Simulation Method for Critical Conditions of Gravity Erosion on Loess Slopes with Dynamic Moisture Content Correction
By using a dynamic strength model and a seepage-stress coupling framework, combined with multiple criteria, we have achieved accurate simulation of gravity erosion of loess slopes. This solves the problems of static strength parameters, insufficient coupling, and single criteria in existing technologies, and provides a reliable early warning threshold.
Patent Information
- Application Number
- CN202611384881.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-08
- Publication Date
- 2026-11-17
AI Technical Summary
Existing simulation methods cannot accurately reflect the nonlinear strength degradation caused by changes in soil moisture content due to rainfall. The fluid-structure interaction mechanism is insufficient, the determination of critical state is not reliable enough, and it is easy to misjudge the critical conditions of gravity erosion on loess slopes.
By calibrating the dynamic strength model and updating the shear strength parameters unit by unit, a seepage-stress bidirectional coupling and permeability coefficient dynamic feedback loop is constructed. Combined with the triple criteria of displacement mutation, plastic zone penetration and iteration non-convergence, the gravity erosion of loess slopes can be accurately simulated.
It improves the accuracy and reliability of gravity erosion simulation of loess slopes, outputs engineering-operable early warning thresholds, and supports early warning and disaster prevention decisions.
Smart Images

Figure CN123116608A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of slope engineering stability technology, specifically to a method for simulating critical conditions for gravity erosion of loess slopes with dynamic moisture content correction. Background Technology
[0002] Loess in the loess hilly and gully region has engineering characteristics such as strong structure, well-developed pores and vertical joints, and significant water sensitivity. High and steep slopes formed by power transmission, transportation and water conservancy projects are prone to shallow landslides, collapses or landslides due to gravity erosion under the action of rainfall infiltration. Therefore, it is necessary to reliably simulate its evolution process and trigger threshold.
[0003] However, existing simulation methods have three shortcomings:
[0004] In terms of strength parameter characterization, numerical simulation usually uses fixed cohesion and internal friction angle throughout the calculation process, which cannot reflect the nonlinear strength deterioration caused by the spatiotemporal changes in soil moisture content due to rainfall. In particular, it is difficult to characterize the accelerated decay law of shear strength in the range of 15% to 18% mass moisture content, which is not compatible with the water sensitivity of loess.
[0005] In terms of fluid-structure interaction mechanism, existing methods mostly use single seepage field calculation or only perform one-way coupling, which cannot form a two-way feedback between the strength deterioration and stress redistribution caused by changes in pore water pressure and matrix suction and the changes in pore structure and permeability parameters caused by soil deformation, thus severing the intrinsic linkage process between rainfall infiltration, strength attenuation and deformation failure.
[0006] In determining the critical state, relying solely on a single safety factor or a single numerical convergence state as the instability criterion lacks coordinated monitoring of precursors to failure, such as sudden changes in slope displacement and the penetration of the plastic zone. This can easily lead to misjudgments due to fluctuations in indicators or unusual values, thus affecting the reliability of critical state identification.
[0007] To address the aforementioned shortcomings, there is an urgent need in this field for a complete critical condition simulation method that addresses issues such as static strength parameters, insufficient coupling, single criteria, and fragmented processes, in order to achieve a more accurate, reliable, and engineering-applicable evaluation of gravity erosion on loess slopes. Summary of the Invention
[0008] This invention aims to address the technical deficiencies of existing technologies by providing a method for simulating critical conditions of gravity erosion on loess slopes with dynamic moisture content correction. By calibrating a dynamic strength model and updating shear strength parameters unit by unit, a seepage-stress bidirectional coupling and dynamic feedback loop of the permeability coefficient are constructed. This solves the problems of inaccurate characterization of the local moisture content impact of rainfall and insufficient fluid-structure interaction simulation. A triple criterion of displacement mutation, plastic zone penetration, and iterative non-convergence is used to jointly determine the critical state, reducing the risk of misjudgment based on a single criterion. Engineering thresholds such as critical moisture content, rainfall duration, and safety factor are output, providing quantifiable evidence for early warning of gravity erosion on loess slopes.
[0009] This invention provides the following technical solution: a method for simulating critical conditions for gravity erosion of loess slopes with dynamic moisture content correction, comprising the following steps:
[0010] S1. Slope numerical model construction: Obtain topographic mapping data, engineering geological data and rainfall boundary data of the loess slope to be analyzed, and establish a slope numerical model through a two-dimensional plane strain model.
[0011] S2. Dynamic strength model calibration: For loess samples of the loess slope to be analyzed, different mass moisture content and different dry density conditions are set up. Direct shear test and triaxial consolidation undrained test are carried out to determine the cohesion, internal friction angle and unsaturated hydraulic parameters under each condition. The dynamic strength model is calibrated based on the test results of cohesion and internal friction angle.
[0012] S3. Update the strength parameters of each element: Configure the dynamic strength model to each soil element of the slope numerical model, read the real-time mass moisture content and dry density of each soil element at each calculation step, update the cohesion and internal friction angle parameters of the soil element element by element, and use the updated parameters as the stress field solution input for the current calculation step.
[0013] S4. Two-way fluid-structure interaction calculation of seepage field and stress field: Based on the Biot consolidation relationship, a two-way fluid-stress coupling framework is established. The time step is set. Within the same calculation time step, the unsaturated seepage field is solved first to obtain pore water pressure, matrix suction and real-time mass water content. Then, it is input into step S3 to complete the unit strength parameter update and solve the stress field to obtain stress, displacement and pore structure changes. The permeability coefficient is then corrected according to the pore structure changes and fed back to the seepage field. The process is iterated until the preset convergence condition is reached.
[0014] S5. Critical state determination based on triple criteria: In each step of the two-way fluid-structure interaction calculation, the strength reduction calculation is performed using the Mohr-Coulomb elastoplastic constitutive relation. Simultaneously, the three criteria of slope displacement mutation, plastic zone penetration, and numerical iteration non-convergence are monitored. When all three criteria are met, the loess slope to be analyzed is determined to have reached the critical state of gravity erosion. The corresponding critical time, sliding zone location, and safety factor are recorded.
[0015] S6. Early warning threshold output: Extract the mass moisture content at the critical moment and the location of the sliding zone, and output the critical mass moisture content range, critical rainfall duration, and safety factor in combination with the corresponding slope gradient, dry density, and rainfall conditions.
[0016] Furthermore, the slope numerical model includes slope geometry, soil partitioning, grid, displacement boundary, and seepage boundary;
[0017] The two-dimensional plane strain model has a bidirectional displacement constraint at the bottom, a normal displacement constraint on the left and right sides, a variable rainfall infiltration flow boundary on the slope, a drainage boundary at the bottom of the model, and mesh refinement for the slope, vertical joints, and slope toe area.
