A high-position slope chain risk assessment method based on numerical simulation

Abstract

The application discloses a high-position slope chain risk assessment method based on numerical simulation and belongs to the technical field of landslide numerical simulation, and comprises the following steps: S1, collecting landslide susceptibility results and landslide catalog data; S2, screening a landslide starting area; S3, numerical simulation of the landslide; and S4, spatial analysis and risk assessment.The high-position slope chain risk assessment method based on numerical simulation has the advantages that the landslide starting area is accurately screened through numerical simulation, and scientific basis is provided for the prevention and disaster reduction of landslide disasters.

Description

Technical Field

[0001] This invention relates to the field of landslide numerical simulation technology, and in particular to a method for assessing the risk of chain reaction of high-altitude slopes based on numerical simulation. Background Technology

[0002] Landslides are common and highly destructive geological hazards worldwide, and their occurrence is usually closely related to the stability of high-altitude slopes. Traditional landslide risk assessment methods rely heavily on experience and field investigations, often suffering from insufficient accuracy and inefficiency when dealing with complex terrain and large areas. With the development of computer technology and numerical simulation methods, landslide risk assessment based on numerical simulation has gradually become a research hotspot. Among them, the FLOW-R model, as an advanced landslide numerical simulation tool, can simulate the initiation, propagation, and deposition processes of landslides, providing an effective technical means for landslide risk assessment. However, directly using landslide susceptibility assessment results as the initiation zone input for the FLOW-R model may lead to significant deviations between the simulation results and actual conditions, as susceptibility assessment results are usually based on statistical analysis. Therefore, a method is urgently needed to accurately screen landslide initiation zones in the FLOW-R model to improve the accuracy and reliability of landslide risk assessment. Summary of the Invention

[0003] The purpose of this invention is to provide a method for assessing the risk of landslide chain reaction on high-altitude slopes based on numerical simulation. This method accurately screens landslide initiation zones through numerical simulation, providing a scientific basis for the prevention and mitigation of landslide disasters.

[0004] To achieve the above objectives, this invention provides a method for assessing the chain reaction risk of high-altitude slopes based on numerical simulation, comprising the following steps:

[0005] S1. Landslide susceptibility results and landslide cataloging data collection;

[0006] S2, Landslide initiation zone screening;

[0007] S3. Numerical simulation of landslides;

[0008] S4. Spatial Analysis and Risk Assessment.

[0009] Preferably, in S1, the scope of the study area is determined, and geological, topographical, and climatic data related to landslide susceptibility are collected, including elevation, slope, soil thickness distribution, soil type, rainfall, and seismic activity. The probability of landslides occurring in the area is assessed using an infinite slope stability model, and landslide cataloging data for the study area is collected, including detailed records of historical landslide events, including the location, scale, occurrence time, and triggering factors of the landslides.

[0010] Preferably, S2 specifically includes the following steps:

[0011] S2.1 Integrate the landslide susceptibility results and landslide cataloging data generated in S1;

[0012] S2.2. Based on the data integrated in S2.1, the kernel density estimation method is used to perform statistical analysis on the spatial distribution of historical landslide events.

[0013] Preferably, S2.2 specifically includes the following steps:

[0014] S2.2.1 Using the geographic coordinates in the landslide catalog data as the input point set, the density is estimated using the Gaussian kernel function. Based on the landslide distribution density and spatial dimension of the study area, the kernel density map is output.

[0015] S2.2.2 The Segment Anything (SAM) model is used to analyze the kernel density map to obtain potential landslide initiation data.

[0016] Preferably, in S3, the FLOW-R model is used to simulate the movement process of landslide material. The FLOW-R model is based on the Digital Elevation Model (DEM) and its derivatives for regional sensitivity analysis. It calculates the propagation path and deposition range of landslide material through numerical iteration. Its core formula is:

[0017]

[0018] Where F is the net force of the landslide material; m is the mass of the landslide material; g is the gravitational acceleration; θ is the terrain slope; μ is the internal friction coefficient, which controls the frictional resistance of the landslide along the slope; ρ is the density of the landslide material; v is the current velocity of the landslide material; and ξ is the turbulence coefficient.

[0019] Preferably, S4 specifically includes the following steps:

[0020] S4.1 Based on the landslide impact range generated by numerical simulation using FLOW-R, the landslide impact range and disaster-bearing body data are comprehensively evaluated through spatial overlay analysis;

[0021] S4.2 Construct a risk assessment indicator system;

[0022] S4.3. Develop scientific risk management strategies based on risk assessment results.

