Method for solving parameters of rock constitutive characteristic VG model

Abstract

The invention discloses a rock constitutive characteristic VG model parameter solving method, which comprises the following steps: S1, placing a rock sample soaked in water on a flat test bed, keeping the surface of the rock in a saturated state all the time by supplementing water, wiping the surface at set intervals, weighing, and calculating the mass increment of the rock sample; s2, converting the mass increment of the rock sample into the volume delta V of water, solving to obtain the depth change delta zf of the wetting peak according to the sectional area S of the rock and the measured porosity epsilon, measuring a plurality of groups of data, fitting a Greep-Ampt model according to the data, and solving to obtain the matrix suction Sfi of the wetting peak of the rock under the water content; s3, i equations and four unknown quantities m, n, l and alpha are obtained; solving the equation set to obtain VG model parameters; according to the method, the parameters of the Green-Ampt infiltration model are measured, and the parameters of the Van-Genuchten model are indirectly solved by utilizing the conversion relation between the parameters and the parameters of the VG model, so that the soil-water characteristics of the rock are obtained.

Description

Technical Field

[0001] The invention relates to the technical field of geotechnical engineering, and in particular to a method for solving parameters of a VG model of rock constitutive characteristics. Background Art

[0002] In geotechnical engineering, if only the saturated area is considered, it will not only increase the difficulty of analysis, but also make it impossible to solve complex problems. Some scholars at home and abroad are committed to the numerical calculation method of comprehensive consideration of saturated-unsaturated seepage area. In the numerical calculation of unsaturated seepage, the determination of unsaturated seepage parameters is a very important part of the numerical calculation.

[0003] For porous media such as rocks, the important difference between the movement of water in saturated media and unsaturated media lies in the difference in permeability coefficients. When the medium reaches saturation, all pores are filled with water, and the permeability coefficient is relatively large and generally constant. The pores in unsaturated media are partially occupied by gas, so its permeability coefficient is lower than that when saturated. The permeability coefficient of unsaturated media changes with water content. If you want to obtain the seepage characteristics of unsaturated porous media, you must understand the relationship between permeability coefficient and matrix potential or water content. This relationship is generally described by the characteristic curve of hydraulic parameters of unsaturated porous media, that is, the soil-water characteristic curve (SWCC).

[0004] The soil-water characteristic curve characterizes the relationship between the matrix suction and moisture content (such as gravity moisture content, volumetric moisture content) or saturation of unsaturated soil. It contains the amount of water contained in the pores and pore size distribution information of the soil at any suction. The SWCC can not only determine the water holding capacity of the soil, but also be used to calculate the unsaturated permeability coefficient, which is of great significance for studying the migration and retention of soil moisture and the water absorption performance of plants. The relationship between the moisture content and water absorption capacity of rock and soil media can also be characterized by this curve. The measurement methods can be divided into direct and indirect methods. Most of the measurement methods are for soil and are widely used in agriculture.

[0005] The methods for directly measuring SWCC (laboratory environment) mainly include the hanging water column method (or sand core funnel method), tensiometer method, pressure membrane method, eccentricity method, electric heat conduction technology method, equilibrium water vapor pressure method, etc. The soil samples used in these tests can be both disturbed soil and original soil. The sand core funnel method establishes a hydraulic connection with the soil sample through the sand core funnel and the clay plate, and can measure the moisture absorption curve and drainage curve of the soil sample. The maximum measured suction is 0.1MPa. The tensiometer method can only measure the suction range of 0-0.08MPa. It is widely used in the dynamic monitoring of field soil moisture, but due to the low accuracy of the tensiometer head, its measurement range and accuracy are limited. The pressure membrane method is to apply pressure to make the water in the soil flow out, so that the soil matrix potential is reduced until the pressure balance is reached. This method is suitable for the dynamic simulation of soil moisture and has high accuracy, but its measurement cycle is long and there will be problems such as changes in bulk density. The eccentricity method is essentially to conduct a soil column experiment in a centrifuge. The water content corresponding to the suction can be obtained by weighing on a balance. The equilibrium time required for measuring the matrix is ​​determined by experiment. This method has high measurement accuracy, but the bulk density of the soil will change greatly, especially for clay particles. In general, the direct measurement method for obtaining soil SWCC has the disadvantages of long cycle, high cost and discrete data.

