Method for assessing the danger of high-speed long-range landslides based on friction weakening and numerical simulation

Abstract

The application is a high-speed long-range landslide risk assessment method based on friction weakening and numerical simulation, which establishes a quantitative relationship formula of friction coefficient mu and shear rate V, normal stress sigma and shear displacement d; the PFC software is used to simulate the movement process of the high-speed long-range landslide of the target landslide model, the sliding speed, normal stress and sliding distance of the particles at each time point can be read through the command stream in the movement process, and then the friction coefficient formula of the target landslide is used to calculate the friction coefficient of the landslide at each time point; the sliding distance, sliding speed and accumulation layer thickness are obtained under the premise of real-time updating of the friction coefficient; the risk of high-speed long-range landslide is evaluated according to the three indexes of sliding distance, sliding speed and accumulation layer thickness. The method solves the disadvantage that the change of friction coefficient cannot be considered in the current single landslide risk assessment.

Description

Technical Field

[0001] This invention belongs to the field of geological disaster prevention and control research, specifically involving a method for assessing the risk of high-speed long-distance landslides based on friction weakening formula and PFC numerical simulation. Background Technology

[0002] Global warming has led to the large-scale melting of snow-capped mountains, threatening not only the loss of natural landscapes but also the lives and property of people in high-altitude regions. High-speed, long-distance landslides caused by melting snow and ice are a particularly dangerous natural disaster, characterized by high energy, rapid speed, and long sliding distances. These characteristics make the analysis of high-speed, long-distance landslide processes extremely complex. The weakening of friction between the landslide mass and the underlying bedrock during the sliding process is a challenging problem in this field. A key issue is how to quantitatively consider this frictional weakening process in the hazard assessment of high-speed, long-distance landslides.

[0003] Current research on landslide hazard assessment mainly focuses on the regional scale. Guo Zizheng et al. (Guo Zizheng et al., Rapid Assessment Model and Application of Rainfall-Induced Shallow Landslide Hazard) developed a formula for calculating the probability of landslide failure considering different rainfall conditions, achieving landslide hazard assessment at the regional scale (area greater than 100 square kilometers). However, they could not assess the hazard of individual landslides. They used GIS software to calculate landslide hazard by combining landslide susceptibility with effective rainfall intensity, but this method is only applicable to slow landslides at the regional scale and not to high-speed, long-distance landslides. This is because slow landslides do not involve significant movement. While statistical models can be used in GIS software for hazard assessment, individual landslides are small, and high-speed, long-distance landslides involve movement, requiring consideration of kinematic laws. Existing numerical simulation methods struggle to quantitatively account for landslide weakening phenomena. In the numerical simulation of high-speed, long-distance landslides, frictional weakening cannot be ignored; otherwise, the final simulation results will differ significantly from reality.

[0004] This invention is based on a quantitative formula for calculating the weakening of the friction coefficient over time during numerical simulation. Then, the sliding distance, sliding velocity, and accumulation layer thickness of a high-speed long-distance landslide are simulated in PFC software using this formula, thereby realizing the risk assessment of a single landslide. Summary of the Invention

[0005] To address the aforementioned problems, this invention provides a high-speed, long-distance landslide hazard assessment method based on friction weakening and numerical simulation, which overcomes the shortcoming of current single landslide hazard assessments that cannot consider changes in the friction coefficient.