[0023] Preferably, the risk assessment in S4.2 is based on three core elements: hazard, vulnerability, and exposure, and specifically includes the following steps:

[0024] S4.2.1 Based on the landslide simulation results, extract the impact range of the landslide and classify the hazard level;

[0025] S4.2.2 Assess the vulnerability of the disaster-bearing body by combining its type and socioeconomic data;

[0026] S4.2.3 Quantify the population and infrastructure of the affected area.

[0027] Therefore, this invention adopts the above-mentioned method for assessing the risk of landslide chain formation on high-level slopes based on numerical simulation. Through numerical simulation, it accurately screens landslide initiation zones, providing a scientific basis for the prevention and mitigation of landslide disasters.

[0028] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0029] Figure 1 This is a flowchart of a method for assessing the risk of chain formation in high-altitude slopes based on numerical simulation, according to the present invention.

[0030] Figure 2 This is a schematic diagram of the location and DEM of the research area of ​​the numerical simulation-based method for assessing the risk of chain formation in high-altitude slopes according to the present invention.

[0031] Figure 3 This is an affected distribution map of a numerical simulation-based method for assessing the chain-like risk of high-altitude slopes, as described in this invention. Figure 3 Map (a) shows the extent of potential landslide impact along a certain highway. Figure 3 (b) in the figure is a map showing the extent of potential landslide impact on a river along a certain highway. Figure 3 (c) in the figure shows the extent of the landslide impact on buildings along a certain highway. Figure 3 (d) in the figure is an analysis map of the distribution range of buildings affected in a certain area along a highway. Figure 3 (e) in the figure is a map showing the potential impact range of a landslide along a certain highway;

[0032] Figure 4 This invention relates to a method for assessing the risk of chain-like formation of high-altitude slopes based on numerical simulation, and presents kernel density results for a portion of a highway along a certain area. Detailed Implementation

[0033] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0034] Unless otherwise defined, the technical or scientific terms used in this invention shall have the ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains.

[0035] Example 1

[0036] like Figure 1 As shown, this invention provides a method for assessing the risk of chain reaction on high-altitude slopes based on numerical simulation. Taking a 20km buffer zone along a highway as an example, the method includes the following steps:

[0037] S1. Landslide susceptibility results and landslide cataloging data collection;

[0038] The study area was defined, and geological, topographical, and climatic data related to landslide susceptibility were collected, including elevation, slope, soil thickness distribution, soil type, rainfall, and seismic activity. An infinite slope stability model was used to assess the likelihood of landslides in the area. Landslide cataloging data for the study area was collected, including detailed records of historical landslide events, such as location, size, timing, and triggering factors.

[0039] S2. Landslide initiation zone screening; specifically including the following steps:

[0040] S2.1 Integrate the landslide susceptibility results and landslide cataloging data generated in S1;

[0041] The landslide susceptibility results generated in the first step are integrated with landslide catalog data to achieve effective analysis of multi-source data. The landslide susceptibility results provide the probability distribution of landslide occurrences within the study area, while the landslide catalog data records the specific locations, scales, and triggering factors of historical landslides. By spatially overlaying and analyzing these two types of data in a Geographic Information System (GIS), high-susceptibility areas and historically frequent landslide-affected areas are preliminarily identified, laying the foundation for subsequent refined screening. This process ensures spatial consistency of the data and provides a reliable kernel input for density analysis.

[0042] S2.2. Based on the data integrated in S2.1, the kernel density estimation method is used to perform statistical analysis on the spatial distribution of historical landslide events; specifically, the following steps are included:

[0043] S2.2.1. Using geographical coordinates from the landslide catalog data as the input point set, each point represents a historical landslide event; a Gaussian kernel function is used for density estimation, and the bandwidth selection is optimized using cross-validation, with the range set to 100 meters. Based on the landslide distribution density and spatial dimensions of the study area, the kernel density result map is output, as shown below. Figure 4 As shown, by setting an appropriate bandwidth, areas with density values ​​below a preset value are considered insignificant areas, thus eliminating landslide points that are scattered, small in scale, and lack the characteristics of initiation zones. The output resolution of the kernel density map is consistent with the input DEM (e.g., 10 meters or 30 meters), and the pixel values ​​reflect the landslide density per unit area. High-value areas potentially correspond to landslide initiation zones.