[0006] The direct method of determining SWCC is time-consuming and expensive, and it is difficult to implement in the field, which limits its application in practice. Therefore, many scholars at home and abroad have tried to deduce SWCC through indirect methods. The indirect method here refers to the method of using basic soil physical properties data, such as soil particle size distribution, organic matter content, bulk density, etc., and then predicting SWCC through empirical or semi-empirical relationships. In terms of predicting SWCC, soil transfer functions (PTFs) use known basic soil physical properties and then use a certain algorithm to construct a relationship between predicted suction and water content. From this perspective, almost all indirect methods for predicting SWCC can be summarized as soil transfer function methods.

[0007] For dense porous media such as rock, it is impossible to obtain parameters such as particle size distribution like soil, and naturally it is impossible to use the traditional method of predicting the soil-water characteristic curve based on soil transfer functions (PTs); the porosity and permeability of rock are relatively low compared to soil, and the measurement error using instruments is large. In addition, the processing is difficult, and the dense particles are difficult to insert into the instrument, all of which limit the measurement of the rock soil-water characteristic curve.

[0008] In order to characterize the soil-water characteristic curve and facilitate engineering calculations, predecessors have derived some theoretical models based on test results to deduce the relationship between capillary pressure (matrix suction) and water content. The VG model is one of the most widely applicable models. The expression is as follows: ; In the formula,S e is the effective saturation, S w is the volume fraction of saturated water phase in porous media, S wr is the volume fraction of residual water phase in porous media, p c is the capillary pressure, p ec is the inlet capillary pressure, l is the pore distribution index.

[0009] Relevant scholars have conducted unsaturated seepage tests, and the results show that the VG model can simulate the measured data well. However, compared with soil, which can be tested to determine the SWCC curve and then fit to obtain the VG model parameters, it is difficult to directly obtain the SWCC of rock media through experimental determination due to its compact particles and poor permeability; and since it is impossible to split the rock and study its particle characteristics, it is impossible to indirectly obtain SWCC through its particle distribution characteristics.

[0010] The existing VG model is usually obtained by fitting the soil-water characteristic curve: (1) Determination of soil-water characteristic curve: The existing VG model parameters are often determined by measuring the soil-water characteristic curve fitting. The soil-water characteristic curve test is usually performed on soil. The existing equipment is all for soil and the operation is relatively complicated. Due to the poor permeability of rock, the equipment accuracy requirements are high. At present, there is no soil-water characteristic curve measurement device specifically for rock.

[0011] (2) VG model parameter fitting: After the soil-water characteristic curve is obtained by measurement, since there are two unknown quantities in the model and the function is in the form of a non-integer power series, it needs to be performed through data processing software such as Matlab and Orgin, which is difficult to operate. Summary of the invention

[0012] The purpose of the present invention is to overcome the above-mentioned shortcomings and provide a method for solving the VG model parameters of rock constitutive characteristics. By measuring the Green-Ampt infiltration model parameters and utilizing the conversion relationship between the parameters and the VG model parameters, the Van-Genuchten model parameters are indirectly solved to obtain the soil-water characteristics of the rock.

[0013] In order to solve the above technical problems, the technical solution adopted by the present invention is: a method for solving the parameters of the VG model of rock constitutive characteristics, comprising the following steps: S1. Place the soaked rock sample on a flat test bench, keep the rock surface saturated by adding water, wipe the surface dry and weigh it at regular intervals, and calculate the mass increase of the rock sample; S2. Convert the mass increase of the rock sample into the volume of water ΔV, according to the rock cross-sectional area S and the measured porosity e , and solve for the change in wetting peak depth Δ z f , measure multiple sets of data, fit the Greep-Ampt model according to the data, and solve the wetting peak matrix suction of rock under different water contents S fi ; S3, prepare samples with different moisture contents, the corresponding moisture contents are i 1, i 2, i 3. i 4…… i i , to obtain S f1 , S f2 , S f3 , S f4 … S fi , and the saturated moisture content i s , residual moisture content i r And the initial moisture content of different samples i i Substitute the average suction at the wetting peak into S f In the relationship between the parameters of the VG model, we get i equations and 4 unknown quantities m , n , l , α ; Solve the system of equations to obtain the VG model parameters.

