Monitoring methods for landslides

Abstract

This invention discloses a method for monitoring landslides, relating to the field of power technology. The method includes: identifying landslide hazard areas based on slope information, vegetation cover type, and precipitation distribution in the power grid area; predicting the landslide volume in the hazard area to obtain a predicted landslide volume value; determining the spatiotemporal variation rate of moisture and the range of moisture anomalies in the hazard area based on the moisture status, wherein the moisture status includes at least the soil moisture content; and determining the monitoring results based on the predicted landslide volume, the spatiotemporal variation rate of moisture, and the range of moisture anomalies, wherein the monitoring results characterize the degree of danger of the landslide hazard area. This invention solves the technical problem of low accuracy in landslide monitoring.

Description

Technical Field

[0001] This invention relates to the field of power technology, and more specifically, to a method for monitoring landslides. Background Technology

[0002] Power grid equipment, such as transmission lines, substations, and distribution facilities, located in mountainous areas, may be threatened by geological disasters such as landslides. Therefore, monitoring landslide-prone areas is crucial and indispensable for providing early warnings of high-risk areas and ensuring the safe operation of the power grid.

[0003] When assessing the impact of precipitation on landslides and identifying potential landslide zones, relevant technologies typically rely on ground measurements. However, ground measurements have limited efficiency, making it impossible to achieve large-scale, high-frequency monitoring. The limited data obtained results in low accuracy in monitoring landslides.

[0004] There is currently no effective solution to the above problems. Summary of the Invention

[0005] This invention provides a method for monitoring landslides, which at least solves the technical problem of low accuracy in monitoring landslides.

[0006] According to one aspect of the present invention, a method for monitoring landslides is provided, comprising: determining a landslide hazard area based on slope information, vegetation cover type, and precipitation distribution in the area where the power grid is located, wherein the landslide hazard area is an area in the power grid area with low mountain stability and a risk of landslides; predicting the landslide volume in the landslide hazard area to obtain a predicted landslide volume value; determining the spatiotemporal variation rate of moisture and the range of moisture anomalies in the landslide hazard area based on the moisture status of the landslide hazard area, wherein the moisture status includes at least the soil moisture content; and determining monitoring results based on the predicted landslide volume value, the spatiotemporal variation rate of moisture, and the range of moisture anomalies, wherein the monitoring results are used to characterize the degree of danger of the landslide hazard area.

[0007] Optionally, predicting the landslide volume in the landslide hazard area to obtain a predicted landslide volume value includes: acquiring the mountain structure characteristics of the landslide hazard area, wherein the mountain structure characteristics include at least one of the following: mountain elevation, mountain slope, mountain aspect, vegetation index, precipitation, and rock edge movement rate; inputting the mountain structure characteristics into the landslide volume prediction model, and using the landslide volume prediction model to predict the landslide volume in the landslide hazard area to obtain a predicted landslide volume value.

[0008] Optionally, based on the moisture status of the landslide hazard area, the spatiotemporal rate of moisture variation and the range of moisture anomalies in the landslide hazard area are determined, including: identifying at least one key analysis area within the landslide hazard area, wherein the at least one key analysis area is an area with anomalies in surface temperature within the landslide hazard area; acquiring surface temperature, atmospheric temperature, vegetation index, and precipitation of the at least one key analysis area at different times; inputting the surface temperature, atmospheric temperature, vegetation index, and precipitation of the at least one key analysis area at different times into a moisture prediction model, and using the moisture prediction model to predict the moisture status of the at least one key analysis area at different times, thereby obtaining the moisture status of the at least one key analysis area at different times; and determining the spatiotemporal rate of moisture variation and the range of moisture anomalies based on the moisture status of the at least one key analysis area at different times.

[0009] Optionally, based on the moisture status of at least one key analysis area at different times, the spatiotemporal rate of moisture change and the range of moisture anomalies are determined, including: determining the moisture change rate of at least one key analysis area based on the moisture status of at least one key analysis area at different times; determining the spatiotemporal rate of moisture change based on the moisture change rate of at least one key analysis area; comparing the moisture change rate of at least one key analysis area with a preset change rate to determine whether at least one key analysis area is in a moisture anomaly state, wherein the moisture anomaly state is used to characterize that the moisture change rate is greater than the preset change rate; and determining the range of moisture anomalies based on the range of key analysis areas in a moisture anomaly state.

[0010] Optionally, at least one key analysis area is identified within the landslide hazard area, including: acquiring satellite thermal infrared images of the landslide hazard area; processing the satellite thermal infrared images to obtain thermal infrared radiance; processing the thermal infrared radiance using a single scattering inversion algorithm to obtain surface temperature data of the landslide hazard area; and using a clustering algorithm to identify anomalous data in the surface temperature data, determining the areas corresponding to the anomalous data as key analysis areas.

[0011] Optionally, the monitoring results are determined based on the predicted landslide volume, the spatiotemporal rate of moisture change, and the range of moisture anomalies, including: determining the weights of the predicted landslide volume, the spatiotemporal rate of moisture change, and the range of moisture anomalies respectively; determining the hazard coefficient of the landslide hazard area based on the predicted landslide volume, the spatiotemporal rate of moisture change, the range of moisture anomalies, the weights of the predicted landslide volume, the weights of the spatiotemporal rate of moisture change, and the weights corresponding to the range of moisture anomalies; and determining the monitoring results based on the hazard coefficient of the landslide hazard area.

[0012] Optionally, determining landslide hazard areas based on mountain slope information, vegetation cover type, and precipitation distribution in the area where the power grid is located includes: determining at least one slope anomalous area based on mountain slope information in the area where the power grid is located; inputting the vegetation cover type and precipitation distribution of the at least one slope anomalous area into a mathematical model, using the mathematical model to predict the mountain stability index of the at least one slope anomalous area, and obtaining the mountain stability index of the at least one slope anomalous area; and determining landslide hazard areas within the at least one slope anomalous area based on the mountain stability index of the at least one slope anomalous area.

[0013] Optionally, determining at least one slope anomaly area based on the slope information of the power grid area includes: dividing the power grid area into multiple areas to be determined; obtaining the slope information of the multiple areas to be determined; determining whether the slope information of the multiple areas to be determined is within a preset slope range; and determining the areas to be determined whose slope information is within the preset slope range as slope anomaly areas.

[0014] Optionally, based on the mountain stability index of at least one slope anomaly area, a landslide hazard area is determined in at least one slope anomaly area, including: determining slope anomaly areas with mountain stability indices within a preset range as candidate landslide hazard areas; determining the rock edge movement rate of candidate landslide hazard areas based on the bare rock distribution of candidate landslide hazard areas; and determining candidate landslide hazard areas as landslide hazard areas if the rock edge movement rate of candidate landslide hazard areas is greater than a preset rate.

[0015] Optionally, the method further includes: determining landslide response strategies based on monitoring results; and carrying out emergency treatment on power grid equipment within the landslide hazard area according to the landslide response strategies.

[0016] According to another aspect of the present invention, a landslide monitoring device is also provided, comprising: a first determining module, configured to determine a landslide hazard area based on the slope information, vegetation cover type, and precipitation distribution of the area where the power grid is located, wherein the landslide hazard area is an area in the area where the power grid is located with low mountain stability and a risk of landslide; a predicting module, configured to predict the landslide volume of the landslide hazard area and obtain a predicted landslide volume value; a second determining module, configured to determine the spatiotemporal variation rate of moisture and the range of moisture anomalies in the landslide hazard area based on the moisture status of the landslide hazard area, wherein the moisture status includes at least the soil moisture content; and a third determining module, configured to determine the monitoring result based on the predicted landslide volume value, the spatiotemporal variation rate of moisture, and the range of moisture anomalies, wherein the monitoring result is used to characterize the degree of danger of the landslide hazard area.