[0044] S2.2.2 The Segment Anything (SAM) model is used to analyze the kernel density map to obtain potential landslide initiation. SAM is a deep learning-based segmentation model. Before applying SAM, the kernel density map needs to be constructed: the kernel density map is converted into a grayscale image, and the pixel values ​​are normalized to 0-255. High-density areas correspond to high pixel values, and low-density areas correspond to low pixel values, providing input for SAM.

[0045] SAM's segmentation process relies on point cues. Kernel density is used as the central cue point for high-density clusters, and a high-precision detection algorithm automatically identifies kernel density values ​​as local maxima. To ensure comprehensive coverage, additional cue points can be added for regions with a susceptibility score of 0.8. SAM segments the kernel density map based on these cues, generating multiple segmentation gap codes. Its multi-readout feature extraction capability can capture irregular boundaries in high-density regions, subsequently improving the intelligence and reliability of traditional sparse value segmentation methods.

[0046] The segmentation reservation codes generated by SAM require post-processing optimization. First, areas smaller than 100 square meters are clearly removed to reduce noise. Second, screening criteria include: areas with an average kernel density value lower than 1 event per event are excluded; areas with less than 50% overlap with areas having a susceptibility score of 0.8 are considered low-risk; areas with a slope less than 5° are excluded, as sparse areas do not meet the criteria for landslide initiation. The selected initiation areas retain high-risk and high-clustering characteristics. After screening, the initiation areas are classified and prioritized based on the average kernel density value and potential impact factors (such as area and distance to nearby affected bodies), with areas having high kernel density values ​​prioritized for subsequent analysis.

[0047] S3. Numerical simulation of landslides;

[0048] The FLOW-R model, based on the depth-integrated continuity equation and the Voellmy rheological model, simulates the movement of landslide material. This model simplifies the three-dimensional hydrodynamic equations, treating landslide material as a continuous medium, with its propagation dynamics along the surface controlled by frictional resistance and turbulence. The Voellmy rheological model combines the characteristics of solid friction and fluid turbulence, making it suitable for simulating the trajectories of different landslide types (such as debris flows and rockfalls). The FLOW-R model uses a digital elevation model (DEM) and its derivatives for regional sensitivity analysis, calculating the propagation path and deposition range of landslide material through numerical iteration. Its core formula is:

[0049]

[0050] Where F is the net force of the landslide material; m is the mass of the landslide material; g is the gravitational acceleration; θ is the terrain slope; μ is the internal friction coefficient, which controls the frictional resistance of the landslide along the slope; ρ is the density of the landslide material; v is the current velocity of the landslide material; and ξ is the turbulence coefficient.

[0051] The FLOW-R model uses a high-resolution DEM (e.g., 10-meter or 30-meter) as the terrain basis to ensure accurate representation of terrain features. The selected landslide initiation zone is used as the initial sliding source input to the model; the boundary and volume information of the initiation zone are directly imported from the results of the second step. To configure key model parameters, referring to the soil and rock parameters for different types of landslides provided on the FLOW-R website, preliminary software parameters are set, including the turbulence coefficient ξ and initial velocity conditions.

[0052] To ensure the accuracy of the simulation results, typical historical landslide events within the study area were selected as calibration cases. Landslide events occurring in Attabad were used, and their actual impact range was extracted as reference data. Through preliminary simulations, the differences between the landslide paths and depositional extents output by the model and historical records were compared. Key parameters were iteratively adjusted, and the calibration process employed a trial-and-error method combined with sensitivity analysis to determine the optimal combination of parameters.

[0053] After calibration, the FLOW-R model is run to simulate the landslide impact range through numerical iterative calculations. The simulation results are saved in Shp format, generating a landslide impact range map, as shown below. Figure 2 As shown, the data includes the landslide path boundaries and the extent of the depositional zone. To facilitate subsequent analysis, GIS software such as ArcGIS or QGIS was used to visualize the results, and key areas (such as maximum outflow distance and depositional depth) were marked.