[0014] Preferably, S1 is as follows: the rock sample is processed into a cylindrical shape with both ends flat, the rock dry weight is M, a saturated sample is prepared by vacuuming or boiling, and the saturated mass is recorded. m s , dry the sample to dry state, and record the dry mass m 0; Place the soaked rock sample on a flat test bench, keep the rock surface saturated by adding water, wipe it dry and weigh it regularly, and record the mass at different time points m i , calculate the mass increase of the rock sample.

[0015] Preferably, the rock surface is kept saturated by adding water, and the rock surface is wiped dry and weighed regularly, and the mass at different time points is recorded. m i , at this time: ; ; .

[0016] Preferably, in S2, the Greep-Ampt model is: ; In the formula, f — infiltration rate; K s —permeability coefficient; S f —Average suction at the wetting front; z f —generalized wetting front depth; h —Total water potential at the surface.

[0017] More preferably, the average suction at the wetting peak is S f The relationship between the VG model parameters is as follows: ; In the formula, m =1-1 / n , m is the constitutive property fitting parameter, n is the pore size distribution parameter, l is the matrix suction fitting parameter, α is an empirical parameter, i s is the saturated moisture content, i r is the residual moisture content, that is, the percentage of residual moisture after the sample is dried. i i is the initial moisture content of different samples.

[0018] More preferably, the infiltration rate f About the generalized moist front depth z f The function is: ; Determination of permeability coefficient through rock permeability and porosity tests K s And porosity e , the initial moisture content of different samples was obtained by experimental fitting i i Lower wetting peak matrix suction S fi .

[0019] More preferably, the specific process of S3 is as follows: Prepare a set of samples with different moisture contents and calculate their saturated moisture content i s , residual moisture content i r , initial moisture content of different samples i i , conduct natural seepage experiments, assuming that the infiltration rate of the rock sample at each time interval during the seepage process is f Keep constant, record the mass growth of the rock sample, and the seepage rate per unit time is the infiltration rate f , by converting the weight gain into volume and dividing by the porosity e , and obtain the generalized depth of the wetting peak z f ; when Z fi = S fi At this time f =2 K s , then we can obtain the wetting peak matrix suction S f1 ; Prepare another moisture content sample and repeat the above steps to obtain S f2 , S f3 , S f4 , according to the average suction at the wetting peak S f Relationship with VG model parameters: ; Four sets of expressions are obtained: ; In the formula m are constitutive model parameters, n =1-1 / m , l is the matrix suction parameter, and the other three are unknown quantities. The number of equations is greater than the number of unknown quantities, and the unknown quantities, i.e. the required model parameters, are obtained by solving them by computer.

[0020] Beneficial effects of the present invention: the present invention obtains data through simple water seepage experiments and basic physical measurements (such as mass, volume, time, etc.). Compared with the complex experimental equipment and high cost required for directly measuring the VG model parameters, this method significantly reduces the experimental difficulty and cost; the experimental steps of the present invention are clear and easy to operate, which reduces the errors and uncertainties that may occur during the experiment; the present invention can accurately calculate the water seepage rate and the wetting peak depth by directly measuring the mass changes of rock samples at different time points, and these data are the basis for subsequent parameter fitting and solution; the direct acquisition of data of the present invention avoids the errors that may be caused by indirect calculation, and improves the accuracy and reliability of the results; the present invention uses the mathematical relationship between the Green-Ampt model parameters and the VG model parameters, and indirectly solves the VG model parameters that are difficult to directly measure by measuring the relatively easy-to-obtain Green-Ampt model parameters; the method of the present invention not only solves the problem of difficulty in obtaining VG model parameters, but also expands the application range of model parameters; the indirect solution method of the present invention reduces the requirements for experimental conditions and equipment, so that more researchers can use this method to study the hydrological characteristics of porous media. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 This is a schematic diagram of the natural seepage test; Figure 2 is the infiltration rate f About the generalized moist front depth z f The function curve of . DETAILED DESCRIPTION

[0022] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0023] Example 1: In order to study the unsaturated seepage law of rock, it is necessary to obtain its soil-water characteristic curve, that is, the relationship between matrix suction and saturation. Usually, the practical model Van-Genuchten (hereinafter referred to as VG model) curve is used for fitting. However, due to the particularity of rock media, it is difficult to prepare it into any shape like soil, and the equipment cannot be directly inserted. It is difficult to directly measure it with pressure membrane, filter paper, tensiometer and other equipment. This method measures the Green-Ampt infiltration model parameters, and solves the Van-Genuchten model parameters through the conversion relationship between the parameters and the VG model parameters, thereby obtaining the soil-water characteristics of the rock.