[0006] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: A method for assessing the hazard of high-speed, long-distance landslides based on friction weakening and numerical simulation, the method comprising the following steps: S1, coefficient of friction With shear rate V and normal stress The quantitative relationship between the shear displacement d and the shear displacement d is: (1) in, For is V and The steady-state value of the friction coefficient when it is infinite; The peak value of the friction coefficient; i, j, k, and r are constants; , These are constants obtained through fitting; S2. By conducting high-speed ring shear tests, the relationship between the friction coefficient and shear displacement of the target landslide is obtained, and the peak value of the friction coefficient of the target landslide is determined. S3. Conduct shear experiments on the target landslide soil under room temperature and ambient humidity conditions, including shear experiments with constant shear rate but different normal stress conditions, and shear experiments with constant normal stress but different shear rates, to obtain the corresponding steady-state values ​​of the friction coefficient. ; Then, based on the peak and steady-state values ​​of the friction coefficient of the target landslide Obtain the sliding weakening distance Dc under different shear velocities and normal stresses; The values ​​of i, j, k, r, and in formula (1) are determined by data fitting. , , The values ​​are used to obtain a friction coefficient formula that fits the target landslide; S4. Input the friction coefficient formula of the target landslide determined in step S3 into the PFC software for command flow programming, which can calculate the friction coefficient at each moment. S5. Set the macro and micro parameters of the target landslide, establish the model of the target landslide, and use PFC software to simulate the high-speed long-distance landslide motion process of the target landslide model. During the motion, the sliding rate, normal stress and sliding distance of the particles at each time point can be read through the command flow. Then, the friction coefficient of the landslide at each moment is calculated using the friction coefficient formula of the target landslide. Under the premise of real-time updating of the friction coefficient, three indicators are obtained: sliding distance, sliding speed, and accumulation layer thickness; the risk of high-speed long-distance landslides is assessed based on these three indicators.

[0007] Compared with the prior art, the beneficial effects of the present invention are: This invention is primarily applied to the risk assessment of high-speed, long-distance landslides that consider frictional weakening phenomena, thereby enabling better prediction and early warning of geological disasters. This invention constructs a quantitative calculation formula for landslide frictional weakening that simultaneously considers three parameters: the actual sliding distance, sliding rate, and normal stress. Using this formula and PFC software, the sliding distance, sliding velocity, and deposit thickness of a high-speed, long-distance landslide were simulated, thus achieving landslide risk assessment. This invention offers the following advantages: 1. A quantitative calculation formula for friction weakening during landslide movement is proposed for the first time. This formula simultaneously considers the influence of three parameters—the actual sliding distance, sliding rate, and normal stress—on the friction coefficient, a feature not seen in previous studies. This quantitative calculation formula can more accurately describe the movement process of high-speed, long-distance landslides and the variation law of the friction coefficient during this process. 2. By combining the above formula with PFC software, it is possible to simulate the three indicators of high-speed long-distance landslides: sliding distance, sliding velocity, and deposit thickness, thereby achieving hazard assessment of individual landslides. Previous studies have mostly focused on landslide hazard assessment at the regional scale based on GIS platforms. This patent realizes the hazard assessment of individual landslides considering the weakening phenomenon of friction coefficient, which is a very useful supplement to this field.

[0008] 3. This patent provides a quantitative calculation formula for the friction coefficient and these three parameters, and proposes an assessment method for the risk of high-speed long-distance landslides by considering the phenomenon of friction weakening. This method can not only more accurately describe the change law of the friction coefficient during the process of high-speed long-distance landslides, but also increase the accuracy and reliability of the risk assessment of high-speed long-distance landslides at the single-unit scale. Attached Figure Description

[0009] Figure 1 This is a schematic diagram showing the changes in shear displacement d and friction coefficient μ.

[0010] Figure 2 The figure shows the fitting relationship between the steady-state values ​​of shear rate, normal stress, and friction coefficient in the embodiment.

[0011] Figure 3 The shear rate, normal stress, and sliding weakening distance in the examples are ( ). The fitting relationship diagram.

[0012] Figure 4 The variation law of friction coefficient during the sliding process of Yigong landslide.

[0013] Figure 5 This is a comparison chart of the simulation results of the sliding velocity of the Yigong landslide.

