic gradient.

[0122] By comparing the target loading curve and the pore pressure response curve over time, the dynamic impedance effect within the soil can be analyzed. The target loading curve (time-varying hydraulic gradient curve) represents the active excitation source of the system input, equivalent to the applied external force, while the pore pressure response curve reflects the real-time resistance state of the soil's internal framework and pores. Comparing these two curves reveals whether, in the cyclical flow of water, the soil's internal structure becomes looser (decreasing pore pressure difference, increasing flow velocity) or becomes locally blocked (sudden increase in pore pressure).

[0123] Particle mass-time curves for each particle size range can be used for:

[0124] Quantitative assessment of synergistic erosion effects: Erosion of graded soil is often not a single-size loss, but rather a process of "large particles leading small particles" or "small particles leading large particles." By comparing the rise amplitude and speed of the three curves, it is possible to quantitatively analyze which particle size range of soil particles is lost first and which particle size range suffers the most severe loss under the same hydraulic gradient.

[0125] Determining the state of soil skeleton failure: If the mass curve of coarse particles begins to rise sharply, it indicates that the coarse particle skeleton of the soil has disintegrated, and the soil has experienced destructive piping or collapse; if only the curve of fine particles fluctuates, it indicates that only free fine particles in the pores are migrating (which is a temporary erosion).

[0126] By integrating and accumulating the particle mass-time curves for each particle size range over the entire time period, and using the total test run time or number of cycles as the abscissa, the dry soil mass calculated from the first, second, and third collection units is accumulated over the entire time period. By fitting the growth function relationship with the cycle period, the cumulative erosion evolution model of the graded soil under continuous cyclic seepage can be established, determining whether it belongs to stable erosion or sudden destructive erosion.

[0127] Computer reads two consecutive acquisition times and dry soil quality and The particle migration rate within a given time period is calculated using the difference formula, yielding a particle migration rate curve. This curve serves as a direct tool for identifying the particle initiation threshold and migration lag effect. Since the mass curve is an additive, gently rising curve, it's difficult to visually pinpoint when it begins to rise. In contrast, the migration rate curve is zero when particles are not initiating movement; once initiation occurs, the curve jumps abruptly, making the inflection point easily detectable. Superimposing the sine curve of the hydraulic gradient fluctuation with the particle migration rate curve allows for clear observation of whether the particle migration rate reaches its peak simultaneously with the hydraulic gradient peak. If the rate peak lags behind the hydraulic gradient peak, the lag time of particle movement within the soil pores can be quantitatively calculated.

[0128] The embodiments described above are merely illustrative of implementation methods of the present invention and should not be construed as limiting the scope of protection of the present invention. Those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and all such modifications and improvements should fall within the scope of protection of the present invention.

Claims

1. A test apparatus for simulating continuous circulating hydraulic gradient erosion of graded soil, characterized in that, include: The pneumatic pressure controlled circulating hydraulic gradient loading system includes a computer control terminal (3), a pneumatic pressure controller (1) electrically connected to the computer control terminal (3), and a sealed water tank (2) with an air inlet at the top and a water outlet at the bottom; the air inlet is connected to the pneumatic pressure controller (1) through an air pipe; The data monitoring system includes a sample tube (4) with a pebble layer (8), a gravel layer (7), a metal filter (9), and a test soil body (6) laid out from bottom to top inside; the outlet of the sealed water tank (2) is connected to the inlet at the bottom of the sample tube (4) through a pipeline to form an upward seepage flow from bottom to top; several pore water pressure sensors (5) are buried at different heights of the test soil body (6), and each of the pore water pressure sensors (5) is electrically connected to the computer control terminal (3); The multi-stage sieving and grading collection system includes a series sieving device (10) independently set behind the water outlet of the sample tube (4) and several grading collection units; the top water inlet of the series sieving device (10) is connected to the top water outlet of the sample tube (4); the inside of the series sieving device (10) is arranged vertically from top to bottom with multiple layers of screens with decreasing apertures; each layer of screen has a corresponding screen retention outlet on the side wall of the series sieving device (10); each screen retention outlet is connected to the corresponding grading collection unit through an independent connecting channel.

2. The test apparatus for simulating continuous circulating hydraulic gradient erosion graded soil according to claim 1, characterized in that: Each of the aforementioned staged collection units adopts a mechanically decoupled, full-water, constant-volume collection structure, and each staged collection unit specifically includes: External mounting bracket and base; the external mounting bracket is fixed to the base; A flow guide funnel is rigidly fixed to the top of an external fixing frame, and the top of the flow guide funnel is connected to the corresponding screen. The overflow cylinder is suspended below the load cell (19). The load cell (19) is fixed on an external mounting bracket and electrically connected to the computer control terminal (3). The load cell does not contact the guide funnel. The lower outlet of the guide funnel is inserted into the overflow cylinder and maintains a physical gap between it and the inner and outer walls of the overflow cylinder to achieve mechanical decoupling. An outer cylinder is placed on a base and located around the overflow cylinder, without contacting the overflow cylinder; an overflow port is provided on the side wall of the overflow cylinder, and the overflow port is connected to the inside of the outer cylinder.

