Method, device and equipment for rapidly evaluating risk of landslide induced by regional earthquake, storage medium and product

Abstract

The invention provides a regional earthquake-induced landslide risk rapid evaluation method and device, equipment, a storage medium and a product, relates to the field of geological disaster evaluation, and aims to solve the problems that an earthquake-induced landslide risk evaluation method in related technologies is difficult to realize rapid evaluation and is relatively small in application range. The method comprises the following steps: acquiring or predicting a distribution diagram of corresponding seismic intensity parameters and background condition data aiming at a landslide event induced by a sudden earthquake or a predicted landslide event induced by an earthquake; according to the distribution diagram of the seismic intensity parameters, seismic intensity levels of a plurality of areas included in the earthquake-induced landslide event are obtained through division; selecting a landslide risk evaluation model corresponding to the seismic intensity level according to the respective seismic intensity level and background condition data of the plurality of areas, and performing risk evaluation on each area to obtain landslide risk evaluation results of the plurality of areas included in the earthquake-induced landslide event; and displaying the landslide risk evaluation result.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of geological disaster assessment, and specifically, to a method, device, equipment, storage medium and product for rapidly assessing the risk of regional earthquake-induced landslides. Background Art

[0002] Earthquake-induced landslides are one of the secondary disasters of earthquakes and can cause serious harm to life and property in the landslide area. Due to the periodic law of earthquake tectonics and the periodicity of earthquake phenomena, there is a high possibility of earthquake-induced landslides occurring again in areas where earthquake-induced landslides have occurred before. Therefore, it is necessary to carry out the assessment of the risk of earthquake-induced landslides. However, the methods for assessing the risk of earthquake-induced landslides in related technologies need to collect ground motion after an earthquake and evaluate the sensitivity of factors affecting earthquake-induced landslides to carry out the assessment, making it difficult to achieve rapid assessment, and there is no general model for assessing the risk of earthquake-induced landslides, resulting in a small scope of application. Summary of the Invention

[0003] The embodiments of the present application aim to provide a method, device, equipment, storage medium and product for rapidly assessing the risk of regional earthquake-induced landslides, aiming to solve the problems that the methods for assessing the risk of earthquake-induced landslides in related technologies are difficult to achieve rapid assessment and have a small scope of application.

[0004] The first aspect of the embodiments of the present application provides a method for rapidly assessing the risk of regional earthquake-induced landslides, including:

[0005] For a sudden earthquake-induced landslide event or a predicted earthquake-induced landslide event, obtain or predict the distribution map of corresponding earthquake intensity parameters and the background condition data, and the distribution map of the earthquake intensity data at least includes: earthquake intensity distribution map, peak ground acceleration distribution map;

[0006] According to the distribution map of the earthquake intensity parameters, divide to obtain the earthquake intensity levels of each of the multiple regions included in the earthquake-induced landslide event;

[0007] According to the earthquake intensity levels of the multiple regions and the background condition data, select a landslide risk assessment model corresponding to the earthquake intensity level, and respectively conduct risk assessment on each region to obtain the landslide risk assessment results of the multiple regions included in the earthquake-induced landslide event;

[0008] Display the landslide risk assessment results.

[0009] In an optional implementation manner, the training steps of the landslide risk assessment model include:

[0010] Obtain the target data of multiple historical earthquake-induced landslide events, and the target data includes: earthquake intensity level data, landslide distribution data, background condition data;

[0011] Establish a database according to the target data;

[0012] Perform data processing and grading on the target data of multiple historical earthquake-induced landslide events in the database to obtain the background condition data and landslide distribution data corresponding to each earthquake intensity level;

[0013] Establish the earthquake-induced landslide hazard assessment model corresponding to each earthquake intensity level according to the background condition data and landslide distribution data corresponding to each earthquake intensity level.