[0017] According to another aspect of the present invention, an electronic device is also provided, comprising: a memory storing an executable program; and a processor for running the program, wherein the program executes the above-described landslide monitoring method during runtime.

[0018] According to another aspect of the present invention, a computer-readable storage medium is also provided, comprising: the computer-readable storage medium including a stored executable program, wherein, when the executable program is executed, it controls the device where the storage medium is located to perform the above-described landslide monitoring method.

[0019] According to another aspect of the present invention, a computer program product is also provided, comprising: a computer program that, when executed by a processor, implements the above-described method for monitoring landslides.

[0020] In this embodiment of the invention, landslide hazard areas are identified based on the slope information, vegetation cover type, and precipitation distribution of the area where the power grid is located. These landslide hazard areas are regions within the power grid area where the mountain stability is low and there is a risk of landslides. The landslide volume of the hazard areas is predicted to obtain a predicted landslide volume value. Based on the moisture status of the hazard areas, the spatiotemporal variation rate of moisture and the range of moisture anomalies are determined. The monitoring results are then determined based on the predicted landslide volume value, the spatiotemporal variation rate of moisture, and the range of moisture anomalies. It is noteworthy that by identifying landslide hazard areas using mountain slope information, vegetation cover type, and precipitation distribution, and using the predicted landslide volume value, the spatiotemporal variation rate of moisture, and the range of moisture anomalies to determine the monitoring results, the influence of the mountain's slope and vegetation cover on landslides is fully considered. Furthermore, the landslide volume is predicted, overcoming the limitation of ground measurements in achieving large-scale, high-frequency monitoring. This achieves the technical effect of improving the accuracy of landslide monitoring, thereby solving the technical problem of low accuracy in landslide monitoring. Attached Figure Description

[0021] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0022] Figure 1 This is a flowchart of a landslide monitoring method according to an embodiment of the present invention;

[0023] Figure 2 This is a flowchart of an optional landslide monitoring method according to an embodiment of the present invention;

[0024] Figure 3 This is a flowchart of an optional method for identifying a landslide hazard area according to an embodiment of the present invention;

[0025] Figure 4 This is a flowchart of another optional method for identifying potential landslide hazard areas according to an embodiment of the present invention;

[0026] Figure 5 This is a schematic diagram of a landslide monitoring device according to an embodiment of the present invention. Detailed Implementation

[0027] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0028] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0029] Example 1

[0030] According to an embodiment of the present invention, an embodiment of a method for monitoring landslides is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0031] Figure 1 This is a flowchart of a landslide monitoring method according to an embodiment of the present invention, such as... Figure 1 As shown, the method includes the following steps:

[0032] Step S102: Based on the slope information, vegetation cover type and precipitation distribution of the area where the power grid is located, landslide hazard areas are identified. The landslide hazard areas are areas in the area where the power grid is located where the mountain stability is low and there is a risk of landslide.

[0033] The slope information mentioned in the above steps refers to the slope of the mountains in the area where the power grid is located. The slope is the ratio of the vertical direction to the horizontal direction of the earth's surface. It is usually used to describe the degree of inclination of the terrain. Steeper slopes may be more prone to landslides.

[0034] The vegetation cover type mentioned above refers to the plant species covering the surface of the area where the power grid is located. Different types of vegetation, such as coniferous forests, broad-leaved forests, and grasslands, have different physical and ecological characteristics. These characteristics affect the stability of the soil and its ability to retain moisture, thereby affecting the probability of landslides.

[0035] The precipitation distribution described in the above steps refers to the spatial distribution of precipitation in different regions or at different times over a period of time. Precipitation is one of the important factors that trigger landslides; heavy precipitation may saturate the soil, increasing the risk of landslides.

[0036] In one alternative embodiment, the Normalized Difference Vegetation Index (NDVI) can be obtained from satellite remote sensing imagery that reflects vegetation cover and vitality to determine the vegetation cover type in the power grid area. Then, the area is divided according to different vegetation cover types to obtain the distribution areas of different vegetation types. A Digital Elevation Model (DEM) is then used to determine the slope information of the mountain slopes in the distribution areas of different vegetation types.

[0037] For example, the NDVI of coniferous forests is generally between 0.5 and 0.8. If the NDVI of a certain area extracted from satellite remote sensing imagery is 0.6, then that area can be identified as a coniferous forest distribution area. Next, a digital elevation model (DEM) can be used to determine the slope information of these coniferous forest areas. Each pixel in the DEM data corresponds to an elevation value. The elevation change between adjacent pixels can be calculated to obtain the slope information. If the obtained slope is 35 degrees, then the area is determined to be a coniferous forest distribution area with a slope of 35°.

[0038] Based on the slope information of different vegetation types, areas with slopes greater than a slope threshold are identified. Satellite cloud imagery data of the periods with the highest cloud cover in these areas is obtained, and the precipitation distribution of these areas is determined using this data. The vegetation cover type, slope information, and precipitation distribution of each area are then input into a trained model. The model predicts the stability of the mountains in these areas, and areas with stability below a stability threshold are identified as potential landslide zones.

[0039] It should be noted that the slope threshold and stability threshold mentioned above can be set according to the actual application. No specific value is limited here, and the values ​​in the examples above are only examples. The values ​​involved in actual applications may not be limited to these.

[0040] Step S104: Predict the landslide volume in the landslide hazard area to obtain the predicted landslide volume value.

[0041] The predicted landslide volume in the above steps is the predicted landslide volume when a landslide occurs in the landslide-prone area. The landslide volume is the amount of earth and rock that slides down from the original location in the landslide-prone area during the entire landslide process.

[0042] In one optional embodiment, high-resolution satellite imagery of the landslide hazard area is acquired. Digital elevation model (DEM) data within the image coverage area is extracted from the high-resolution satellite imagery. GPS point cloud data of the landslide hazard area is obtained using Global Positioning System (GPS) technology. Based on the DEM data and GPS point cloud data, a three-dimensional digital elevation model is established using a triangulation mesh construction method. Mountain structure data is then extracted from the three-dimensional digital elevation model.

[0043] Historical mountain structure data from landslide-prone areas are used to train a neural network model, enabling it to predict landslide volume based on this data. The trained neural network model is then used to process the mountain structure data to generate predicted landslide volume values.

[0044] Step S106: Based on the moisture status of the landslide hazard area, determine the spatiotemporal variation rate of moisture and the range of moisture anomalies in the landslide hazard area, wherein the moisture status includes at least the soil moisture content.

[0045] The rate of change of moisture in the above steps is the rate of change of soil moisture content in the landslide hazard area. It is usually expressed by dividing the change by time, that is, the change in moisture content per unit time.

[0046] The abnormal water range in the above steps refers to the area of ​​the landslide hazard zone that is in an abnormal water state. An abnormal water state is a state in which the change in soil moisture content exceeds the normal or expected range.

[0047] The soil moisture content mentioned in the above steps refers to the amount of water contained in the soil.

[0048] In one optional embodiment, a trained gradient boosting decision tree model is used to predict the soil moisture content of the landslide hazard area at different times. Then, the soil moisture content of the landslide hazard area at different times is divided by the time interval between the different times to determine the spatiotemporal variation rate of moisture in the landslide hazard area. By comparing the spatiotemporal variation rate of moisture with a variation rate threshold, it is determined whether the landslide hazard area is in a state of abnormal moisture. If it is in a state of abnormal moisture, the area of ​​the landslide hazard area is taken as the range of abnormal moisture.