[0018] Furthermore, in step S4, the unsaturated seepage field is solved:
[0019] The unsaturated seepage process on a slope under rainfall conditions can be solved using the mass conservation law, and can be expressed as follows: In the formula, θ is the volumetric water content, K(θ) is the unsaturated permeability coefficient which is a function of the volumetric water content, h is the pressure head, z is the position head, and q is the rainfall infiltration source term;
[0020] The relationship between matrix suction and water content is described using the Van Genuchten soil-water characteristic relationship:
[0021] In the formula, S e θ represents the effective saturation, s represents the matrix suction, and θ represents the effective saturation. r θ represents the residual volumetric water content. s The saturated volumetric water content is α, and the parameters of the soil-water characteristic curve are n and n.
[0022] The unsaturated permeability coefficient is determined based on the effective saturation: In the formula, K(S) e K is the unsaturated permeability coefficient, which is a function of the effective saturation. s The saturated permeability coefficient;
[0023] Convert the volumetric moisture content of each unit to mass moisture content:
[0024] In the formula, w i Let ρ be the real-time mass moisture content of the i-th soil element. w ρ is the density of water. d,i This refers to the real-time dry density of the unit.
[0025] In the formula, p w Let g be the pore water pressure, and g be the acceleration due to gravity.
[0026] Furthermore, in step S4, the unit-by-unit strength parameter update involves updating the real-time mass moisture content w of each soil unit. i and real-time dry density ρ d,i Input step S3 updates element by element to obtain the real-time cohesion c of each soil element in the current calculation step. i and real-time internal friction angle φ i :
[0027] ;
[0028] ;
[0029] In the formula, This is the dynamic strength model function of cohesion. This is the dynamic strength model function for the internal friction angle.
[0030] Furthermore, in step S4, the stress field is solved by calculating the stress field based on the Biot effective stress relationship, which is expressed as: And satisfy the mechanical equilibrium relationship: In the formula, σ is the total stress. The effective stress is I, the unit tensor is α. B Here, 'b' is the Biot coefficient, and 'b' is the volume force.
[0031] Solving for the nodal displacement u, within the finite element numerical framework, the mechanical equilibrium equations, the Mohr-Coulomb elastoplastic constitutive model, and the displacement boundary conditions are assembled to obtain a set of nonlinear finite element equations: , Here is the elastic-plastic stiffness matrix. F represents the nodal displacement increment. ext Let F be the external load vector. int Given the internal force vector, the Newton-Raphson iterative algorithm is used to solve the above nonlinear equations. After the iteration converges, the nodal displacement u of the current iteration step is obtained.
[0032] Solving for the element strain tensor from nodal displacements: In the formula For displacement gradient, The element strain tensor is input into the Mohr-Coulomb elastoplastic constitutive model to calculate the effective stress of the element. Substituting into the Biot effective stress relationship, we obtain the total stress σ.
[0033] Furthermore, in step S4, the permeability parameters are corrected based on the changes in pore structure and fed back to the seepage field: for the i-th element, the deformed Jacobian determinant J is solved. i Characterizes the degree of deformation of the unit volume. Update porosity Update dry density In the formula Let n be the displacement gradient of the i-th element. i The porosity after deformation. The initial porosity, The dry density after deformation. The initial dry density;
[0034] Using a Kozeny-Carman type relationship based on porosity variation, the expression is:
[0035] In the formula, k i Let k be the updated permeability coefficient of the i-th unit. i,0 Let be the initial permeability coefficient of the i-th unit.
[0036] Furthermore, the preset convergence condition in step S4 is expressed as follows:
[0037] ;
[0038] ;
[0039] ;
[0040] In the formula, , , These represent pore water pressure, permeability coefficient, and relative displacement error, respectively, with pref, kref, and uref serving as reference values. Preset convergence tolerance;
[0041] when When the current computation time step reaches convergence, the bidirectional fluid-structure interaction computation reaches convergence and proceeds to the next computation time step.
[0042] Furthermore, in step S5, the strength reduction calculation is as follows: Let the strength reduction factor be F, then the reduced cohesion of the i-th soil element is: ;
[0043] The reduced internal friction angle is: ;
[0044] In the formula, c i (t) Cohesion of the i-th soil element at time t after dynamic correction for water content. Let be the internal friction angle of the i-th soil element at time t after dynamic correction based on moisture content;
[0045] The reduction factor is gradually increased according to the set step size to reduce the cohesion and internal friction angle respectively. When the reduction factor increases to the point that the slope changes from stable to overall unstable, the critical reduction factor is the safety factor corresponding to the current rainfall time.
[0046] Furthermore, the abrupt change in slope displacement includes the increase in horizontal displacement at the top of the slope exceeding a preset displacement increment threshold between adjacent calculation steps; the connection of the plastic zone includes the connection between the plastic zone at the toe of the slope and the plastic zone in the upper middle part of the slope to form a continuous potential sliding zone; and the non-convergence of numerical iteration includes the failure of the residual to reach a preset convergence tolerance within a preset number of iterations.
[0047] Furthermore, in step S2, the loess sample is set with a moisture content of 8%-22% and a dry density of 1.40-1.70 g / cm³.
[0048] Compared with the prior art, the beneficial effects of the present invention are:
[0049] 1. Achieve dynamic updates of intensity parameters on a unit-by-unit basis, accurately responding to the non-uniformity of rainfall infiltration.
[0050] This invention calibrates the dynamic strength model of loess through indoor geotechnical tests (direct shear test and triaxial consolidated undrained test), establishes the quantitative relationship between cohesion, internal friction angle, and mass moisture content and dry density; and embeds this model into each soil element of the slope numerical model. In each calculation step, the system automatically reads the real-time mass moisture content and dry density of each element and updates the shear strength parameters independently for each element. This mechanism enables the local moisture content differences caused by rainfall to be directly and instantly mapped to the shear strength of the corresponding element, overcomes the defects of traditional methods that use uniform or empirically reduced parameters, and significantly improves the spatiotemporal resolution of the strength field description.
[0051] 2. Construct a closed-loop coupled framework of "seepage field - unit-by-unit update - stress field - permeability parameter feedback" to realistically reflect the evolution process of rainfall erosion.
[0052] Based on Biot's consolidation theory, a two-way fluid-structure interaction calculation system for seepage field and stress field is established. Within the same time step, unsaturated seepage solution, element-by-element strength update, stress field solution and dynamic correction of permeability coefficient are connected in series to form a complete closed feedback loop. This framework allows rainfall-induced water infiltration (seepage field) to first change the local water content, and then reduce the resistance of the corresponding element through strength update, thereby inducing stress redistribution and deformation. The changes in pore structure caused by deformation then reversely correct the permeability coefficient and affect subsequent seepage. This whole-process coupling mechanism can effectively capture the key characteristics of gravity erosion of loess slopes, making the simulation results closer to engineering reality.
[0053] 3. A triple-criteria approach is used to collaboratively determine critical states, effectively mitigating the risk of misjudgment.