[0054] S4. Spatial Analysis and Risk Assessment; specifically including the following steps:

[0055] S4.1 Based on the landslide impact range generated by numerical simulation using Flow-R, spatial overlay analysis is used to comprehensively assess the landslide impact range and the data of disaster-bearing bodies. This includes spatially overlaying the landslide impact range map with data on disaster-bearing bodies such as residential areas, roads, and buildings to identify threatened areas and calculate the affected population, infrastructure length, and land use area, obtaining the spatial distribution results of the affected buildings, roads, and rivers in the study area, such as... Figure 3 As shown, where, Figure 3 Map (a) shows the extent of potential landslide impact along a certain highway. Figure 3 (b) in the figure is a map showing the extent of potential landslide impact on a river along a certain highway. Figure 3 (c) in the figure shows the extent of the landslide impact on buildings along a certain highway. Figure 3(d) in the figure is an analysis map of the distribution range of buildings affected in a certain area along a highway. Figure 3 (e) in the figure is a map showing the potential landslide impact range along a certain highway.

[0056] S4.2 Construct a risk assessment indicator system;

[0057] Risk assessment is based on three core elements: hazard, vulnerability, and exposure, and includes the following steps:

[0058] S4.2.1 Based on the landslide simulation results, extract the impact range of the landslide and classify the hazard level;

[0059] S4.2.2 Assess the vulnerability of the disaster-bearing body by combining its type and socioeconomic data;

[0060] S4.2.3 Quantify the exposure indicators of the affected area, such as population and infrastructure. Use the Analytic Hierarchy Process (AHP) to determine the weights of each indicator, generating areas with different risk levels, and highlighting high-risk areas and key disaster-bearing structures. Based on this, conduct risk visualization and mapping.

[0061] S4.3. Develop scientific risk management strategies based on risk assessment results.

[0062] Based on the risk level map, suggestions for restricting development in high-risk areas are proposed, land use planning is optimized, and key areas for landslide prevention and control are identified, with priority given to implementing engineering measures.

[0063] Therefore, this invention adopts the above-mentioned method for assessing the risk of landslide chain formation on high-level slopes based on numerical simulation. Through numerical simulation, it accurately screens landslide initiation zones, providing a scientific basis for the prevention and mitigation of landslide disasters.

[0064] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for evaluating the chain risk of high-position slope based on numerical simulation, characterized in that: Comprising the following steps: S1, landslide susceptibility result and landslide inventory data collection; S2, landslide initiation area screening; Specifically comprising the following steps: S2.1, integrating the landslide susceptibility results generated in S1 and the landslide inventory data; S2.2, based on the data integrated in S2.1, using kernel density estimation method to statistically analyze the spatial distribution of historical landslide events; S2.2.1, using the geographic coordinates in the landslide inventory data as input point set, using Gaussian kernel function for density estimation, according to the landslide distribution density and spatial dimension of the study area, outputting the kernel density map; S2.2.2, using Segment Anything model SAM to analyze the kernel density map to obtain potential landslide initiation; S3, landslide numerical simulation; Using FLOW-R model to simulate the movement process of landslide material, FLOW-R model based on digital elevation model DEM and its derivatives for regional sensitivity analysis, through numerical iteration method to calculate the propagation path and deposition range of landslide material, its core formula is: ； wherein, is the net force of the landslide mass; is the mass of the landslide mass; is the acceleration due to gravity; is the terrain slope; is the internal friction coefficient, controlling the frictional resistance of the landslide along the slope surface; is the density of the landslide mass; is the current velocity of the landslide mass; is the turbulence coefficient; S4, spatial analysis and risk assessment; Specifically comprising the following steps: S4.1, based on the landslide impact range generated by using FLOW-R for numerical simulation, through spatial overlay analysis, the landslide impact range is comprehensively evaluated with the hazard bearing body data; S4.2, constructing risk assessment index system; Risk assessment is based on three core elements: danger, vulnerability and exposure, specifically comprising the following steps: S4.2.1, based on the landslide simulation results, extracting the impact range, dividing the danger level; S4.2.2, combined with the type of hazard bearing body and social and economic data, assess its vulnerability; S4.2.3, quantifying the population, infrastructure in the affected area; S4.3, based on the risk assessment results, developing scientific risk management strategies.

2. The numerical simulation-based high-position slope chain risk assessment method according to claim 1, characterized in that: In S1, the scope of the study area is determined, and the data related to landslide susceptibility, including geology, topography, and climate, are collected, including elevation, slope, soil thickness distribution, soil type, rainfall, and seismic activity. The infinite slope stability model is used to evaluate the possibility of landslide occurrence in the region. The landslide inventory data of the study area is collected, which includes detailed records of historical landslide events, including the location, size, occurrence time, and triggering factors of the landslide.

Images