[0049] In another optional embodiment, the landslide hazard area is divided into different zones. A trained gradient boosting decision tree model is used to predict the soil moisture content of each zone, and then the spatiotemporal change rate of moisture in each zone is calculated. Based on the relationship between the change rate threshold and the spatiotemporal change rate of moisture in different zones, it is determined whether each zone is in an abnormal moisture state. The area of ​​each zone in an abnormal moisture state is determined, and the areas of all zones in an abnormal moisture state are summed to obtain the range of the abnormal moisture state.

[0050] It should be noted that the above-mentioned rate of change threshold can be set according to the actual application situation, and no specific value is limited here.

[0051] Step S108: Determine the monitoring results based on the predicted landslide volume, the spatiotemporal variation rate of moisture, and the range of moisture anomalies. The monitoring results are used to characterize the degree of danger in the landslide hazard area.

[0052] The monitoring results in the above steps may include, but are not limited to, high-risk, medium-risk, and low-risk.

[0053] In one optional embodiment, the predicted slope volume, the spatiotemporal rate of change of moisture, and the range of moisture anomalies can be normalized to between 0 and 1. Then, a weighted sum is calculated from the predicted slope volume, the spatiotemporal rate of change of moisture, and the range of moisture anomalies to obtain a weighted result. The weights can be set based on industry experience, reference materials, or actual conditions; no specific numerical limit is imposed on the weights here. Based on a pre-set mapping relationship between the weighted result and the monitoring result, the corresponding monitoring result is determined. For example, the mapping relationship between the weighted result and the monitoring result can be: a weighted result greater than or equal to 4 indicates a high-risk landslide hazard area; a weighted result greater than or equal to 2 and less than 4 indicates a medium-risk landslide hazard area; and a weighted result less than 2 indicates a low-risk landslide hazard area. If the weighted result is 4.7, the monitoring result indicates a high-risk landslide hazard area. It should be noted that the values ​​in the above examples are merely illustrative, and the values ​​involved in actual applications are not limited to these.

[0054] In this embodiment of the invention, landslide hazard areas are identified based on the slope information, vegetation cover type, and precipitation distribution of the area where the power grid is located. These landslide hazard areas are regions within the power grid area where the mountain stability is low and there is a risk of landslides. The landslide volume of the hazard areas is predicted to obtain a predicted landslide volume value. Based on the moisture status of the hazard areas, the spatiotemporal variation rate of moisture and the range of moisture anomalies are determined. The monitoring results are then determined based on the predicted landslide volume value, the spatiotemporal variation rate of moisture, and the range of moisture anomalies. It is noteworthy that by identifying landslide hazard areas using mountain slope information, vegetation cover type, and precipitation distribution, and by using the predicted landslide volume value, the spatiotemporal variation rate of moisture, and the range of moisture anomalies to determine the monitoring results, the impact of the mountain's slope and vegetation cover on landslides is fully considered. This achieves the technical effect of improving the accuracy of landslide monitoring, thereby solving the technical problem of low accuracy in landslide monitoring.

[0055] Optionally, predicting the landslide volume in the landslide hazard area to obtain a predicted landslide volume value includes: acquiring the mountain structure characteristics of the landslide hazard area, wherein the mountain structure characteristics include at least one of the following: mountain elevation, mountain slope, mountain aspect, vegetation index, precipitation, and rock edge movement rate; inputting the mountain structure characteristics into the landslide volume prediction model, and using the landslide volume prediction model to predict the landslide volume in the landslide hazard area to obtain a predicted landslide volume value.

[0056] The mountain structure features mentioned in the above steps are the physical and geometric properties of the mountain.

[0057] The mountain elevation mentioned above is the height of the highest point of the mountain relative to sea level or a certain reference surface, which can be used to determine the absolute height of the mountain and the undulation of the terrain.

[0058] The slope of the mountain mentioned above refers to the degree of inclination of the earth's surface on a horizontal plane, usually expressed as an angle or percentage. The greater the slope, the steeper the inclination of the earth's surface.

[0059] The slope aspect mentioned above refers to the direction of inclination of the mountain slope, usually measured clockwise from a reference direction. Slope aspect is crucial for understanding solar radiation, precipitation distribution, and the direction of surface water flow.

[0060] The vegetation index mentioned in the above steps is an indicator that measures vegetation density and health status, also known as NDVI.

[0061] The precipitation mentioned above refers to the total amount of water that falls to the ground within a certain period of time, usually measured in millimeters or inches. Precipitation is an important factor affecting landslide occurrence; increased precipitation increases soil moisture content, which may reduce soil stability.

[0062] The rock edge movement rate mentioned above refers to the speed at which the rock edge in the mountain moves due to weathering, erosion or other geological processes. A higher rock edge movement rate can indicate potential landslide activity.

[0063] The landslide volume prediction model in the above steps can be a backpropagation neural network (BP neural network model), a radial basis function network (RBF), or a convolutional neural network (CNN), but is not limited to these.

[0064] In one alternative embodiment, the elevation, slope, aspect, vegetation index, precipitation, and rock edge movement rate of the landslide hazard area can be extracted from a three-dimensional digital elevation model constructed based on DEM numerical elevation data and GPS point cloud data as the structural features of the mountain.

[0065] A backpropagation (BP) neural network model, also known as a landslide volume prediction model, is constructed. Historical mountain structure characteristics of the landslide-prone area are obtained as training data to train the BP neural network model. The model parameters are adjusted to achieve the best fit, resulting in a well-trained BP neural network model. The mountain structure characteristics are then input into the trained BP neural network model to obtain the predicted landslide volume.

[0066] For example, a dataset for training the BP neural network model was constructed, with landslide volume as the dependent variable and mountain elevation, slope, aspect, vegetation index, precipitation, and rock edge movement rate as independent variables. Mountain aspect was defined using azimuth angles, with true north as 0 degrees and measured clockwise: southeast from approximately 45 to 135 degrees, northeast from approximately 135 to 225 degrees, southwest from approximately 225 to 315 degrees, and northwest from approximately 315 to 45 degrees. 1000 sample points were selected as the training set, and another 500 sample points were used as the test set to train the BP neural network model, and the model parameters were adjusted to obtain the best fit.

[0067] During the prediction phase, the elevation of the mountain within the landslide hazard area was 950 meters, the slope was 18°, the aspect was 45°, the vegetation index was 0.6, the precipitation was 50 mm, and the rock edge movement rate was 1 meter per year. The structural characteristics of the mountain within the landslide hazard area were input into a trained BP neural network model to obtain the predicted landslide volume.

[0068] It should be noted that the values ​​in the examples above are merely illustrative, and the values ​​involved in actual applications may not be limited to these.

[0069] Optionally, based on the moisture status of the landslide hazard area, the spatiotemporal rate of moisture variation and the range of moisture anomalies in the landslide hazard area are determined, including: identifying at least one key analysis area within the landslide hazard area, wherein the at least one key analysis area is an area with anomalies in surface temperature within the landslide hazard area; acquiring surface temperature, atmospheric temperature, vegetation index, and precipitation of the at least one key analysis area at different times; inputting the surface temperature, atmospheric temperature, vegetation index, and precipitation of the at least one key analysis area at different times into a moisture prediction model, and using the moisture prediction model to predict the moisture status of the at least one key analysis area at different times, thereby obtaining the moisture status of the at least one key analysis area at different times; and determining the spatiotemporal rate of moisture variation and the range of moisture anomalies based on the moisture status of the at least one key analysis area at different times.

[0070] The surface temperature mentioned in the above steps refers to the temperature of natural features on the Earth's surface, such as rocks, soil, and water. It can be measured using thermal infrared remote sensing and is usually expressed in degrees Celsius.