[0054] Within each coupled calculation time step, strength reduction calculations are performed based on the Mohr-Coulomb elastoplastic constitutive relation, and three independent criteria—slope displacement abrupt change, plastic zone connectivity, and numerical iteration non-convergence—are monitored simultaneously. Only when all three criteria are met simultaneously is the slope determined to have reached the critical state of gravity erosion. This multi-index joint judgment strategy significantly reduces the misjudgment of the critical state caused by numerical disturbances or local non-uniformity due to a single criterion (such as relying solely on displacement abrupt change or relying solely on plastic zone connectivity), and improves the reliability and uniqueness of determining the critical moment, slip zone location, and safety factor.
[0055] 4. Outputs actionable early warning thresholds for engineering purposes, directly supporting monitoring design and disaster prevention decision-making.
[0056] The mass moisture content at the sliding zone location under critical conditions is extracted, and combined with slope gradient, dry density and rainfall conditions, the critical mass moisture content range, critical rainfall duration and corresponding safety factor are comprehensively output. This facilitates direct connection with on-site moisture content monitoring, rainfall early warning system and protection and reinforcement design standards, and provides a quantitative basis for early warning and proactive prevention and control of geological disasters on loess slopes.
[0057] In summary, this invention systematically improves the accuracy, robustness, and practicality of simulating critical conditions for gravity erosion on loess slopes from four aspects: parameter dynamization, coupling the entire process, multidimensional criteria, and engineering output. Attached Figure Description
[0058] Figure 1 This is a flowchart of the present invention;
[0059] Figure 2 This is a schematic diagram of the numerical model and boundary conditions for the slope in the example.
[0060] Figure 3 The diagram shows the relationship between dynamic strength updates per element and two-way fluid-structure interaction.
[0061] Figure 4 This is a timeline of events from rainfall infiltration to critical instability, as illustrated in the example. Detailed Implementation
[0062] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.
[0063] like Figure 1-4 As shown, it illustrates a specific embodiment of the present invention:
[0064] Example 1:
[0065] Slope of a 1000kV ultra-high voltage transmission line tower foundation in the loess hilly and gully area of western Shanxi: This example focuses on the excavated slope of a 1000kV ultra-high voltage transmission line tower foundation in the loess hilly and gully area of western Shanxi. The slope is composed of Q3 Malan loess, with a measured slope height of 12m and a slope angle of 75°. Vertical joints are well-developed, and vertical cracks with a width of 2-5mm exist. The initial dry density is 1.55g / cm³, and the initial surface moisture content is 9.2%.
[0066] A method for simulating critical conditions for gravity erosion of loess slopes with dynamic moisture content correction includes the following steps: Figure 1 As shown:
[0067] S1. Slope numerical model construction: Obtain topographic mapping data, engineering geological data and rainfall boundary data of the loess slope to be analyzed, and establish a slope numerical model through a two-dimensional plane strain model.
[0068] The slope numerical model includes slope geometry, soil partitioning, grid, displacement boundary, and seepage boundary;
[0069] The two-dimensional plane strain model has a bidirectional displacement constraint at the bottom, a normal displacement constraint on the left and right sides, a variable rainfall infiltration flow boundary on the slope, a drainage boundary at the bottom of the model, and mesh refinement for the slope, vertical joints, and slope toe area.
[0070] like Figure 2 As shown, in this embodiment, the model has a total width of 21.2m, a total height of 20m, a slope height of 12m, a horizontal section length of 8m at the top of the slope, a horizontal section length of 10m at the bottom of the slope, and a slope angle of 75° (horizontal projection of the slope surface is 3.2m). Bidirectional displacement constraints are applied at the bottom, and normal displacement constraints are applied on the left and right sides. A variable rainfall infiltration flow boundary is set on the slope surface, simulating a rainfall intensity of 0.6mm / min and a duration of 0–8h. A free drainage boundary is set at the bottom, and a permeable boundary is set on the top surface and the non-infiltration area of the slope surface. A quadrilateral structured mesh is used, with densification applied to the slope surface, vertical joints, and the bottom of the slope. The minimum element size is 0.2m, with a total of 12,860 elements and 13,124 nodes.
[0071] S2. Dynamic strength model calibration: For loess samples of the loess slope to be analyzed, different mass moisture content and different dry density conditions are set up. Direct shear test and triaxial consolidation undrained test are carried out to determine the cohesion, internal friction angle and unsaturated hydraulic parameters under each condition. The dynamic strength model is calibrated based on the test results of cohesion and internal friction angle.
[0072] Uncirculated loess samples were collected on-site, and five sets of dry density and six sets of moisture content gradients were prepared, totaling 30 sets of test data. Direct shear tests and triaxial consolidated undrained tests were conducted to determine the cohesion c and internal friction angle under various working conditions. The hydraulic parameters required for the model were measured or calibrated. The fitted sample data are shown in Table 1, and the experimentally measured values of cohesion and internal friction angle under each working condition are listed in full; Table 2 shows the geotechnical test design and parameter acquisition for the example, and Table 3 shows an example of dynamic strength parameters at a dry density of 1.55 g / cm³.
[0073] Table 1:
[0074]
[0075] Based on experimental data, multivariate nonlinear fitting, correlation verification, and residual correction were used to obtain the dynamic correction relationships for cohesion and internal friction angle adopted in this embodiment:
[0076]
[0077]
[0078] In the formula, ρ d ρ is the dry density of loess, in g / cm³; w is the mass moisture content, in %; c is the real-time cohesion, in kPa. is the real-time internal friction angle, in °; exp(·) is the natural exponential function.
[0079] Cohesion model coefficient of determination: ;
[0080] The coefficient of determination for the internal friction angle model: ;
[0081] In the formula, N is the number of data sets participating in fitting and validation; in this embodiment, N=30; c i and These are the measured cohesion and internal friction angle of the i-th group of experiments, respectively. and These are the calculated values corresponding to the dynamic intensity model; and These are the average values of the cohesion and internal friction angle of all tests, respectively.
[0082] By substituting each of the 30 sets of experimentally measured data into the above formula, and statistically analyzing the residuals between the measured values and the model-calculated values, the coefficient of determination R for cohesion fitting was obtained. c The coefficient of determination R² for the internal friction angle is 0.994. φ The coefficient of determination R² is 0.975. 2 A value close to 1 indicates a good fit.
[0083] Table 2:
[0084]
[0085] Table 3:
[0086]
[0087] Determination of unsaturated hydraulic parameters:
[0088] (1) Saturated water content θ s The saturated moisture content of undisturbed loess samples was determined using the ring cutter method. The ring cutter sample was placed in a saturator, saturated with air (vacuum degree -100 kPa, for 24 hours), weighed, and the saturated moisture content was calculated.
[0089] (2) Residual moisture content θᵣ: The desiccation branch of the soil-water characteristic curve was determined by the pressure plate method. The volumetric moisture content corresponding to the increase of matrix suction to 1500 kPa was taken as the residual moisture content.