[0024] The present invention provides a method for solving the parameters of the VG model of rock constitutive characteristics, comprising the following steps: S1. Place the soaked rock sample on a flat test bench, keep the rock surface saturated by adding water, wipe the surface dry and weigh it at regular intervals, and calculate the mass increase of the rock sample; S2. Convert the mass increase of the rock sample into the volume of water ΔV, according to the rock cross-sectional area S and the measured porosity e , and solve for the change in wetting peak depth Δ z f , measure multiple sets of data, fit the Greep-Ampt model according to the data, and solve the wetting peak matrix suction of rock under different water contents S fi ; S3, prepare samples with different moisture contents, the corresponding moisture contents are i 1, i 2, i 3. i 4…… i i , to obtain S f1 , S f2 , S f3 , S f4 … S fi , and the saturated moisture content i s , residual moisture content i r And the initial moisture content of different samples i i Substitute the average suction at the wetting peak into S f In the relationship between the parameters of the VG model, we get i equations and 4 unknown quantities m , n , l , α ; Solve the system of equations to obtain the VG model parameters.

[0025] The present invention adopts the Greep-Ampt model to describe the natural infiltration law of porous media. The model assumes that the soil above the wetting front is saturated and the initial moisture content is maintained below it. At the same time, combined with the relationship between the permeability coefficient and the saturation, the relationship between the average suction Sf at the wetting front and the VG model parameters is derived.

[0026] During the experimental preparation and measurement process: Process the rock sample into a cylindrical shape with both ends flat, and record the rock dry weight M. Use the vacuum method or boiling method to prepare a saturated sample and record the saturated mass ms . Dry the sample to dry state and record the dry mass m0. Soak the samples in water and record the quality at different time points my , calculate the mass increase of the rock sample.

[0027] Calculation of wetting peak depth and infiltration rate: Through water replenishment experiments, keep the rock surface saturated, wipe it dry and weigh it regularly, and record the mass increase of the rock sample. Convert the mass increase of the rock sample into the volume of water ΔV, and calculate the change of wetting peak depth Δz by combining the rock cross-sectional area S and porosity ε. f Calculate the change rate of the wetting peak depth Δz f / t, i.e. infiltration rate f .

[0028] Parameter fitting and solution: Repeat the above experiment for samples with different initial moisture contents θi to obtain the corresponding Sfi . Substitute multiple sets of data into Sf The relationship between the VG model parameters and the saturated water content θs , residual moisture content θr and initial moisture content I will When the data volume i is greater than or equal to 4, solve the equation system to solve the VG model parameters m、n、l、a .

[0029] Therefore, the present invention indirectly solves the VG model parameters: when it is difficult to directly obtain the VG model parameters, the Green-Ampt model parameters (such as the infiltration rate f , wetting front depth z f The VG model parameters are indirectly solved by using the mathematical relationship between these parameters and the VG model parameters. Compared with the complex experimental conditions required for directly measuring the VG model parameters, this method can obtain the required parameters through simple water seepage experiments and data processing, reducing the experimental difficulty and cost.

[0030] The specific process is as follows: The infiltration of porous media has special rules. Generally speaking, the seepage of unsaturated porous media is affected by gravity and matrix effects, which are usually characterized by infiltration rate. The matrix effect is different under different water contents.

[0031] The Greep-Ampt model expresses the natural infiltration law of porous media with equation (1). This equation assumes that the soil area above the wetting front (wetting peak) reaches the saturated water content, while below the wetting front, the soil moisture content remains at the initial moisture content. i i .

[0032] (1) In the formula, f — infiltration rate; Ks —permeability coefficient; S f —Average suction at the wetting front; z f —generalized wetting front depth; h —Total water potential at the surface.

[0033] The average suction force at the wetting peak S f Using the integral form of the permeability coefficient and combining the relationship between the permeability coefficient and saturation, we can finally get S f Relationship with VG model parameters: (2) In the formula, m =1-1 / n , m is the constitutive property fitting parameter, n is the pore size distribution parameter, l is the matrix suction fitting parameter, α is an empirical parameter, i s is the saturated moisture content, i r is the residual moisture content (the percentage of residual moisture after the sample is dried), i i is the initial moisture content of different samples.