[0014] In an optional implementation manner, performing data processing and grading on the target data of multiple historical earthquake-induced landslide events in the database to obtain the background condition data and landslide distribution data corresponding to each earthquake intensity level includes:

[0015] Vectorize the target data in the database to obtain vector data;

[0016] Convert the vector data into raster data, and through coordinate system conversion and projection conversion, obtain a unified projection coordinate system;

[0017] Divide each earthquake-induced landslide event to obtain the areas included in multiple earthquake-induced landslide events corresponding to each earthquake intensity level;

[0018] Based on the projection coordinate system and the areas included in multiple earthquake-induced landslide events corresponding to each earthquake intensity level, determine the spatial positions corresponding to each earthquake intensity level, and according to the spatial positions, obtain the corresponding spatial ranges;

[0019] Based on the projection coordinate system, obtain the background condition data and landslide distribution data corresponding to each earthquake intensity level according to the spatial range corresponding to each earthquake intensity level.

[0020] In an optional implementation manner, dividing each earthquake-induced landslide event to obtain the areas included in multiple earthquake-induced landslide events corresponding to each earthquake intensity level includes:

[0021] According to the division standard of earthquake intensity in the earthquake intensity level data and the conversion formula between seismic acceleration and earthquake intensity, obtain the grading standard for earthquake-induced landslide events;

[0022] According to the grading standard, divide each earthquake-induced landslide event to obtain the earthquake intensity levels of the multiple areas included in each earthquake-induced landslide event;

[0023] Based on the earthquake intensity levels of each of the multiple regions included in each earthquake-induced landslide event, the regions included in multiple earthquake-induced landslide events corresponding to each earthquake intensity level are obtained.

[0024] In an alternative embodiment, after obtaining the background condition data and landslide distribution data corresponding to each earthquake intensity level, the method further includes:

[0025] The background condition data corresponding to each earthquake intensity level is classified and divided to obtain the classified background condition data.

[0026] In an alternative embodiment, according to the background condition data and landslide distribution data corresponding to each earthquake intensity level, establishing the earthquake-induced landslide hazard assessment model corresponding to each earthquake intensity level includes:

[0027] Performing multiple collinearity analysis and importance analysis on the landslide distribution data and background condition data in the database to obtain the target landslide influencing factors;

[0028] According to the target landslide influencing factors, the classified background condition data and landslide distribution data corresponding to each earthquake intensity level are selected to obtain the target background condition data and target landslide distribution data corresponding to each earthquake intensity level.

[0029] For each earthquake intensity level in the database, with the corresponding target background condition data as the input and the target landslide distribution data as the output, performing hazard assessment through multiple models to obtain the AUC index of the ROC curve of the evaluation results of multiple models;

[0030] For each earthquake intensity level, the model corresponding to the maximum AUC value is selected as the landslide hazard assessment model for that earthquake intensity level.

[0031] The second aspect of the embodiments of the present application provides a rapid regional earthquake-induced landslide hazard assessment device, including:

[0032] An acquisition module, for a sudden earthquake-induced landslide event or a predicted earthquake-induced landslide event, acquires or predicts the distribution map of the corresponding earthquake intensity parameters and the background condition data, and the distribution map of the earthquake intensity data at least includes: the earthquake intensity distribution map, the earthquake peak acceleration distribution map;

[0033] A division module, configured to divide, according to the distribution map of the earthquake intensity parameters, the earthquake intensity levels of each of the multiple regions included in the earthquake-induced landslide event.

[0034] An evaluation module, configured to select a landslide hazard evaluation model corresponding to the earthquake intensity level according to the earthquake intensity levels and background condition data of multiple regions, respectively perform hazard evaluations on each region, and obtain the landslide hazard evaluation results of multiple regions included in the earthquake-induced landslide event;

[0035] A display module, configured to display the landslide hazard evaluation results.

[0036] A third aspect of the embodiments of the present application provides an electronic device, including: a processor, a memory, and a computer program stored on the memory and executable on the processor, where when the processor executes the computer program, the rapid evaluation method for regional earthquake-induced landslide hazards in the first aspect of the embodiments of the present application is implemented.

[0037] A fourth aspect of the embodiments of the present application provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the rapid evaluation method for regional earthquake-induced landslide hazards in the first aspect of the embodiments of the present application is implemented.