[0090] (3) VanGenuchten model parameters α and n: The volumetric water content data corresponding to different matrix suction (0, 5, 10, 20, 50, 100, 200, 500, 1000, 1500 kPa) were obtained by pressure plate test, and the VanGenuchten model was fitted by nonlinear least squares method.
[0091] In the formula, ψ represents the matrix suction (kPa). The optimal estimates of α and n are obtained by fitting.
[0092] (4) Saturated permeability coefficient Ks: Determined by variable head permeability test. The undisturbed loess sample was loaded into the permeameter, and the permeation flow rate under different head differences was measured under saturated conditions. The saturated permeability coefficient was calculated according to Darcy's law.
[0093] In this embodiment, the unsaturated hydraulic parameters obtained by the above method are θᵣ=0.068 and θ s =0.452, α=0.012kPa -1 n=1.45, K s =8.2×10 -7 m / s.
[0094] S3. Update the strength parameters of each element: Configure the dynamic strength model to each soil element of the slope numerical model, read the real-time mass moisture content and dry density of each soil element at each calculation step, update the cohesion and internal friction angle parameters of the soil element element by element, and use the updated parameters as the stress field solution input for the current calculation step.
[0095] like Figure 3As shown, the calculation time step is set to Δt = 10 min. After each seepage substep, the real-time mass moisture content and dry density of each unit are read. The dynamic correction relationship of this embodiment or the equivalent dynamic strength relationship obtained from the same set of test data are called to update the cohesion and internal friction angle of each soil unit respectively. The update results are written into the current material state of the soil unit, and the uniform parameters for the entire slope are not used.
[0096] S4. Two-way fluid-structure interaction calculation of seepage field and stress field: Based on the Biot consolidation relationship, a two-way fluid-stress coupling framework is established. The time step is set. Within the same calculation time step, the unsaturated seepage field is solved first to obtain pore water pressure, matrix suction and real-time mass water content. Then, it is input into step S3 to complete the unit strength parameter update and solve the stress field to obtain stress, displacement and pore structure changes. The permeability coefficient is then corrected according to the pore structure changes and fed back to the seepage field. The process is iterated until the preset convergence condition is reached.
[0097] (1) Solving for the unsaturated seepage field:
[0098] The unsaturated seepage process on a slope under rainfall conditions can be solved using the mass conservation law, and can be expressed as follows: In the formula, θ is the volumetric water content, K(θ) is the unsaturated permeability coefficient which is a function of the volumetric water content, h is the pressure head, z is the position head, and q is the rainfall infiltration source term;
[0099] The relationship between matrix suction and water content is described using the Van Genuchten soil-water characteristic relationship:
[0100] In the formula, S e Where s is the effective saturation and s is the matrix suction.
[0101] θ r θ represents the residual volumetric water content. s The saturated volumetric water content is α, and the parameters of the soil-water characteristic curve are n and n.
[0102] The unsaturated permeability coefficient is determined based on the effective saturation: In the formula, K(S) e K is the unsaturated permeability coefficient, which is a function of the effective saturation. s The saturated permeability coefficient;
[0103] Convert the volumetric moisture content of each unit to mass moisture content:
[0104] In the formula, w i Let ρ be the real-time mass moisture content of the i-th soil element. w ρ is the density of water. d,i This refers to the real-time dry density of the unit.
[0105] In the formula, p w Let g be the pore water pressure, and g be the acceleration due to gravity.
[0106] (2) Update the strength parameters of each unit: update the real-time mass moisture content w of each soil unit. i and real-time dry density ρ d,i Input step S3 updates element by element to obtain the real-time cohesion c of each soil element in the current calculation step. i and real-time internal friction angle φ i :
[0107] ;
[0108] ;
[0109] In the formula, This is the dynamic strength model function of cohesion. This is the dynamic strength model function for the internal friction angle.
[0110] (3) Solve for the stress field: Calculate the stress field according to the Biot effective stress relationship. The relationship is as follows: And satisfy the mechanical equilibrium relationship: In the formula, σ is the total stress. The effective stress is I, the unit tensor is α. B Here, 'b' is the Biot coefficient, and 'b' is the volume force.
[0111] Solving for the nodal displacement u, within the finite element numerical framework, the mechanical equilibrium equations, the Mohr-Coulomb elastoplastic constitutive model, and the displacement boundary conditions are assembled to obtain a set of nonlinear finite element equations: , Here is the elastic-plastic stiffness matrix. F represents the nodal displacement increment. ext Let F be the external load vector. int Given the internal force vector, the Newton-Raphson iterative algorithm is used to solve the above nonlinear equations. After the iteration converges, the nodal displacement u of the current iteration step is obtained.
[0112] Solving for the element strain tensor from nodal displacements: In the formula For displacement gradient, The element strain tensor is input into the Mohr-Coulomb elastoplastic constitutive model to calculate the effective stress of the element. Substituting into the Biot effective stress relationship, we obtain the total stress σ.
[0113] (4) Correct the permeability parameters based on the changes in pore structure and feed them back to the seepage field: For the i-th element, solve the deformed Jacobian determinant J. iCharacterizes the degree of deformation of the unit volume. Update porosity Update dry density In the formula Let n be the displacement gradient of the i-th element. i The porosity after deformation. The initial porosity, The dry density after deformation. The initial dry density;
[0114] Using a Kozeny-Carman type relationship based on porosity variation, the expression is:
[0115] In the formula, k i Let k be the updated permeability coefficient of the i-th unit. i,0 Let be the initial permeability coefficient of the i-th unit.
[0116] The preset convergence condition is expressed as:
[0117] ;
[0118] ;
[0119] ;
[0120] In the formula, , , These represent pore water pressure, permeability coefficient, and relative displacement error, respectively, with pref, kref, and uref serving as reference values. Preset convergence tolerance;
[0121] when When the current computation time step reaches convergence, the bidirectional fluid-structure interaction computation reaches convergence and proceeds to the next computation time step.
[0122] This embodiment establishes a seepage-stress bidirectional coupling framework based on the Biot consolidation relationship and uses Van Genuchten soil-water characteristic curves to describe the hydraulic properties of unsaturated loess. Site parameters are: residual water content θᵣ = 0.068, saturated water content θ s =0.452, intake air value parameter α=0.012kPa -1 Pore distribution index n=1.45, saturated permeability coefficient K s =8.2×10 -7 m / s. At each time step, the seepage field is solved first, and the water content is passed to step S3 to complete the element-by-element update; then the stress field is solved, and the soil deformation and void ratio change are converted into permeability coefficient correction values and fed back to the seepage field, iterating until convergence.
[0123] After 3 hours of rainfall, dominant flow channels formed at the vertical joints of the slope. The water content at a depth of 2-4 meters within the joints first rose to 18.3%, and the matrix suction decreased from an initial approximately 3167 kPa to 269 kPa. This result serves as the state variable transferred from the coupled calculation in step S4 to the critical determination in step S5. Table 4 shows the calculation data for the example. Figure 4 This is a timeline of rainfall infiltration to the critical instability event, as shown in the example.