[0071] The atmospheric temperature mentioned in the above steps refers to the average kinetic energy of gas molecules in the atmosphere at a certain altitude, and is usually expressed as air temperature.

[0072] The moisture prediction model in the above steps can be a gradient boosting decision tree model, but it is not limited to this. Gradient boosting decision tree is an ensemble learning algorithm that improves the predictive performance of a model by combining multiple weak predictors from decision trees. The basic idea of ​​gradient boosting is to first build a base decision tree model, and then gradually add new decision trees based on the model's prediction error. The prediction direction of the new trees is the gradient direction that reduces the overall model error. A well-trained gradient boosting decision tree model can be used to predict the moisture status of at least one key analysis area at different times.

[0073] In one optional embodiment, the surface temperature of the landslide hazard area can be obtained, and areas with abnormal surface temperatures can be identified as key analysis areas, with at least one key analysis area.

[0074] The measured soil moisture content and corresponding surface temperature, atmospheric temperature, vegetation index, and precipitation at different historical moments in the key analysis area are used as training and testing datasets. The training and testing datasets are divided proportionally, with a preferred ratio of 7:3, but not limited to this. The gradient boosting decision tree model is trained using the training dataset and tested using the testing dataset. The learning rate and the number of trees are tuned using a grid search method to optimize the model's performance, resulting in a well-trained gradient boosting decision tree model.

[0075] The surface temperature of at least one key analysis area at different times is obtained using satellite thermal infrared imagery. Atmospheric temperature of the same area at different times is acquired using satellite sensors. NDVI (Normally Displaced Vegetation Index) of the same area at different times is obtained using satellite remote sensing imagery. Precipitation of the same area at different times is obtained using satellite cloud imagery. The surface temperature, atmospheric temperature, NDVI, and precipitation of the same area at different times are then input into a trained gradient boosting decision tree model to obtain the moisture status of the same area at different times as predicted by the gradient boosting decision tree model.

[0076] For example, surface temperature data at times A, B, and C are obtained using satellite thermal infrared imagery. The surface temperature at time A is 28°C, at time B it is 32°C, and at time C it is 30°C. Then, atmospheric temperature data, vegetation index, and precipitation are acquired using satellite sensors. At time A, the atmospheric temperature is 25°C, the vegetation index is 0.5, and the precipitation is 10 mm; at time B, the atmospheric temperature is 28°C, the vegetation index is 0.6, and the precipitation is 15 mm; and at time C, the atmospheric temperature is 27°C, the vegetation index is 0.7, and the precipitation is 12 mm. The data at time A, specifically a surface temperature of 28°C, an atmospheric temperature of 25°C, a vegetation index of 0.5, and precipitation of 10 mm, are input into a trained gradient boosting decision tree model. The model predicts a soil moisture content of 12%. The predictions for soil moisture content at times B and C are similar and will not be described further. It should be noted that the values ​​in the examples above are merely illustrative, and the values ​​involved in actual applications may not be limited to these.

[0077] After obtaining the moisture status of at least one key analysis area at different times, i.e., the layer moisture content, the spatiotemporal variation rate of moisture and the range of moisture anomalies can be determined based on the moisture status of at least one key analysis area at different times.

[0078] Optionally, based on the moisture status of at least one key analysis area at different times, the spatiotemporal rate of moisture change and the range of moisture anomalies are determined, including: determining the moisture change rate of at least one key analysis area based on the moisture status of at least one key analysis area at different times; determining the spatiotemporal rate of moisture change based on the moisture change rates of multiple key analysis areas; comparing the moisture change rate of at least one key analysis area with a preset change rate to determine whether at least one key analysis area is in a moisture anomaly state, wherein the moisture anomaly state is used to characterize that the moisture change rate is greater than the preset change rate; and determining the range of moisture anomalies based on the range of key analysis areas in a moisture anomaly state.

[0079] The preset rate of change in the above steps is a pre-set rate of change that can be preset according to the actual application. No specific value of the predicted rate of change is restricted here.

[0080] The moisture change rate in the above steps can characterize the moisture changes in the key analysis area, and the moisture change rates of multiple key analysis areas can be used to determine the spatiotemporal change rate of moisture.

[0081] In one optional embodiment, based on the soil moisture content of at least one key analysis area at different times, the spatiotemporal variation rate of moisture in at least one key analysis area at different times is calculated. The formula for calculating the moisture variation rate is as follows:

[0082] k = ΔM1 / (t2-t1);

[0083] Where k is the spatiotemporal variation rate of moisture, ΔM1 is the difference between the soil moisture content at time t2 and the soil moisture content at time t1, and time t2 is later than time t1.

[0084] One approach is to calculate the average rate of moisture change across multiple key analysis areas within a landslide hazard zone and use it as the spatiotemporal rate of moisture change. Alternatively, one can arbitrarily select a key analysis area within a landslide hazard zone and use the rate of moisture change in that key analysis area as the spatiotemporal rate of moisture change. However, the methods for determining the spatiotemporal rate of moisture change are not limited to these approaches.

[0085] The moisture change rate of the key analysis area is compared with the preset change rate. If the moisture change rate is greater than the preset change rate, the key analysis area is in a state of moisture anomaly; if the moisture change rate is not greater than the preset change rate, the key analysis area is not in a state of moisture anomaly. The areas of the key analysis areas in a state of moisture anomaly are added together to obtain the area of ​​moisture anomaly, that is, the range of moisture anomaly.

[0086] Optionally, at least one key analysis area is identified within the landslide hazard area, including: acquiring satellite thermal infrared images of the landslide hazard area; processing the satellite thermal infrared images to obtain thermal infrared radiance; processing the thermal infrared radiance using a single scattering inversion algorithm to obtain surface temperature data of the landslide hazard area; and using a clustering algorithm to identify anomalous data in the surface temperature data, determining the areas corresponding to the anomalous data as key analysis areas.

[0087] The satellite thermal infrared imagery mentioned above is a surface temperature image acquired from a satellite using an infrared-band sensor. Since different objects emit different amounts of thermal radiation at different temperatures, this satellite thermal infrared imagery can reveal the temperature distribution on the Earth's surface.

[0088] The thermal infrared radiance mentioned above is a physical quantity characterizing the intensity of radiant energy emitted by the Earth's surface in the thermal infrared band. Thermal infrared radiance is the radiative characteristic of each pixel in a thermal infrared image and is directly related to the Earth's surface temperature.

[0089] The single-scattering inversion algorithm in the above steps is an algorithm that inverts the surface temperature based on thermal infrared radiance. It estimates the surface temperature by solving the nonlinear relationship between thermal infrared radiance and surface temperature.

[0090] The clustering algorithm in the above steps can be Density-Based Spatial Clustering of Applications with Noise (DBSCAN). DBSCAN is a density-based clustering algorithm that can divide regions with sufficiently high density into clusters, without requiring a pre-specified number of clusters. It can identify clusters of arbitrary shapes and has good robustness to noisy data. It can also be a hierarchical clustering algorithm, or a Mean Shift algorithm, but is not limited to these.

[0091] In one optional embodiment, satellite thermal infrared imagery of the landslide hazard area is acquired. The original satellite thermal infrared imagery is preprocessed, including radiometric correction and denoising, to obtain preprocessed thermal infrared radiance. The thermal infrared radiance is then input into a single-scattering inversion algorithm, which iteratively calculates the nonlinear relationship between thermal infrared radiance and surface temperature to obtain surface temperature data.

[0092] The DBSCAN algorithm is used to identify whether there are abnormal data in the surface temperature data of the landslide hazard area. If abnormal surface temperature data is detected, the landslide hazard area is extracted as the key analysis area.