3. The test apparatus for simulating continuous circulating hydraulic gradient erosion of graded soil according to claim 1, characterized in that: Each layer of screen in the series screening device (10) is inclined in the same direction. The series screening device (10) is also equipped with a vibration module (17). The series screening device (10) is supported by a fixed bracket (18), and a silicone pad (20) is provided at the bottom of the fixed bracket (18).

4. The test apparatus for simulating continuous circulating hydraulic gradient erosion of graded soil according to claim 1, characterized in that: The series screening device (10) is vertically arranged from top to bottom with an upper screen (11) with a 2 mm aperture, a middle screen (12) with a 0.25 mm aperture, and a lower screen (13) with a 0.075 mm aperture. The graded collection unit includes a first collection unit (14) corresponding to the upper screen (11), a second collection unit (15) corresponding to the middle screen (12), and a third collection unit (16) corresponding to the lower screen (13).

5. The test apparatus for simulating continuous circulating hydraulic gradient erosion of graded soil according to claim 1, characterized in that: The sample tube (4) is a transparent or cross-sectional cylindrical container, and its upper and lower ends are sealed by flanges or caps.

6. A test method for simulating continuously circulating hydraulic gradient erosion graded soil, characterized in that, Includes the following steps: Step 1: Experiment Preparation; Based on the purpose of the experiment, the soil sample gradation, dry density and moisture content were designed, and the theoretical loading parameters of the target hydraulic gradient were calculated. The test soil (6) was filled in the sample tube (4), and several pore water pressure sensors (5) were buried at different heights in the test soil (6). Step 2: Device connection and initialization; complete the physical connection and signal connection of each device component in the experiment; fill the sealed water tank (2) with water and remove the air from the pipeline; add pure water to the overflow tube of the graded collection unit used to collect the seepage particles until it just overflows from the overflow port, and after the liquid level has settled and stabilized, record the initial total mass M0 of each graded collection unit filled with water through the weighing sensor (19). Step 3: Start circulating hydraulic gradient loading; input and configure the theoretical loading parameters obtained in Step 1 into the computer control terminal (3) to generate a digital target loading curve; Start the air pressure controller (1) to make the air pressure in the sealed water tank (2) change continuously and smoothly according to the target loading curve, thereby forming a continuous circulating hydraulic gradient at the bottom of the sample tube (4) and driving the seepage to pass through the test soil (6) from bottom to top. Step 4: Gradually screen the leached particles; start the vibration module (17) on the computer control terminal (3) and the series screening device (10). The water-soil mixture discharged from the top outlet of the sample tube (4) enters the series screening device (10). Under the action of vibration, it is screened step by step through the internal screens with decreasing aperture from top to bottom. The intercepted material and entrained water of each screen are introduced into the corresponding overflow tube through the corresponding independent connection channel and guide funnel. Fine particles smaller than the aperture of the bottom screen and wastewater are discharged through the drain outlet at the bottom of the series screening device (10). Step 5: Online continuous mass conversion and dynamic closed-loop correction; During the experiment, the computer control terminal (3) synchronously and frequently collects data from each pore water pressure sensor (5) and real-time mass data from each weighing sensor (19). ; The computer control terminal (3) automatically calculates the real-time hydraulic gradient inside the soil based on the real-time water pressure data fed back by each pore water pressure sensor (5). and will The target hydraulic gradient at this moment If a deviation is found, a control command is issued in real time to adjust the output air pressure of the air pressure controller (1) to achieve dynamic closed-loop correction. At the same time, the computer control terminal (3) calculates the mass of dry soil particles lost in each particle size range in real time online according to the density difference formula. And a continuous curve of the change in dry soil loss mass over time was plotted.

7. The test method for simulating continuous circulating hydraulic gradient erosion graded soil according to claim 6, characterized in that: The target hydraulic gradient curve mentioned in step one is: i(t) = i0 + A·sin(2πft); where i(t) is the target hydraulic gradient at time t, i0 is the average hydraulic gradient, A is the amplitude, and f is the loading frequency.

8. The test method for simulating continuous circulating hydraulic gradient erosion graded soil according to claim 6, characterized in that: In step five, the density difference formula is: in: Density of soil particles; The density of water; The difference between the real-time measurement value of the weighing sensor and the initial total mass of the water-filled area.

Images