[0124] Table 4:
[0125]
[0126] S5. Critical state determination based on triple criteria: In each step of the two-way fluid-structure interaction calculation, the strength reduction calculation is performed using the Mohr-Coulomb elastoplastic constitutive relation. Simultaneously, the three criteria of slope displacement mutation, plastic zone penetration, and numerical iteration non-convergence are monitored. When all three criteria are met, the loess slope to be analyzed is determined to have reached the critical state of gravity erosion. The corresponding critical time, sliding zone location, and safety factor are recorded.
[0127] (1) Strength reduction calculation: Let the strength reduction factor be F, then the cohesion of the i-th soil element after reduction is: ;
[0128] The reduced internal friction angle is: ;
[0129] In the formula, c i (t) Cohesion of the i-th soil element at time t after dynamic correction for water content. Let be the internal friction angle of the i-th soil element at time t after dynamic correction based on moisture content;
[0130] The reduction factor is gradually increased according to the set step size to reduce the cohesion and internal friction angle respectively. When the reduction factor increases to the point that the slope changes from stable to overall unstable, the critical reduction factor is the safety factor corresponding to the current rainfall time.
[0131] After each fluid-structure interaction time step is completed, strength reduction calculations are performed starting from F=1.00, and the original implementation is set as follows: ΔF=0.02, and the calculations are performed sequentially using: 1.00, 1.02, 1.04, 1.06, ...
[0132] When one reduction factor remains stable while the next reduction factor begins to show overall instability, this critical reduction factor is determined as the safety factor corresponding to the current rainfall moment. To further improve the accuracy of the safety factor calculation, the interval between adjacent stable and unstable reduction factors can be further refined.
[0133] (2) Criterion for sudden displacement
[0134] Record the horizontal displacement of the monitoring point at the top of the slope at each time step: u top (t n The horizontal displacement increment between adjacent calculation steps is: When the displacement increment exceeds a preset displacement increment threshold and exhibits significant nonlinear acceleration, the displacement mutation criterion is considered met. In this embodiment, at the critical moment, the horizontal displacement at the top of the slope suddenly increases from 0.8 mm to 12.6 mm, with a horizontal displacement increment of 11.8 mm, demonstrating a significant displacement mutation. Table 5, the combined determination record of the three criteria, has recorded this result.
[0135] Table 5:
[0136]
[0137] (3) Determination of the location of the plastic zone and the sliding zone
[0138] The Mohr-Coulomb yield state is calculated for each finite element. When the plastic elements near the toe of the slope and the plastic elements in the upper part of the slope gradually expand and form a set of plastic elements that are spatially connected, the plastic zone is determined to be connected.
[0139] This set of continuous plastic elements is represented as Ω p The continuously distributed high plastic shear strain region is considered as the potential slip zone Ω. slip The starting point, ending point, maximum burial depth, and spatial location of the potential sliding zone can be obtained from the spatial coordinates of the plastic zone connecting units. Therefore, the "sliding zone location" is not arbitrarily determined manually, but is determined based on the continuous connection state of the plastic units at the critical moment. In this embodiment, only a local plastic yielding zone appears at the toe of the slope after 3.8 hours of rainfall, and no connection has yet been formed; after 6.8 hours of rainfall, the middle and upper plastic zones form a continuous connection with the plastic zone at the toe of the slope, thus this continuous plastic zone is determined to be the critical potential sliding zone.
[0140] (4) Criterion for numerical non-convergence
[0141] Record the calculated residual R and the number of iterations N during each nonlinear solution process. When the preset maximum number of iterations N=N is reached... max The condition R > R is still satisfied. cr If the residual cannot be reduced to within the specified convergence tolerance, then the calculation is deemed not to converge under the current intensity state.
[0142] (5) Determination of critical time
[0143] Perform the above intensity reduction calculation for each rainfall moment to obtain the change of the safety factor with rainfall duration: F S (t1),F S (t2),⋯,F S (t n)
[0144] As rainfall continues to infiltrate, the slope's moisture content increases, the matrix suction decreases, and the shear strength weakens, resulting in a gradual decrease in the safety factor.
[0145] The critical moment is defined as the time when the following conditions are simultaneously met:
[0146] A sudden change occurred in the horizontal displacement at the top of the slope;
[0147] The plastic zone at the toe of the slope is connected to the plastic zone in the upper middle part of the slope.
[0148] Numerical iterations failed to reach the specified convergence tolerance;
[0149] The safety factor corresponding to the original strength state is approximately 1.00.
[0150] In this embodiment, only localized dominant flow occurred at 3.0h, and no plastic zone was formed. At 3.8h, a localized plastic zone appeared at the toe of the slope, but the calculation still converged. At 6.8h, the horizontal displacement at the top of the slope suddenly increased from 0.8mm to 12.6mm, the plastic zone in the middle and upper part connected with the plastic zone at the toe of the slope, and numerical non-convergence occurred. Therefore, all three criteria were met simultaneously for the first time, and it was determined that t cr =6.8h corresponds to: F Scr =1.00, meaning 6.8h is the critical time for gravity erosion of the slope under the above rainfall conditions. Table 3 has recorded the criteria status at the three stages of 3.0h, 3.8h, and 6.8h.
[0151] S6. Early warning threshold output: Extract the mass moisture content at the critical moment and the location of the sliding zone, and output the critical mass moisture content range, critical rainfall duration, and safety factor in combination with the corresponding slope gradient, dry density, and rainfall conditions.
[0152] The mass moisture content of each unit in the potential sliding zone at the critical moment of 6.8 hours was extracted, yielding a critical mass moisture content range of 15.6%–17.8% for this slope. Combined with a slope angle of 75°, dry density of 1.55 g / cm³, and rainfall intensity of 0.6 mm / min, the critical state and its operating conditions were output as the engineering early warning threshold. To verify the duration of critical rainfall, the following method was used: In the formula, T c For the critical rainfall duration, α is the slope gradient, and ρ is the slope gradient. d For dry density, w c The mean critical moisture content is given by , and w0 is the initial moisture content. The calculated time is 6.7 h, which has a relative error of 1.5% compared to the 6.8 h obtained from the two-way fluid-structure interaction numerical simulation. This formula is only used for verifying the threshold of the specific embodiment. Table 6 shows a comparison between the warning threshold output and verification of the specific embodiment:
[0153] Table 6:
[0154]
[0155] Table 6 shows that in-situ monitoring indicated that a small, shallow landslide occurred 6.5 hours after the measured rainfall intensity of 0.58 mm / min. The results of this method at 6.8 hours showed a 95.6% agreement with the monitoring data, while the results of the traditional fixed-parameter method at 9.1 hours showed a 71.4% agreement. This verifies the feasibility of this method.