[0034] It is worth noting that the above three moisture contents cannot be directly measured because the residual moisture content corresponding to the residual moisture content cannot be directly measured.

[0035] Simplifying equation (1) and considering that the surface water potential is very small and can be ignored, we have: (3) The permeability coefficient can be determined by rock permeability and porosity tests K s And porosity e , at this time, equation (3) becomes the infiltration rate f About Wetting Peak Depth z f The function of the initial moisture content can be fitted through experiments. i i Lower wetting peak matrix suction S fi .

[0036] In addition, before the test, the rock sample was processed into a cylindrical shape with both ends flat. Assuming the dry weight of the rock is M, a saturated sample was prepared by vacuuming or boiling. At this time, the saturated mass of the rock is ms , natural water seepage treatment, placed in a dryer at 105° for 24 hours, then a dry sample is obtained with a mass of m 0. Soak the rocks in water and record their quality at different time points m i , at this time: (4) like Figure 1 As shown in the figure, the rock sample after soaking in water is placed on a flat test bench, and water is added through a nozzle to keep the upper surface of the rock in a saturated state. There is a thin layer of water. The surface is wiped dry and weighed at regular intervals to calculate the mass increase of the rock sample. It is considered that the water seepage rate is constant during this period of time, that is, f Constant, at this time, the mass increase of the rock sample is converted into the volume of water ΔV, according to the rock cross-sectional area S and the measured porosity e , the change in wetting peak depth Δ can be obtained z f , Δ z f / t is f ,Σ(Δ z f ) is the wetting peak depth, K s Through permeability test, multiple sets of data are measured, and the curve of formula (1) is fitted according to the data to obtain the rock water content under this water content. S fi .

[0037] Furthermore, the present invention prepares samples with different moisture contents, and the corresponding moisture contents are i 1, i 2, i 3. i 4…… i i , correspondingly obtain S f1 , S f2 , S f3 , S f4 … S fi , put each group of data into formula (2), and at the same time, the saturated moisture content i s , residual moisture content i r And the initial moisture content of different samples i i Substitute the average suction at the wetting peak into S fThe relationship between the VG model parameters is: S f The relationship between the VG model parameters is equation (2), and we get i equations and 4 unknown quantities. m , n , l , α When i is greater than or equal to 4, solve the equations to obtain the VG model parameters. The specific process is described as follows: Prepare a set of samples with different moisture contents and calculate their saturated moisture content i s , residual moisture content i r , initial moisture content of different samples i i , conduct natural seepage experiments, assuming that the infiltration rate of the rock sample at each time interval during the seepage process is f Keep constant, record the mass growth of the rock sample, and the seepage rate per unit time is the infiltration rate f , by converting the weight gain into volume and dividing by the porosity e , and the wetting front depth is z f , infiltration rate f About the generalized moist front depth z f The function curve is as follows Figure 2 As shown; when Z fi = S fi At this time f =2 K s , then we can obtain the wetting peak matrix suction S f1 ; Prepare another moisture content sample and repeat the above steps to obtain S f2 , S f3 , S f4 , according to the average suction at the wetting peak S f Relationship with VG model parameters: ; Four sets of expressions are obtained: ; In the formula m are constitutive model parameters, n =1-1 / m , lis the matrix suction parameter, and the other three are unknown quantities. The number of equations is greater than the number of unknown quantities, and the unknown quantities, i.e. the required model parameters, are obtained by solving them by computer.

[0038] Under the background that it is difficult to obtain the rock VG model, the present invention utilizes the Green-Ampt infiltration model parameters with fewer parameters determined by experiments, and utilizes the relationship between the parameters and the VG model parameters of the soil-water characteristic empirical model to indirectly solve and obtain the VG model parameters.

[0039] This method uses the natural seepage test of rocks and the water infiltration model Green-Ampt. It only needs to record the water absorption in a certain period of time, convert it through known parameters, and fit the Green-Ampt model curve with a simpler form of -1 power to obtain the required parameters. It is highly operational and has low requirements for equipment. During the test, only mass and time are measured, there are few interference factors, and the measurement accuracy is relatively high.

[0040] Since the unknown parameters of the VG model are constants, multiple parameters can be obtained by conducting natural seepage tests on rocks under different water contents, and a set of equations can be constructed to solve each parameter immediately.