[0014] Figure 6 This is a comparison chart of the simulation results of the sliding distance and the thickness of the deposit layer in the Yigong landslide. Detailed Implementation

[0015] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. Of course, the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0016] This invention relates to a method for assessing the hazard of high-speed, long-distance landslides based on friction weakening and numerical simulation, comprising the following steps: S1, coefficient of friction With shear rate V and normal stress A quantitative formula relating the shear displacement d to the shear displacement d. S2. Plot the relationship between the friction coefficient and displacement of the target landslide through high-speed ring shear test, and determine the peak value of the friction coefficient of the target landslide; S3. Conduct conventional shear tests on the target landslide soil under constant shear rate and different normal stress conditions at room temperature and ambient humidity, as well as conventional shear tests under constant normal stress and different shear rates. Based on the experimental results, fit the measured shear rate, normal stress, and shear displacement values ​​to determine the steady-state value of the friction coefficient of the target landslide. Then, based on the peak friction coefficient and steady-state friction coefficient of the target landslide... Dc was obtained under different shear rates and normal stresses; The values ​​of i, j, k, r, and in formula (1) are determined by data fitting. , , The values ​​are used to obtain a friction coefficient formula that fits the target landslide; S4. Input the friction coefficient formula of the target landslide obtained above into the PFC software for command flow programming. Read the sliding rate, normal stress and sliding distance of the particles at each time point through the command flow, and then use the friction coefficient formula of the target landslide to calculate the landslide friction coefficient at each time point. S5. Using the landslide friction coefficient and other macroscopic and microscopic parameters calculated above, the movement process of a high-speed, long-distance landslide is simulated using PFC software to obtain three indicators: sliding distance, sliding velocity, and deposit thickness. The sliding distance can be used to assess the impact range of the landslide hazard; comparing the landslide sliding velocity with the escape speed of a person can assess the magnitude of the danger posed by the landslide to people within the landslide's impact range; comparing the deposit thickness with the height of a person and the height of a building can assess the magnitude of the danger posed by the landslide to people and buildings at specific heights within the landslide's impact range.

[0017] In this invention, shear displacement and shear rate refer to physical quantities during indoor tests, while sliding distance and sliding rate are the actual values ​​when a landslide occurs, both represented by d and V.

[0018] The coefficient of friction is affected by the actual sliding distance, sliding speed, and normal stress of the landslide. The effect of this. Experimental results show that after a small shear displacement, the friction coefficient reaches its peak ( As the shear displacement increases, the friction coefficient gradually decreases until it reaches a steady-state value. The peak value of the friction coefficient. and the steady-state value of friction coefficient The shear displacement d corresponding to 95% of the difference is defined as ,See Figure 1 , where the horizontal axis represents the shear displacement and the vertical axis represents the friction coefficient. The experimentally obtained relationship between the shear rate, shear displacement, normal stress, and friction coefficient is expressed as equation (1): (1) In the formula, d is the experimentally obtained shear displacement, and V is the experimentally obtained shear rate. It is the normal stress obtained from the experiment. It is V and The steady-state value of the friction coefficient at infinity, and the peak value of the friction coefficient. Compared to It is a constant. , The constants are obtained by fitting, while i, j, k, and r are material constants.

[0019] (3), It's about the shear rate ( ) and normal stress ( The function is expressed as follows: (6) In the formula It can be obtained from the shear displacement-friction coefficient diagram of landslide samples in indoor tests, and Dc is the shear displacement corresponding to 95% of the difference between the aforementioned peak friction coefficient and the steady-state friction coefficient.

[0020] Formula (1) and Substituting part of the formula yields formula (2), which in turn gives... and and The relation, in the formula It's about the shear rate ( ) and normal stress ( ) function (2) Will Split into and , and The formula expands to: (4) (5) In the formula and For the shear rate ( ) and normal stress ( These are the steady-state values ​​of the friction coefficient when it is infinite. , It is a constant obtained by fitting.

[0021] Example: (1) Overview of the experimental case The Yigong landslide was selected as a case study. On April 9, 2000, a massive, high-speed, long-distance landslide occurred in Zhamunonggou, Yigong Township, Bomi County. The landslide center was located at latitude N30°12′3″ and longitude E94°58′03″. Rocks cascaded down a 5km high mountain, with a sliding distance of 8km, a maximum elevation difference of approximately 3.3km, and a maximum horizontal displacement of approximately 8.0-10.0km. The maximum width of the landslide's leading edge was 3.0km, its longitudinal length was 4.6km, and its thickest point reached 80m. The total volume of the landslide was... This massive mass of rock and soil slid downwards at a maximum speed of 37-39 m / s, eventually accumulating to form a natural dam with a length and width of 2.5 km and an average thickness of 60 m.