[0156] For slopes of 60°, 75°, and 90°, initial dry density of 1.40–1.70 g / cm³, rainfall intensity of 0.2–1.0 mm / min, and rainfall duration of 1–12 h, S1 to S6 are repeated. The critical mass moisture content range, critical rainfall duration, safety factor, and triple criterion status under each condition can be summarized into a set of warning thresholds according to the condition index. In actual engineering applications, the parameters should be recalibrated using the samples and monitoring conditions of the corresponding project.
[0157] Example 2:
[0158] Excavated slope of Q3 Malan loess highway cut in Yan'an, Shaanxi (slope 60°): The research object of this embodiment is an excavated loess slope of a highway cut in Yan'an, Shaanxi. The stratum is Q3 Malan loess, the slope height is 14m, the slope angle is 60°, and the vertical joints are moderately developed. The initial dry density of the soil measured on site is 1.45g / cm³, and the initial mass moisture content of the surface layer is 8.5%. This slope is prone to shallow landslides and gravity erosion disasters under the action of rainfall. The critical condition simulation method of gravity erosion of loess slope with dynamic moisture content correction of the present invention is used to carry out numerical simulation by fully executing the S1 to S6 process.
[0159] S1. Construction of Numerical Model for Slope
[0160] Obtain the topographic mapping results, engineering geological survey report, and regional rainfall monitoring statistics of the target slope, and establish a numerical model of the slope using a two-dimensional plane strain model.
[0161] Model geometry: Total width 26.1m, total height 22m; slope height 14m, horizontal section length at the top 9m, horizontal section length at the bottom 9m, slope angle 60°. Boundary settings: Bidirectional displacement constraint at the bottom of the model, normal displacement constraint on the left and right sides; variable rainfall infiltration flow boundary set on the slope surface, simulating rainfall intensity of 0.4mm / min; free drainage boundary set at the bottom of the model; permeable boundary set at the top of the slope and non-infiltration area. Mesh refinement was applied to the slope surface, bottom of the slope, and jointed zones, with a minimum element size of 0.2m; Mesh statistics: 13420 elements in total, 13706 nodes in total.
[0162] S2, Dynamic Strength Model Calibration
[0163] Uncirculated Q3 Malan loess samples were collected on-site, and 30 working conditions were set up with 5 dry densities (1.40, 1.45, 1.50, 1.60, 1.70 g / cm³) and 6 moisture contents (8%, 12%, 15%, 18%, 20%, 22%). Direct shear tests and triaxial consolidated undrained tests were conducted to determine the cohesion, internal friction angle, and unsaturated hydraulic parameters for each group. Based on the test data, a dynamic strength model coupled with moisture content and dry density was calibrated. Table 7 shows all sample data from the geotechnical tests and model fitting in Example 2.
[0164] Table 7:
[0165]
[0166] Example 2: Dynamic intensity fitting equation:
[0167]
[0168]
[0169] Model fit goodness: cohesion R c ² = 0.997; Angle of internal friction R φ ² = 0.985.
[0170] Unsaturated hydraulic parameter calibration results: residual volumetric water content θr = 0.062, saturated volumetric water content θs = 0.436, Van Genuchten inlet parameters α = 0.0105 kPa⁻¹, porosity distribution index n = 1.42, saturated permeability coefficient Ks = 6.7 × 10⁻ 7 m / s.
[0171] S3, Update element-by-element strength parameters
[0172] Set the calculation time step Δt=10min; after each seepage substep is completed, read the real-time mass moisture content and real-time dry density of each soil element, call the dynamic strength model of this embodiment, and update the soil cohesion and internal friction angle of each element; the updated parameters are directly assigned to the corresponding soil element as the input for the stress field solution of this time step, and the uniform strength parameters of the whole slope are not used.
[0173] S4. Two-way fluid-structure interaction calculation of seepage field and stress field
[0174] A two-way coupled seepage-stress framework was constructed based on Biot's consolidation theory. Within the same calculation time step, the unsaturated seepage field was first solved to obtain pore water pressure, matrix suction, and real-time mass water content of the elements. The output water content parameter was then transferred to S3 to update the shear strength of each element. Next, the stress field was solved to obtain slope stress, nodal displacement, and element pore structure deformation. Based on the Kozeny-Carman formula, the permeability coefficient was corrected according to the change in element porosity, and the corrected permeability coefficient was fed back to the seepage field. This process was iterated until the convergence criteria were met. Table 8 shows the key time step coupled calculation status data for Example 2.
[0175] Table 8:
[0176]
[0177] S5, Triple Criterion Critical State Determination
[0178] After each fluid-structure interaction calculation step is completed, strength reduction calculations are performed based on the Mohr-Coulomb elastoplastic constitutive model, with a reduction coefficient step size ΔF=0.02. Simultaneously, three criteria are monitored: abrupt slope displacement, plastic zone penetration, and numerical iteration non-convergence. When all three criteria are simultaneously satisfied, the slope is determined to have reached the critical state of gravity erosion, and the critical moment, spatial location of the slip zone, and safety factor are recorded. Table 9 shows the complete record of the three criteria in Example 2.
[0179] Table 9:
[0180]
[0181] The critical time t in this embodiment cr =5.7h, critical safety factor F Scr =1.00; The position of the sliding zone is determined by the spatial coordinates of the through plastic element set.
[0182] S6, Warning Threshold Output
[0183] The mass moisture content of all units in the sliding zone at the critical moment was extracted; combined with the working conditions of a slope of 60°, initial dry density of 1.45 g / cm³, and rainfall intensity of 0.4 mm / min, the usable early warning threshold for the project was output; and compared with the field measured disaster records and the results of traditional fixed parameter simulation. As shown in Table 10, the comparison table of early warning threshold output and verification in Example 2 is presented.
[0184] Table 10:
[0185]
[0186] The simulated critical rainfall duration of this invention was 5.7 hours, while the actual field measurement was 5.4 hours, with a relative error of 5.6%, proving that this method has high simulation accuracy for loess slopes with a moderate slope of 60°.
[0187] Example 3:
[0188] Nearly vertical cut slope (90° collapse-type slope) on the edge of Q3 loess plateau in Dongzhiyuan, Longdong: The object of this embodiment is an artificially excavated vertical loess cut slope on the edge of Dongzhiyuan, Longdong. The stratum is Q3 Malan loess, the slope height is 10m, the slope is nearly vertical at 90°, and the vertical fissures are extremely well developed; the initial dry density measured on site is 1.60g / cm³, and the initial mass moisture content of the surface layer is 8.8%; under rainfall conditions, this type of slope is very prone to gravity erosion disasters in the form of collapse; all simulation steps S1 to S6 of this invention are fully implemented.