[0041] The above embodiments are only preferred technical solutions of the present invention and should not be regarded as limiting the present invention. The protection scope of the present invention shall be the technical solutions recorded in the claims, including equivalent replacement solutions of the technical features in the technical solutions recorded in the claims. That is, equivalent replacement improvements within this scope are also within the protection scope of the present invention.

Claims

1. A method for solving the parameters of the VG model of rock constitutive properties, characterized by: The following steps are involved: S1. Place the soaked rock sample on a flat test bench, keep the rock surface saturated by adding water, wipe the surface dry and weigh it at regular intervals, and calculate the mass increase of the rock sample; S2. Convert the mass increase of the rock sample into the volume of water ΔV, according to the rock cross-sectional area S and the measured porosity ε , and solve for the change in wetting peak depth Δ z f , measure multiple sets of data, fit the Greep-Ampt model according to the data, and solve the wetting peak matrix suction of rock under different water contents S fi ; S3, prepare samples with different moisture contents, the corresponding moisture contents are θ 1, θ 2, θ 3. θ 4…… θ i , to obtain S f1 , S f2 , S f3 , S f4 … S fi , and the saturated moisture content θ s , residual moisture content θ r And the initial moisture content of different samples θ i Substitute the average suction at the wetting peak into S f In the relationship between the parameters of the VG model, we get i equations and 4 unknown quantities m , n , l , α ; Solve the system of equations to obtain the VG model parameters.

2. The method for solving the rock constitutive characteristic VG model parameters according to claim 1, characterized in that: Specifically, S1 includes: processing the rock sample into a cylindrical shape with flat ends, the rock dry weight M, preparing a saturated sample by vacuuming or boiling, and recording the saturated mass m s , dry the sample to dry state, and record the dry mass m 0; Place the soaked rock sample on a flat test bench, keep the rock surface saturated by adding water, wipe it dry and weigh it regularly, and record the mass at different time points m i , calculate the mass increase of the rock sample.

3. The method for solving the rock constitutive characteristic VG model parameters according to claim 2, characterized in that: Keep the rock surface saturated by adding water, wipe it dry and weigh it regularly, and record the mass at different time points m i , at this time: ； ； 。 4. The method for solving the rock constitutive characteristic VG model parameters according to claim 1, characterized in that: In S2, the Greep-Ampt model is: ； In the formula, f — infiltration rate; K s —permeability coefficient; S f —Average suction at the wetting front; z f —generalized wetting front depth; h —Total water potential at the surface.

5. A method for solving the rock constitutive characteristic VG model parameters according to claim 4, characterized in that: Average suction at the wetting peak S f The relationship between the VG model parameters is as follows: ； In the formula, m =1-1 / n , m is the constitutive property fitting parameter, n is the pore size distribution parameter, l is the matrix suction fitting parameter, α is an empirical parameter, θ s is the saturated moisture content, θ r is the residual moisture content, that is, the percentage of residual moisture after the sample is dried. θ i is the initial moisture content of different samples.

6. A method for solving the rock constitutive characteristic VG model parameters according to claim 5, characterized in that: Infiltration rate f About the generalized moist front depth z f The function is: ； Determination of permeability coefficient through rock permeability and porosity tests K s And porosity ε , the initial moisture content of different samples was obtained by experimental fitting θ i Lower wetting peak matrix suction S fi .

7. A method for solving the rock constitutive characteristic VG model parameters according to claim 6, characterized in that: The specific process of S3 is as follows: Prepare a set of samples with different moisture contents and calculate their saturated moisture content θ s , residual moisture content θ r , initial moisture content of different samples θ i , conduct natural seepage experiments, assuming that the infiltration rate of the rock sample at each time interval during the seepage process is f Keep constant, record the mass growth of the rock sample, and the seepage rate per unit time is the infiltration rate f , by converting the weight gain into volume and dividing by the porosity ε , and obtain the generalized depth of the wetting peak z f ; when Z fi = S fi At this time f =2 K s , then we can obtain the wetting peak matrix suction S f1 ; Prepare another moisture content sample and repeat the above steps to obtain S f2 , S f3 , S f4 , according to the average suction at the wetting peak S f Relationship with VG model parameters: ； Four sets of expressions are obtained: ； In the formula m are constitutive model parameters, n =1-1 / m , l is the matrix suction parameter, and the other three are quantities to be determined; The number of equations is greater than the number of unknown quantities, and the unknown quantities, i.e. the required model parameters, are obtained by solving them using computers.

Images