[0022] (2) Determination of the peak value of the friction coefficient High-speed ring shear tests were conducted on soil from the Yigong landslide in the laboratory. The normal stress range during the experiment was determined to be 0.29, 0.64, 0.99, 1.47, and 1.85 MPa, based on the actual accumulation thickness of the Yigong landslide. The shear rate range was 0.07 m / s to 1.31 m / s. The maximum sliding displacement of the experimental sample was set at 12 m. Based on the experimental results, the relationship between the friction coefficient and displacement of the Yigong landslide was obtained, and the peak friction coefficient of the Yigong landslide was determined to be 0.64.

[0023] (3) Determine the steady-state value of the friction coefficient and the formula for the friction coefficient. Conventional shear tests were conducted on the soil from the Yigong landslide under ambient temperature and humidity conditions. The experiment consisted of two groups: one group measured the friction coefficient under constant normal stress (1.47 MPa in this example) but different shear rates (0.26 m / s to 1.31 m / s in this example); the other group measured the friction coefficient under constant shear rate (0.87 m / s in this example) but different normal stresses (0.29 MPa to 1.85 MPa in this example). In principle, the more experiments conducted, the more accurate the results. The results of the two groups in this example are summarized in Table 1.

[0024] Table 1 Shear test data of Yigong landslide ; Based on the above experimental data, As a fitting function, it can fit the functional relationship between shear rate, normal stress and steady-state value of friction coefficient, as shown in formula (7). The fitting effect diagram is shown in [Figure number missing]. Figure 2 : (7) The experimental results show that It is 1 / 6.4; It is 1 / 1.6; The coefficient of friction is 0.64; steady-state value. It is 0.10.

[0025] Since Dc is the peak value and steady-state value of the friction coefficient The shear displacement corresponding to 95% of the difference is given in this embodiment, where the peak friction coefficient is 0.64. This is based on the steady-state friction coefficient values ​​in Table 1. To obtain Dc under different shear rates and normal stresses, according to By performing fitting, the shear rate, normal stress, and The functional relationship between the three factors is shown in formula (8), and the fitting result is as follows: Figure 3 As shown: (8) At this point, i, j, k, and r are 0.52, -0.84, 2.5, and -1.1, respectively.

[0026] In formula (7) Substituting the expression for Dc in formula (8) into formula (2), we can obtain an empirical equation suitable for describing the steady-state frictional characteristics of the matrix phase of the Yigong landslide, as follows: (9) (4) Fine parameter calibration of PFC software This case study utilizes two-dimensional particle flow discrete element (PFC) software to simulate the movement process of the Yigong landslide, incorporating the friction coefficient as a variable into the Yigong landslide model. It serves as a case study demonstrating a method for assessing the risk of high-speed, long-distance landslides based on friction weakening and numerical simulation.