[0189] S1. Construction of Numerical Model for Slope
[0190] Based on on-site topographic mapping, geological survey, and rainfall monitoring data, a two-dimensional plane strain numerical model was constructed. The model has a total width of 28m, a total height of 18m, a slope height of 10m, and a vertical slope of 90°. Boundary conditions: bidirectional displacement constraints at the bottom of the model, and normal displacement constraints on the left and right sides; rainfall infiltration flow boundary set on the slope surface, with a rainfall intensity of 0.3mm / min; free drainage boundary at the bottom; and a permeable boundary set in the non-infiltrative zone at the top of the slope. The mesh was refined for the vertical slope surface, vertical fissure zone, and slope toe area, with a minimum element size of 0.15m; mesh statistics: a total of 14152 elements and 14437 nodes.
[0191] S2, Dynamic Strength Model Calibration
[0192] Uncirculated loess samples were collected, with 5 dry densities (1.40, 1.50, 1.60, 1.65, 1.70 g / cm³) and 6 moisture contents (8%, 12%, 15%, 18%, 20%, 22%), totaling 30 samples. Direct shear and triaxial consolidated undrained tests were conducted to determine shear strength parameters and unsaturated hydraulic parameters, and the dynamic strength model was calibrated. The data are shown in Table 11, Example 3, Geotechnical Test and Model Fitting Data Table.
[0193] Table 11:
[0194]
[0195] Example 3: Dynamic intensity fitting equation:
[0196]
[0197]
[0198] Goodness of fit: cohesion R c ² = 0.997; Angle of internal friction R φ ² = 0.987.
[0199] Unsaturated hydraulic parameter calibration results: residual volumetric water content θᵣ=0.071, saturated volumetric water content
[0200] θ s =0.461, α=0.013kPa -1 n=1.47, saturated permeability coefficient K s =9.1×10⁻ 7 m / s.
[0201] S3, Update element-by-element strength parameters
[0202] Set the calculation time step Δt=10min; after each seepage substep is solved, read the real-time mass moisture content and real-time dry density of each element; call the dynamic strength model of this embodiment to update the cohesion and internal friction angle element by element; assign the updated parameters to the corresponding soil element as stress field input; the moisture content of the vertical crack element changes drastically, and the shear strength will deteriorate rapidly.
[0203] S4. Two-way fluid-structure interaction calculation of seepage field and stress field
[0204] A two-way coupled framework is built based on Biot consolidation theory; the unsaturated seepage field is solved first at the same time step to obtain pore water pressure, matrix suction, and element water content; the water content is transferred to S3 to complete the strength update; the stress field is solved to obtain stress, displacement, and element pore deformation; the permeability coefficient is corrected according to the Kozeny-Carman relationship based on the porosity change, and the seepage field is fed back and iterated until convergence, as shown in Table 12, the key time step coupled calculation state data table of Example 3.
[0205] Table 12:
[0206]
[0207] S5, Triple Criterion Critical State Determination
[0208] Strength reduction is carried out at each coupling time step, with a reduction step size ΔF=0.02; displacement abrupt change, plastic zone connection, and iteration non-convergence are monitored simultaneously; when all three criteria are met, the slope is judged to enter the critical state of gravity erosion (collapse), and the critical moment, sliding zone position, and safety factor are recorded, as shown in Table 13 Example 3 Triple Criterion Complete Judgment Record Table.
[0209] Table 13:
[0210]
[0211] Critical time t cr =2.9h, critical safety factor F Scr =1.00; The position of the sliding zone is determined by the spatial coordinates of the through plastic element set.
[0212] S6, Warning Threshold Output
[0213] Extract the mass moisture content of all units in the sliding zone at the critical moment; combine the working conditions of vertical slope cutting 90°, initial dry density 1.60g / cm³, and rainfall intensity 0.3mm / min to output the engineering early warning threshold, and compare it with the on-site collapse monitoring records at the edge of the plateau and the simulation results of traditional fixed parameters, as shown in Table 14, Example 3 Early Warning Threshold Output and Verification Comparison Table.
[0214] Table 14:
[0215]
[0216] This invention simulates a critical rainfall duration of 2.9 hours, while the actual measured collapse time in the field is 2.7 hours, with a relative error of 7.4%. For loess slopes with vertical fissures, traditional methods with fixed strength parameters significantly overestimate the duration of critical rainfall. This invention's unit-by-unit dynamic updating strength bidirectional fluid-structure interaction-triple criterion joint judgment system can significantly improve the simulation accuracy of critical conditions for gravity erosion in vertical slope collapse.
[0217] Examples 2 and 3 address two typical loess engineering slopes: a 60° highway cutting slope and a 90° vertical cut slope. All steps S1 to S6 are executed completely. The dynamic strength model is calibrated indoors, shear strength is updated element by element, seepage-stress bidirectional feedback coupling is used, and a triple criterion of displacement-plastic zone-iteration non-convergence is applied to determine the critical moisture content range, critical rainfall duration, and safety factor engineering early warning threshold. Compared with the actual measured disaster occurrence time in the field, the relative error is less than 8%. In contrast, the simulation accuracy using traditional fixed strength parameters is only 66%–71%, further demonstrating that the method of this invention has good applicability and computational reliability for loess slopes with different slopes and dry densities. In practical engineering applications, undisturbed loess samples should be collected for the corresponding engineering project to re-perform geotechnical testing and model calibration; the fitting formula of this example should not be directly reused.
[0218] The preferred embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention. These changes involve related technologies well known to those skilled in the art, and all of them fall within the protection scope of the present invention.
[0219] Many other changes and modifications can be made without departing from the concept and scope of this invention. It should be understood that this invention is not limited to the specific embodiments, and the scope of this invention is defined by the appended claims.
Claims
1. A method for simulating critical conditions for gravity erosion of loess slopes with dynamic moisture content correction, characterized in that, Includes the following steps: S1. Slope numerical model construction: Obtain topographic mapping data, engineering geological data and rainfall boundary data of the loess slope to be analyzed, and establish a slope numerical model through a two-dimensional plane strain model. S2. Dynamic strength model calibration: For loess samples of the loess slope to be analyzed, different mass moisture content and different dry density conditions are set up. Direct shear test and triaxial consolidation undrained test are carried out to determine the cohesion, internal friction angle and unsaturated hydraulic parameters under each condition. The dynamic strength model is calibrated based on the test results of cohesion and internal friction angle. S3. Update the strength parameters of each element: Configure the dynamic strength model to each soil element of the slope numerical model, read the real-time mass moisture content and dry density of each soil element at each calculation step, update the cohesion and internal friction angle parameters of the soil element element by element, and use the updated parameters as the stress field solution input for the current calculation step. S4. Two-way fluid-structure interaction calculation of seepage field and stress field: Based on the Biot consolidation relationship, a two-way fluid-stress coupling framework is established. The time step is set. Within the same calculation time step, the unsaturated seepage field is solved first to obtain pore water pressure, matrix suction and real-time mass water content. Then, it is input into step S3 to complete the unit strength parameter update and solve the stress field to obtain stress, displacement and pore structure changes. The permeability coefficient is then corrected according to the pore structure changes and fed back to the seepage field. The process is iterated until the preset convergence condition is reached. S5. Critical state determination based on triple criteria: In each step of the two-way fluid-structure interaction calculation, the strength reduction calculation is performed using the Mohr-Coulomb elastoplastic constitutive relation. Simultaneously, the three criteria of slope displacement mutation, plastic zone penetration, and numerical iteration non-convergence are monitored. When all three criteria are met, the loess slope to be analyzed is determined to have reached the critical state of gravity erosion. The corresponding critical time, sliding zone location, and safety factor are recorded. S6. Early warning threshold output: Extract the mass moisture content at the critical moment and the location of the sliding zone, and output the critical mass moisture content range, critical rainfall duration, and safety factor in combination with the corresponding slope gradient, dry density, and rainfall conditions.