[0093] Optionally, the monitoring results are determined based on the predicted landslide volume, the spatiotemporal rate of moisture change, and the range of moisture anomalies, including: determining the weights of the predicted landslide volume, the spatiotemporal rate of moisture change, and the range of moisture anomalies respectively; determining the hazard coefficient of the landslide hazard area based on the predicted landslide volume, the spatiotemporal rate of moisture change, the range of moisture anomalies, the weights of the predicted landslide volume, the weights of the spatiotemporal rate of moisture change, and the weights corresponding to the range of moisture anomalies; and determining the monitoring results based on the hazard coefficient of the landslide hazard area.

[0094] The hazard coefficient in the above steps is used to quantify the risk of landslides occurring in landslide-prone areas, that is, to measure the potential risk level of landslide-prone areas.

[0095] In one optional embodiment, the predicted landslide volume, the spatiotemporal rate of change of moisture, and the range of moisture anomalies can be normalized. Then, the weights of the predicted landslide volume, the spatiotemporal rate of change of moisture, and the range of moisture anomalies are preset according to the actual situation. Subsequently, the hazard coefficient of the landslide hazard area is calculated based on the result of the normalization and the weights. The formula for determining the hazard coefficient of the landslide hazard area is as follows:

[0096] M = A*q1 + B*q2 + C*q3;

[0097] Where M is the hazard coefficient, A is the predicted landslide volume after normalization, B is the normalized spatiotemporal variation rate of moisture, C is the normalized range of moisture anomalies, q1 is the weight of the predicted landslide volume, q2 is the weight of the spatiotemporal variation rate of moisture, and q3 is the weight of the range of moisture anomalies.

[0098] For example, the predicted landslide volume in the potential landslide area is 1000 cubic meters, the spatiotemporal variation rate of moisture is 5 cubic meters per day, and the anomalous area, i.e., the moisture anomalous range, is 200 square meters. First, normalization is performed, normalizing the predicted landslide volume, the spatiotemporal variation rate of moisture, and the anomalous area to a range between 0 and 1. After normalization, the predicted landslide volume, the spatiotemporal variation rate of moisture, and the anomalous area are 0.5, 0.5, and 0.4, respectively, with influence weights of 4, 3, and 3, respectively.

[0099] The risk factor is calculated as follows: 0.5*4 + 0.5*3 + 0.4*3 = 2 + 1.5 + 1.2 = 4.7. Landslide events are classified into severity levels based on their risk factors. A risk factor greater than or equal to 4 is considered high risk; a risk factor greater than or equal to 2 and less than 4 is considered medium risk; and a risk factor less than 2 is considered low risk. Based on the calculation results, the risk factor for this landslide hazard area is 4.7, therefore the monitoring results indicate that this landslide risk area is classified as high risk.

[0100] It should be noted that the values ​​in the examples above are merely illustrative, and the values ​​involved in actual applications may not be limited to these.

[0101] Optionally, determining landslide hazard areas based on mountain slope information, vegetation cover type, and precipitation distribution in the area where the power grid is located includes: determining at least one slope anomalous area based on mountain slope information in the area where the power grid is located; inputting the vegetation cover type and precipitation distribution of the at least one slope anomalous area into a mathematical model, using the mathematical model to predict the mountain stability index of the at least one slope anomalous area, and obtaining the mountain stability index of the at least one slope anomalous area; and determining landslide hazard areas within the at least one slope anomalous area based on the mountain stability index of the at least one slope anomalous area.

[0102] The slope anomaly areas mentioned in the above steps refer to areas within the power grid region where slope anomalies require further monitoring.

[0103] The mathematical model in the above steps can be the Transient Rainfall Infiltration and Grid-based Regional Slope-stability model (TRIGRS). The TRIGRS model combines the principles of rainfall infiltration and fluid dynamics. It can predict the possibility of slope instability by calculating the impact of groundwater flow on soil stability and obtain the mountain stability index, but it is not limited to this.

[0104] The mountain stability index mentioned above is a quantitative indicator used to assess the stability of mountain slopes under specific environmental conditions.

[0105] In one alternative embodiment, at least one slope aberration area in the power grid area is determined based on whether the slope information of each zone in the power grid area is within the normal slope range.

[0106] Satellite cloud imagery data of the Guan slope anomaly area at different time periods was acquired, and cloud cover distribution data were compared across these periods. The period with the highest cloud cover was identified, and the cloud imagery for that period was used as the key period cloud imagery. Based on the spatial distribution information displayed in the key period cloud imagery, the cloud region was classified into different types using a deep convolutional neural network. Representative sample areas containing multiple cloud types were obtained as analysis areas for different rainfall distribution types. For each cloud type sample area, the quantitative precipitation estimation model (QPE) was applied to calculate the rainfall distribution corresponding to different cloud types. The calculation results were corrected against ground-measured rainfall data, and the parameters of the QPE model were adjusted to improve accuracy. Finally, the rainfall distribution corresponding to different cloud types in the Guan slope anomaly area was accurately calculated.

[0107] The NDVI (Depth-to-Vegetation Index) was acquired using satellite remote sensing imagery to determine the vegetation cover type in areas with anomalies in slope. Pre-defined slope ranges and vegetation cover types prone to landslides were established to determine whether landslide-prone terrain existed in each cloud type sample area. If such terrain existed, the precipitation distribution, slope, and vegetation cover type within the area at different times were obtained, and the TRIGRS model was used to calculate the mountain stability index under different rainfall conditions.

[0108] For example, satellite cloud imagery data for July, August, and September can be obtained for areas with abnormal slopes, including different types of dense clouds, thin clouds, and cumulonimbus clouds. First, a cloud detection algorithm is used to analyze cloud cover, determining that the highest cloud cover occurs in mid-August. Therefore, the cloud imagery for August 15th is extracted as the key period. Next, the cloud images are classified using a deep convolutional neural network into five categories: thin clouds, thick clouds, and cumulonimbus clouds. Three of these cloud categories are selected as samples, and cloud image slices are extracted for the thick cloud, thin cloud, and cumulonimbus cloud regions respectively. A satellite quantitative precipitation estimation model (QPE) is launched for each cloud slice, which estimates precipitation under the cloud based on parameters such as cloud top temperature and cloud thickness. The inverted precipitation is 25 mm for thick clouds, 5 mm for thin clouds, and 50 mm for cumulonimbus clouds. The results are compared with the actual rainfall observed at the meteorological station. If there is a large error, the parameters within the model need to be adjusted, and the calculation iterated again until the inverted results are close to the measured results. Next, by comparing slope range and vegetation cover type, the presence of landslide-prone terrain in each cloud type sample area was checked. If landslide-prone terrain was found in thick cloud and cumulonimbus cloud areas, its coordinates were extracted. Finally, a quantitative stability assessment model was used, inputting different rainfall distributions, slopes, and vegetation cover types into the TRIGRS model to calculate the mountain stability index under different precipitation conditions. It should be noted that the values ​​in the examples above are only illustrative, and the values ​​involved in practical applications may not be limited to these.

[0109] When identifying potential landslide areas, the stability level of the mountain can be determined based on the mountain stability index. Areas with potentially unstable or unstable slope anomalies are designated as unstable anomaly areas. Multi-temporal satellite imagery of these unstable anomaly areas is acquired, using Landsat satellite (one type) to obtain imagery at four different time phases. Based on this multi-temporal satellite imagery, a Long Short-Term Memory (LSTM) network model is constructed to identify areas where vegetation cover has changed. The NDVI (Depth-Value Vegetation Index) is calculated for each time phase of the imagery, using the following formula:

[0110] NDVI=(NI R-RED) / (NI R+RED);

[0111] Wherein, NDVI is the vegetation index, NI R is the value in the near-infrared band, and RED is the value in the red band.