[0027] The Yigong landslide model in PFC software adopts a parallel bond model. The contact parameters that need to be calibrated include the effective modulus of linear contact (emod), the ratio of linear normal to tangential stiffness (kratio), the effective modulus of parallel bond (pb_emod), the ratio of parallel bond normal to tangential stiffness (pb_kratio), the parallel bond cohesion (pb_coh), the parallel bond friction angle (pb_fa), the parallel bond tensile strength (pb_ten), particle density, and particle radius. First, the linear contact modulus and parallel bond modulus are calibrated using the elastic modulus (E). Then, the stiffness ratio of the linear contact component and the parallel bond component is calibrated using Poisson's ratio. Finally, the uniaxial tensile strength is calibrated... The normal and tangential bond strengths were calibrated. The calibration results are shown in the table below: Table 2 Microscopic parameter calibration ; (4) Simulate the Yigong landslide using PFC software. The Yigong landslide particle flow model was established using a combination of PFC and AutoCAD software to create the landslide surface (wall) and landslide body (ball). The model is 10km long and 33km high, with a total of 3278 particles and a particle radius of 8~10m. The specific method of establishment is as follows: a 1:1 main cross-section of the Yigong landslide was drawn using CAD and saved as a dxf file. The model was established in the PDF using geometry import and wall import filename, and the landslide surface and landslide body regions were established respectively. The microscopic parameters in Table 2 were used to assign particles generated at specified positions (landslide body regions). Then, particle stress balance was performed to make the initial velocity of the particles 0. Formula (9) was used as the particle friction coefficient calculation formula in the command flow. At each time point, the sliding distance, sliding rate and normal stress of the simulated landslide particles at that time were read to calculate the corresponding friction coefficient at that time. Then, the landslide movement process was simulated using PFC software. Subsequently, the sliding distance, sliding rate and normal stress at the next time point were updated, and the friction coefficient and movement process were recalculated. This process was iterated until the landslide movement stopped. The final friction coefficient variation pattern during the Yigong landslide movement is as follows: Figure 4 As shown, it can be seen that there is a significant friction weakening phenomenon during the motion process. The coefficient of friction decreases from 0.67 at the beginning to 0.11 at about 17s, and then remains stable at around this value.

[0028] (5) Risk assessment of Yigong landslide The simulation was performed in PFC software from the starting point of the landslide until the final stop of the landslide, with a total of 97,110 time steps. The sliding distance, sliding velocity and accumulation layer thickness of the Yigong landslide were obtained to assess its hazard. Figure 5 The simulation results of the sliding velocity of the Yigong landslide are shown before (a) and after (b) considering friction weakening. The results indicate that when the friction coefficient formula for the target landslide in this embodiment is used, the landslide velocity rapidly increases to approximately 39 m / s in the short period after the start of sliding, then decreases to around 5 m / s and fluctuates, finally stabilizing at around 0.2 m / s until it stops. In the figure, the landslide reaches its maximum speed of 39.2 m / s at approximately 90 s, and the velocity tends to stabilize after 760 s. However, if a constant value is used instead of considering the change in the friction coefficient, the maximum landslide velocity is only about 30 m / s, which differs significantly from the actual situation (as stated in the experimental case overview, the maximum actual velocity of the Yigong landslide is 37-39 m / s). The normal speed of an adult running away is approximately 2-3 m / s. Therefore, based on the simulation results, the Yigong landslide is most dangerous during the sliding periods of 0-700 s and 1000-1600 s, while its velocity is less than the human's running speed during the remaining time, thus reducing the danger. However, the risk assessment that does not incorporate the friction weakening formula is considered safe at around 1000s, which differs significantly from the actual situation.

[0029] Figure 6 Simulation results of the sliding distance and final deposit thickness of the Yigong landslide before (a) and after (b) considering friction weakening are shown. The results show that when the friction coefficient formula for the target landslide proposed in this patent is used, the maximum sliding distance reaches approximately 7 km, and the maximum horizontal sliding distance is approximately 8.0 km, which is basically consistent with the actual landslide situation. The sliding distance refers to the distance traveled during the sliding process, which is a curved segment, while the horizontal sliding distance is a straight line distance. However, when the friction coefficient formula for the target landslide in this embodiment is not used, the maximum sliding distance is approximately 6.5 km, and the maximum horizontal sliding distance is only about 7.5 km, which differs significantly from the actual horizontal distance. Therefore, the friction coefficient formula proposed in this patent can be well used for simulating high-speed, long-distance landslides. From the perspective of deposit thickness, the deposit thickness is greatest at the rear of the landslide, reaching nearly 100 m, while the deposit thickness in the middle and front parts is relatively consistent, approximately 50 m. Therefore, the rear of the landslide is more dangerous. If we assume a building height of about 3 meters, buildings above the 33rd floor in the rear of the landslide will not be submerged by the landslide accumulation zone. In the middle and front parts, the landslide poses less of a danger to buildings above the 17th floor.