2. The method for simulating critical conditions for gravity erosion of loess slopes with dynamic moisture content correction according to claim 1, characterized in that, The slope numerical model includes slope geometry, soil partitioning, grid, displacement boundary, and seepage boundary; The two-dimensional plane strain model has a bidirectional displacement constraint at the bottom, a normal displacement constraint on the left and right sides, a variable rainfall infiltration flow boundary on the slope, a drainage boundary at the bottom of the model, and mesh refinement for the slope, vertical joints, and slope toe area.
3. The method for simulating critical conditions for gravity erosion of loess slopes with dynamic moisture content correction according to claim 1, characterized in that, The solution of the unsaturated seepage field in step S4: The unsaturated seepage process on a slope under rainfall conditions can be solved using the mass conservation law, and can be expressed as follows: In the formula, θ is the volumetric water content, K(θ) is the unsaturated permeability coefficient which is a function of the volumetric water content, h is the pressure head, z is the position head, and q is the rainfall infiltration source term; The relationship between matrix suction and water content is described using the Van Genuchten soil-water characteristic relationship: In the formula, S e θ represents the effective saturation, s represents the matrix suction, and θ represents the effective saturation. r θ represents the residual volumetric water content. s The saturated volumetric water content is α, and the parameters of the soil-water characteristic curve are n and n. The unsaturated permeability coefficient is determined based on the effective saturation: In the formula, K(S) e K is the unsaturated permeability coefficient, which is a function of the effective saturation. s The saturated permeability coefficient; Convert the volumetric moisture content of each unit to mass moisture content: In the formula, w i Let ρ be the real-time mass moisture content of the i-th soil element. w ρ is the density of water. d,i This refers to the real-time dry density of the unit. In the formula, p w Let g be the pore water pressure and g be the acceleration due to gravity.
4. The method for simulating critical conditions for gravity erosion of loess slopes with dynamic moisture content correction according to claim 3, characterized in that, The element-by-element strength parameter update in step S4 involves updating the real-time mass moisture content w of each soil element. i and real-time dry density ρ d,i Input step S3 to update element by element, and obtain the real-time cohesion c of each soil element in the current calculation step. i and real-time internal friction angle φ i : ; ; In the formula, This is the dynamic strength model function of cohesion. This is the dynamic strength model function for the internal friction angle.
5. The method for simulating critical conditions for gravity erosion of loess slopes with dynamic moisture content correction according to claim 4, characterized in that, The stress field solution in step S4 involves calculating the stress field based on the Biot effective stress relationship, which is expressed as follows: And satisfy the mechanical equilibrium relationship: In the formula, σ is the total stress. The effective stress is I, the unit tensor is α. B Here, 'b' is the Biot coefficient, and 'b' is the volume force. Solving for the nodal displacement u, within the finite element numerical framework, the mechanical equilibrium equations, the Mohr-Coulomb elastoplastic constitutive model, and the displacement boundary conditions are assembled to obtain a set of nonlinear finite element equations: , Here is the elastic-plastic stiffness matrix. F represents the nodal displacement increment. ext Let F be the external load vector. int Given the internal force vector, the Newton-Raphson iterative algorithm is used to solve the above nonlinear equations. After the iteration converges, the nodal displacement u of the current iteration step is obtained. Solving for the element strain tensor from nodal displacements: In the formula For displacement gradient, The element strain tensor is input into the Mohr-Coulomb elastoplastic constitutive model to calculate the effective stress of the element. Substituting into the Biot effective stress relationship, we obtain the total stress σ.
6. The method for simulating critical conditions for gravity erosion of loess slopes with dynamic moisture content correction according to claim 5, characterized in that, In step S4, the permeability parameters are corrected based on the changes in pore structure and fed back to the seepage field: For the i-th element, the deformed Jacobian determinant J is solved. i Characterizes the degree of deformation of the unit volume. Update porosity Update dry density In the formula Let n be the displacement gradient of the i-th element. i The porosity after deformation. The initial porosity, The dry density after deformation. The initial dry density; Using a Kozeny-Carman type relationship based on porosity variation, the expression is: In the formula, k i Let k be the updated permeability coefficient of the i-th unit. i,0 Let be the initial permeability coefficient of the i-th unit.
7. The method for simulating critical conditions for gravity erosion of loess slopes with dynamic moisture content correction according to claim 6, characterized in that, The preset convergence condition in step S4 is expressed as follows: ; ; ; In the formula, , , These represent pore water pressure, permeability coefficient, and relative displacement error, respectively, with pref, kref, and uref serving as reference values. Preset convergence tolerance; when When the current computation time step reaches convergence, the bidirectional fluid-structure interaction computation reaches convergence and proceeds to the next computation time step.
8. The method for simulating critical conditions for gravity erosion of loess slopes with dynamic moisture content correction according to claim 1, characterized in that, The strength reduction calculation in step S5: Let the strength reduction factor be F, then the reduced cohesion of the i-th soil element is: ; The reduced internal friction angle is: ; In the formula, c i (t) Cohesion of the i-th soil element at time t after dynamic correction for water content. Let be the internal friction angle of the i-th soil element at time t after dynamic correction based on moisture content; The reduction factor is gradually increased according to the set step size to reduce the cohesion and internal friction angle respectively. When the reduction factor increases to the point that the slope changes from stable to overall unstable, the critical reduction factor is the safety factor corresponding to the current rainfall time.
9. The method for simulating critical conditions for gravity erosion of loess slopes with dynamic moisture content correction according to claim 1, characterized in that, The abrupt change in slope displacement includes the increase in horizontal displacement at the top of the slope exceeding a preset displacement increment threshold between adjacent calculation steps; the connection of the plastic zone includes the connection between the plastic zone at the toe of the slope and the plastic zone in the upper middle part of the slope to form a continuous potential sliding zone; and the non-convergence of numerical iteration includes the failure of the residual to reach a preset convergence tolerance within a preset number of iterations.
10. The method for simulating critical conditions for gravity erosion of loess slopes with dynamic moisture content correction according to claim 1, characterized in that, In step S2, the loess sample is set with a moisture content of 8%-22% and a dry density of 1.40-1.70 g / cm³.