[0112] Then, the texture features of the image at each time phase are calculated using the gray-level co-occurrence matrix. Using the vegetation index (NDVI) and texture features of the image at each time phase as input values, a trained Long Short-Term Memory (LSTM) model is used to predict the vegetation cover of the image at each time phase. By comparing the prediction results of adjacent time phases, it is determined whether the vegetation cover of each pixel has changed, and the changed areas are extracted as landslide hazard areas.

[0113] Optionally, determining at least one slope anomaly area based on the slope information of the power grid area includes: dividing the power grid area into multiple areas to be determined; obtaining the slope information of the multiple areas to be determined; determining whether the slope information of the multiple areas to be determined is within a preset slope range; and determining the areas to be determined whose slope information is within the preset slope range as slope anomaly areas.

[0114] The areas to be determined in the previous step can be divided according to the type of vegetation cover on the surface of the area, the soil moisture content, or the soil type, but the division method is not limited to these.

[0115] The preset slope range in the above steps is a pre-set slope range that can be preset according to the actual application. No specific value is limited to the preset slope range here.

[0116] In one optional embodiment, the vegetation index (NDVI) is acquired through satellite remote sensing imagery to determine the vegetation cover type of the power grid area. The area is then divided according to the vegetation cover type, resulting in multiple areas to be determined. A digital elevation model (DEM) is used to process these multiple areas, and the slope information of the mountains in each area is determined based on the DEM data output by the DEM. Whether the slope information of the multiple areas falls within a preset slope range is used to determine whether the areas to be determined are slope anomaly areas.

[0117] For example, the preset slope range is 30 to 45 degrees. If the slope information of the area to be determined is 35 degrees, then the area to be determined can be identified as a slope anomaly area. It should be noted that the values ​​in the above example are only examples, and the values ​​involved in actual applications may not be limited to these.

[0118] Optionally, based on the mountain stability index of at least one slope anomaly area, a landslide hazard area is determined in at least one slope anomaly area, including: determining slope anomaly areas with mountain stability indices within a preset range as candidate landslide hazard areas; determining the rock edge movement rate of candidate landslide hazard areas based on the bare rock distribution of candidate landslide hazard areas; and determining candidate landslide hazard areas as landslide hazard areas if the rock edge movement rate of candidate landslide hazard areas is greater than a preset rate.

[0119] The candidate landslide hazard areas in the above steps are slope anomaly areas with low mountain stability index, and the landslide hazard areas are selected from the candidate landslide hazard areas.

[0120] The bare rock distribution mentioned in the above steps refers to the distribution of exposed rocks on the surface within the candidate landslide hazard area.

[0121] The preset index range in the above steps is a pre-set data range that can be preset according to the actual application situation. Here, we do not impose any restrictions on the specific value of the preset index range.

[0122] In one optional embodiment, it is determined whether the mountain stability index of the slope anomaly area is within a preset index range. If it is within the preset index range, the slope anomaly area is determined as a candidate landslide hazard area.

[0123] For example, based on the mountain stability index, mountains are classified into stable, potentially unstable, and unstable levels. The levels are determined by the index value: a stability index above 8 indicates stable, 6-8 indicates potentially unstable, and below 6 indicates unstable. Areas with slope anomalies classified as potentially unstable or unstable are designated as candidate landslide hazard areas. That is, with a preset index range of less than 8, areas with slope anomalies having an index less than 8 are identified as candidate landslide hazard areas. It should be noted that the values ​​in the above example are merely illustrative, and the values ​​involved in practical applications may not be limited to these.

[0124] Based on surface information within the candidate landslide hazard area, it is determined whether bare rock exists within the area. If bare rock is present, multi-temporal images of the bare rock area are acquired to obtain the distribution of bare rock within the candidate landslide hazard area. The Scale-Invariant Feature Transform (SIFT) algorithm is used to match and analyze the multi-temporal images containing the bare rock distribution information to calculate the rock edge movement rate. If the rock edge displacement rate is greater than a preset rate, the candidate landslide hazard area is identified as a landslide hazard area.

[0125] For example, surface information, including reflectance data, texture features, and edge features at different wavelengths, is extracted from multi-temporal satellite imagery of the candidate landslide hazard area to determine whether bare rock is present. The extracted reflectance data, texture features, and edge features are used as input values ​​to construct a discriminant model based on a Support Vector Machine (SVM) to automatically identify bare rock distribution areas and determine their distribution. If bare rock is present, image matching analysis is performed using the SIFT algorithm based on multi-temporal satellite imagery of the bare rock area. The distance between feature points in different temporal images is calculated, and feature point matching is performed. Based on the changes in the registered feature point positions, the displacement information of the rock edge is calculated, including its horizontal and vertical displacement. The displacement data is analyzed to determine the rock edge movement rate, calculated using the following formula:

[0126]

[0127] Where X is the horizontal displacement of the rock, Y is the vertical displacement of the rock, and T is the time interval.

[0128] The preset rate is 5 cm per day. When the displacement rate at the rock edge exceeds 5 cm per day, the candidate landslide hazard area is identified as a landslide hazard area and requires focused monitoring and early warning. It should be noted that the values ​​in the above example are merely illustrative, and the values ​​involved in actual applications may not be limited to these.

[0129] Optionally, the method further includes: determining landslide response strategies based on monitoring results; and carrying out emergency treatment on power grid equipment within the landslide hazard area according to the landslide response strategies.

[0130] The landslide response strategies outlined above are emergency measures to be taken for power grid equipment when monitoring results indicate a landslide hazard in the area where the power grid is located. There is a correlation between landslide response strategies and monitoring results; different monitoring results correspond to different landslide response strategies.

[0131] In one alternative embodiment, landslide hazard areas with high risk can be identified based on monitoring results, and landslide response strategies can be adopted to reduce the impact of landslides on power grid equipment.

[0132] The strategy for responding to landslides classified as high-risk by monitoring is as follows: First, check if there are any power facilities within the landslide hazard area. If power grid equipment exists, calculate the outage area by obtaining the number of transmission lines in the area. Based on the damaged outage area, develop targeted emergency repair and backup power supply plans, and implement preventative measures in advance to ensure power supply. Simultaneously, obtain backup capacity information from other power suppliers in the vicinity of the high-risk landslide hazard area to determine if feeder support can be provided from adjacent power networks. If it is confirmed that an adjacent power grid can be used to absorb part of the load, a pre-set automatic power transfer procedure for network failures will be implemented for rapid response in the event of a power outage.

[0133] If the monitoring results for a landslide hazard area indicate a high risk, emergency measures should be taken for the power grid equipment within the area, following the landslide response strategy described above. Specifically, it's necessary to check if any power facilities are located in the area. If power grid equipment exists, assuming there are three transmission lines in the area, the outage area is calculated based on the number of transmission lines. If each transmission line supplies an average area of ​​5 square kilometers, the total outage area is 3 * 5 = 15 square kilometers. Based on the calculated outage area, a targeted emergency repair and backup power supply plan is developed, and preventative measures are implemented in advance to ensure power supply. Simultaneously, information on the backup capacity of other power suppliers surrounding the high-risk landslide hazard area is obtained to determine whether feeder support can be provided through adjacent power networks. For example, if supplier A has a backup capacity of 500 MW and supplier B has a backup capacity of 800 MW, the backup capacity information is used to determine whether a portion of the load can be taken over by an adjacent power grid. If the power outage area requires 200 MW of power, and Supplier A's backup capacity is 500 MW, exceeding the demand of 200 MW, then Supplier A in the adjacent power grid can take over part of the load. In this case, a pre-set automatic power transfer procedure for network failures is implemented to respond quickly in the event of a power outage and transfer part of the load to Supplier A in the adjacent power grid.