[0030] Any aspects not covered in this invention are applicable to existing technologies.

Claims

1. A method for assessing the hazard of high-speed, long-distance landslides based on friction weakening and numerical simulation, characterized in that, The evaluation method includes the following steps: S1, coefficient of friction With shear rate V and normal stress The quantitative relationship between the shear displacement d and the shear displacement d is: (1) in, For is V and The steady-state value of the friction coefficient when it is infinite; The peak value of the friction coefficient; i, j, k, and r are constants; , These are constants obtained through fitting; S2. By conducting high-speed ring shear tests, the relationship between the friction coefficient and shear displacement of the target landslide is obtained, and the peak value of the friction coefficient of the target landslide is determined. S3. Conduct shear experiments on the target landslide soil under room temperature and ambient humidity conditions, including shear experiments with constant shear rate but different normal stress conditions, and shear experiments with constant normal stress but different shear rates, to obtain the corresponding steady-state values ​​of the friction coefficient. ; Then, based on the peak and steady-state values ​​of the friction coefficient of the target landslide Obtain the sliding weakening distance Dc under different shear velocities and normal stresses; The values ​​of i, j, k, r, and in formula (1) are determined by data fitting. , , The values ​​are used to obtain a friction coefficient formula that fits the target landslide; S4. Input the friction coefficient formula of the target landslide determined in step S3 into the PFC software for command flow programming, which can calculate the friction coefficient at each moment. S5. Set the macro and micro parameters of the target landslide, establish the model of the target landslide, and use PFC software to simulate the high-speed long-distance landslide motion process of the target landslide model. During the motion, the sliding rate, normal stress and sliding distance of the particles at each time point can be read through the command flow. Then, the friction coefficient of the landslide at each moment is calculated using the friction coefficient formula of the target landslide. Under the premise of real-time updating of the friction coefficient, three indicators are obtained: sliding distance, sliding speed, and accumulation layer thickness; the risk of high-speed long-distance landslides is assessed based on these three indicators.

2. The method for assessing the risk of high-speed, long-distance landslides based on friction weakening and numerical simulation as described in claim 1, characterized in that, Sliding distance can be used to assess the impact range of landslide hazards; comparing sliding speed with the escape speed of a person can assess the magnitude of the danger posed by the landslide to people within the landslide impact range; comparing the thickness of the deposit layer with the height of a person and the height of a building can assess the magnitude of the danger posed by the landslide to people and buildings within the landslide impact range.

3. The method for assessing the risk of high-speed, long-distance landslides based on friction weakening and numerical simulation as described in claim 1, characterized in that, The macroscopic and mesoscopic parameters include: linear contact modulus, parallel bond modulus, stiffness ratio of linear contact components, stiffness ratio of parallel bond components, normal bond strength, tangential bond strength, elastic modulus, Poisson's ratio, cohesion, uniaxial compressive strength, internal friction angle, and density.

4. The method for assessing the risk of high-speed long-distance landslides based on friction weakening and numerical simulation according to claim 1, characterized in that, The model construction process for the target landslide is as follows: The landslide surface and landslide body were established by combining PFC software and AutoCAD software. The 1:1 main profile of the target landslide was drawn in CAD and saved as dxf format. The model was built in PDF using geometry import and wall import filename, and the landslide surface and landslide body regions were established respectively. Particles generated at specified positions were assigned based on the micro parameters of the target landslide. Then, particle stress balance was performed to make the initial velocity of the particles 0. The friction coefficient formula of the target landslide determined in step S3 is used as the formula for calculating the particle friction coefficient in the command stream. At each time point, the friction coefficient at that time is calculated by reading the sliding distance, sliding rate and normal stress of the simulated landslide particles. Then, the landslide movement process is simulated using PFC numerical simulation software. Subsequently, the sliding distance, sliding rate and normal stress of the simulated landslide at the next time point are updated, and the friction coefficient and movement process are recalculated. This process is iterated until the landslide movement stops.

Images