[0134] It should be noted that the values ​​in the examples above are merely illustrative, and the values ​​involved in actual applications may not be limited to these.

[0135] The following description uses a preferred embodiment. Figure 2 This is a flowchart of an optional landslide monitoring method according to an embodiment of the present invention, such as... Figure 2 As shown:

[0136] Step S202: Use GPS technology to obtain mountain structure data of the landslide hazard area, establish a three-dimensional digital elevation model, and predict the landslide volume.

[0137] Step S204: Analyze soil moisture changes in the landslide hazard area using satellite infrared thermal imagery, monitor moisture anomalies, and calculate the spatiotemporal rate of moisture change and the area of ​​anomalies.

[0138] The abnormal area in the above steps is also the moisture abnormal area mentioned above.

[0139] Step S206: Determine the severity level of the landslide event based on the predicted landslide volume, the spatiotemporal variation rate of moisture, and the area of ​​the abnormal range. Based on the severity level of the landslide event in the landslide hazard area, determine whether the landslide hazard area is a high-risk landslide area.

[0140] Step S208: Take preventative measures to reduce the impact of the landslide on power grid equipment.

[0141] Figure 3 This is a flowchart illustrating an optional method for identifying potential landslide hazard areas according to an embodiment of the present invention, such as... Figure 3 As shown:

[0142] Step S302: Extract the slope and vegetation cover type of the key areas of interest, and analyze the precipitation distribution using satellite cloud images.

[0143] The key area of ​​concern in the above steps is the area with abnormal slope mentioned above.

[0144] Step S304: Determine the mountain stability level by combining slope and vegetation cover type, and determine the target area based on the mountain stability level.

[0145] The mountain stability level in the above steps can be determined based on the mountain stability coefficient mentioned above.

[0146] Step S306: Use satellite imagery to determine the surface changes in the target area and identify potential landslide hazard zones.

[0147] The target area in the above steps is the candidate landslide hazard area mentioned above.

[0148] Figure 4 This is a flowchart of another optional method for identifying potential landslide hazard areas according to an embodiment of the present invention, such as... Figure 4 As shown:

[0149] Step S402: Acquire multi-temporal satellite images.

[0150] Step S404: Identify areas of vegetation cover change.

[0151] Step S406: Extract the changed region.

[0152] Step S408: Determine the distribution of bare rock.

[0153] Step S410: Obtain multi-temporal images of the bare rock area.

[0154] Step S412: Perform matching analysis using a feature matching algorithm.

[0155] Step S414: Calculate the rock edge movement rate.

[0156] Step S416: Compare with the preset first threshold to locate the landslide hazard area.

[0157] The preset first threshold in the above steps is the preset rate mentioned above.

[0158] Example 2

[0159] According to an embodiment of the present invention, an embodiment of a landslide monitoring device is provided. The device can perform the landslide monitoring method provided in Embodiment 1 above. The specific implementation method and preferred application scenario are the same as those in Embodiment 1 above, and will not be repeated here.

[0160] Figure 5 This is a schematic diagram of a landslide monitoring device according to an embodiment of the present invention, as shown below. Figure 5 As shown, the landslide monitoring device includes:

[0161] The first determining module 50 is used to determine landslide hazard areas based on the mountain slope information, vegetation cover type and precipitation distribution in the area where the power grid is located. The landslide hazard areas are areas in the area where the power grid is located where the mountain stability is low and there is a risk of landslide.

[0162] Prediction module 52 is used to predict the landslide volume in the landslide hazard area and obtain the predicted landslide volume value;

[0163] The second determining module 54 is used to determine the spatiotemporal variation rate of moisture and the range of moisture anomalies in the landslide hazard area based on the moisture status of the landslide hazard area, wherein the moisture status includes at least the soil moisture content.

[0164] The third determining module 56 is used to determine the monitoring results based on the predicted landslide volume, the spatiotemporal change rate of moisture, and the range of moisture anomalies. The monitoring results are used to characterize the degree of danger in the landslide hazard area.

[0165] Optionally, the prediction module includes: a first acquisition unit, used to acquire the mountain structure characteristics of the landslide hazard area, wherein the mountain structure characteristics include at least one of the following: mountain elevation, mountain slope, mountain aspect, vegetation index, precipitation, and rock edge movement rate; and a first prediction unit, used to input the mountain structure characteristics into the landslide volume prediction model, and use the landslide volume prediction model to predict the landslide volume of the landslide hazard area to obtain the predicted landslide volume value.

[0166] Optionally, the second determining module includes: a first determining unit, used to determine at least one key analysis area within the landslide hazard area, wherein the at least one key analysis area is an area with abnormal surface temperature within the landslide hazard area; a second acquiring unit, used to acquire surface temperature, atmospheric temperature, vegetation index, and precipitation of the at least one key analysis area at different times; a second predicting unit, used to input the surface temperature, atmospheric temperature, vegetation index, and precipitation of the at least one key analysis area at different times into a moisture prediction model, and use the moisture prediction model to predict the moisture status of each key analysis area at different times, thereby obtaining the moisture status of the at least one key analysis area at different times; and a second determining unit, used to determine the spatiotemporal rate of moisture variation and the range of moisture anomalies based on the moisture status of the at least one key analysis area at different times.

[0167] Optionally, the second determining unit is further configured to: determine the moisture change rate of at least one key analysis area based on the moisture status of at least one key analysis area at different times; determine the spatiotemporal moisture change rate based on the moisture change rates of multiple key analysis areas; compare the moisture change rate of at least one key analysis area with a preset change rate to determine whether at least one key analysis area is in a moisture anomaly state, wherein the moisture anomaly state is used to characterize that the moisture change rate is greater than the preset change rate; and determine the moisture anomaly range according to the range of the key analysis area in the moisture anomaly state.

[0168] Optionally, the first determining unit is further configured to acquire satellite thermal infrared images of the landslide hazard area; process the satellite thermal infrared images to obtain thermal infrared radiation brightness; process the thermal infrared radiation brightness using a single scattering inversion algorithm to obtain surface temperature data of the landslide hazard area; and use a clustering algorithm to identify abnormal data in the surface temperature data and determine the area corresponding to the abnormal data as the key analysis area.

[0169] Optionally, the third determining module includes: a third determining unit, used to determine the weights of the predicted landslide volume, the spatiotemporal change rate of moisture, and the range of moisture anomalies; a third predicting unit, used to determine the hazard coefficient of the landslide hazard area based on the predicted landslide volume, the spatiotemporal change rate of moisture, the range of moisture anomalies, and the weights corresponding to the predicted landslide volume, the spatiotemporal change rate of moisture, and the range of moisture anomalies; and a fourth determining unit, used to determine the monitoring results based on the hazard coefficient of the landslide hazard area.

[0170] Optionally, the first determining module includes: a fifth determining unit, used to determine at least one slope anomaly area based on the slope information of the mountain in the power grid area; a fourth predicting unit, used to input the vegetation cover type and precipitation distribution of the at least one slope anomaly area into a mathematical model, and use the mathematical model to predict the mountain stability index of the at least one slope anomaly area to obtain the mountain stability index of the at least one slope anomaly area; and a sixth determining unit, used to determine the landslide hazard area in the at least one slope anomaly area based on the mountain stability index of the at least one slope anomaly area.

[0171] Optionally, the fifth determining unit is also used to divide the area where the power grid is located into multiple areas to be determined; obtain the slope information of the multiple areas to be determined; determine whether the slope information of the multiple areas to be determined is within a preset slope range; and determine the areas to be determined whose slope information is within the preset slope range as slope abnormal areas.

[0172] Optionally, the sixth determining unit is also used to determine the slope anomaly area where the mountain stability index is within the preset index range as a candidate landslide hazard area; determine the rock edge movement rate of the candidate landslide hazard area based on the bare rock distribution of the candidate landslide hazard area; and determine the candidate landslide hazard area as a landslide hazard area if the rock edge movement rate of the candidate landslide hazard area is greater than the preset rate.

[0173] Optionally, the third determining module is also used to determine landslide response strategies based on monitoring results; and to carry out emergency treatment on power grid equipment in the landslide hazard area according to the landslide response strategies.

[0174] Example 3

[0175] According to an embodiment of the present invention, an electronic device is also provided, comprising: a memory storing an executable program; and a processor for running the program, wherein the program executes the landslide monitoring method of Embodiment 1 during runtime.

[0176] Example 4

[0177] Embodiments of this application also provide a computer-readable storage medium, which includes a stored executable program, wherein, when the executable program is running, it controls the device where the computer-readable storage medium is located to execute the landslide monitoring method of various embodiments of the present invention.

[0178] Example 5

[0179] Embodiments of this application also provide a computer program product, including a computer program that, when executed by a processor, implements the landslide monitoring method in various embodiments of the present invention.

[0180] Example 6

[0181] Embodiments of this application also provide a computer program product, including a non-volatile computer-readable storage medium for storing a computer program, which, when executed by a processor, implements the landslide monitoring method in various embodiments of the present invention.

[0182] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0183] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0184] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0185] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0186] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0187] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0188] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for monitoring landslides, characterized in that, include: Based on the slope information, vegetation cover type and precipitation distribution of the area where the power grid is located, landslide hazard areas are identified. These landslide hazard areas are areas in the area where the power grid is located where the mountain stability is low and there is a risk of landslides. Predict the landslide volume in the landslide hazard area to obtain the predicted landslide volume value; Based on the moisture status of the landslide hazard area, determine the spatiotemporal variation rate of moisture and the range of moisture anomalies in the landslide hazard area, wherein the moisture status includes at least the soil moisture content; The monitoring results are determined based on the predicted landslide volume, the spatiotemporal rate of change of moisture, and the range of moisture anomalies, wherein the monitoring results are used to characterize the degree of danger of the landslide hazard area; Specifically, based on the moisture status of the landslide hazard area, the spatiotemporal rate of moisture variation and the range of moisture anomalies in the landslide hazard area are determined, including: Identify at least one key analysis area within the landslide hazard area, wherein the at least one key analysis area is an area with abnormal surface temperature within the landslide hazard area; Obtain the surface temperature, atmospheric temperature, vegetation index, and precipitation of at least one key analysis area at different times; The surface temperature, atmospheric temperature, vegetation index and precipitation of the at least one key analysis area at different times are input into the water prediction model. The water prediction model is used to predict the water status of the at least one key analysis area at different times, so as to obtain the water status of the at least one key analysis area at different times. Based on the moisture status of the at least one key analysis area at different times, the spatiotemporal rate of moisture change and the range of moisture anomalies are determined.

2. The method for monitoring landslides according to claim 1, characterized in that, Predicting the landslide volume in the landslide hazard area to obtain the predicted landslide volume value includes: Obtain the structural characteristics of the mountain in the landslide hazard area, wherein the structural characteristics include at least one of the following: mountain elevation, mountain slope, mountain aspect, vegetation index, precipitation, and rock edge movement rate; The mountain structure features are input into the landslide volume prediction model, and the landslide volume prediction model is used to predict the landslide volume in the landslide hazard area to obtain the predicted landslide volume value.

3. The landslide monitoring method according to claim 1, characterized in that, Based on the moisture status of the at least one key analysis area at different times, the spatiotemporal rate of moisture change and the range of moisture anomalies are determined, including: Based on the moisture status of the at least one key analysis area at different times, determine the moisture change rate of the at least one key analysis area; The spatiotemporal rate of moisture change is determined based on the rate of moisture change in the at least one key analysis area. By comparing the rate of change of moisture in the at least one key analysis area with a preset rate of change, it is determined whether the at least one key analysis area is in an abnormal moisture state, wherein the abnormal moisture state is used to characterize that the rate of change of moisture is greater than the preset rate of change. The range of the moisture anomaly is determined based on the extent of the key analysis area that is in a state of moisture anomaly.

4. The method for monitoring landslides according to claim 1, characterized in that, Identify at least one key analysis area within the landslide hazard zone, including: Acquire satellite thermal infrared images of the landslide hazard area; The thermal infrared image of the satellite is processed to obtain the thermal infrared radiation brightness; The thermal infrared radiation brightness was processed using a single scattering inversion algorithm to obtain the surface temperature data of the landslide hazard area; Clustering algorithms are used to identify anomalous data in surface temperature data, and the areas corresponding to the anomalous data are identified as the key analysis areas.

5. The method for monitoring landslides according to claim 1, characterized in that, The monitoring results are determined based on the predicted landslide volume, the spatiotemporal rate of change of moisture, and the range of moisture anomalies, including: The weights of the predicted landslide volume, the spatiotemporal rate of change of moisture, and the range of moisture anomalies are determined respectively. Based on the predicted landslide volume, the spatiotemporal rate of water variation, the range of water anomalies, and the weights corresponding to the predicted landslide volume, the spatiotemporal rate of water variation, and the range of water anomalies, the risk coefficient of the landslide hazard area is determined. The monitoring results are determined based on the risk coefficient of the landslide hazard area.

6. The method for monitoring landslides according to claim 1, characterized in that, Based on information on mountain slope, vegetation cover type, and precipitation distribution in the area where the power grid is located, landslide hazard zones were identified, including: Identify at least one area with an abnormal slope based on the mountain slope information of the power grid area; The vegetation cover type and precipitation distribution of the at least one slope anomaly area are input into the mathematical model, and the mountain stability index of the at least one slope anomaly area is predicted using the mathematical model to obtain the mountain stability index of the at least one slope anomaly area. Based on the mountain stability index of the at least one slope anomaly area, the landslide hazard area in the at least one slope anomaly area is determined.

7. The method for monitoring landslides according to claim 6, characterized in that, Based on the slope information of the mountains in the area where the power grid is located, at least one area with anomaly in slope was identified, including: The area where the power grid is located is divided into multiple areas to be determined; Obtain the slope information of the multiple areas to be determined; Determine whether the slope information of the multiple areas to be determined is within a preset slope range; The area to be determined where the slope information of the mountain is within the preset slope range is identified as the slope anomaly area.

8. The method for monitoring landslides according to claim 6, characterized in that, Based on the mountain stability index of the at least one slope anomaly area, the landslide hazard area within the at least one slope anomaly area is determined, including: Areas with abnormal slopes where the mountain stability index is within a preset range are identified as candidate landslide hazard areas. Based on the distribution of bare rock in the candidate landslide hazard areas, the rock edge movement rate of the candidate landslide hazard areas is determined; If the rock edge movement rate in the candidate landslide hazard area is greater than a preset rate, the candidate landslide hazard area is determined to be the landslide hazard area.

9. The method for monitoring landslides according to any one of claims 1-8, characterized in that, The method further includes: Based on the monitoring results, a landslide response strategy was determined. In accordance with the landslide response strategy, emergency measures were taken for the power grid equipment in the landslide hazard area.

Images