[0038] A fifth aspect of the embodiments of the present application provides a computer program product, including a computer program, and when the computer program is executed by a processor, the rapid evaluation method for regional earthquake-induced landslide hazards in the first aspect of the embodiments of the present application is implemented.

[0039] In this embodiment, by obtaining a distribution map of earthquake intensity parameters and background condition data corresponding to an earthquake-induced landslide event; dividing the earthquake intensity levels of multiple regions included in the earthquake-induced landslide event according to the distribution map of the earthquake intensity parameters; selecting a landslide hazard evaluation model corresponding to the earthquake intensity level according to the earthquake intensity levels and background condition data of multiple regions, respectively performing hazard evaluations on each region, and obtaining the landslide hazard evaluation results of multiple regions included in the earthquake-induced landslide event; displaying the landslide hazard evaluation results. By dividing the earthquake intensity levels of multiple regions of the earthquake-induced landslide event, the uneven distribution of earthquake intensity in different regions is considered, which helps to improve the accuracy and reliability of landslide hazard evaluation; according to different earthquake intensity levels and background condition data, selecting corresponding evaluation models to perform hazard evaluations, there is no need to collect ground motion data again, evaluate the sensitivity of earthquake landslide influencing factors, etc., and rapid and accurate earthquake landslide hazard evaluation can be achieved, and different models can be applied in different regions, with a wide range of applications. Description of the Drawings

[0040] To more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments of the present application. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0041] Figure 1 is a flowchart of the steps of a method for rapid assessment of regional earthquake-induced landslide hazards proposed in an embodiment of the present application;

[0042] Figure 2 is a schematic diagram of the model evaluation results of earthquake-induced landslide events in a method for rapid assessment of regional earthquake-induced landslide hazards proposed in an embodiment of the present application;

[0043] Figure 3 is a schematic diagram of some earthquake-induced landslide attribute data in the database of a method for rapid assessment of regional earthquake-induced landslide hazards proposed in an embodiment of the present application;

[0044] Figure 4 is a schematic diagram of the importance of landslide influencing factors in a method for rapid assessment of regional earthquake-induced landslide hazards proposed in an embodiment of the present application;

[0045] Figure 5 is a schematic diagram of the comparison of the ROC curves and AUC values of the deep learning model and the random forest model for a method for rapid assessment of regional earthquake-induced landslide hazards proposed in an embodiment of the present application;

[0046] Figure 6 is a schematic diagram of an electronic device proposed in an embodiment of the present application. Detailed implementation manners

[0047] The following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some, but not all, of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts belong to the scope of protection of the present application.

[0048] In the drawings, sometimes for clarity, the sizes of the components, the thickness of the layers, or the areas may be exaggerated. Therefore, any implementation of the present disclosure is not necessarily limited to the sizes shown in the figures, and the shapes and sizes of the components in the figures do not reflect the true proportions. In addition, the drawings schematically show ideal examples, and any implementation of the present disclosure is not limited to the shapes or values shown in the drawings.

[0049] Earthquake-induced landslides are landslide phenomena caused by the propagation of seismic waves on the Earth's surface, which cause deformation and damage to surface slopes. After an earthquake landslide occurs, it will cause disasters to life and property in the sliding area, and it is one of the most important secondary disasters of earthquakes. Due to the periodic law of earthquake tectonics and the periodicity of earthquake phenomena, there is a high possibility of earthquake landslides occurring again in areas where earthquake landslides have occurred. Therefore, it is necessary to carry out earthquake landslide hazard assessment. Because earthquakes have the characteristics of short duration and great harm, a large number of earthquake-induced landslides occur in a short time. Therefore, a rapid assessment of earthquake-induced landslide hazards is urgently needed.

[0050] The goal of regional earthquake landslide hazard assessment is to predict the probability, location, and intensity of landslides occurring in a certain area. The primary inducing factor for regional earthquake landslides is the intensity of ground motion. After the action of earthquake intensity on the background conditions such as the geological structure, topography, and hydrogeology of the slope, it causes deformation and damage to the slope, and then triggers landslides. The response relationship between the background conditions and different earthquake intensities is different.

[0051] Regional earthquake landslide disaster risk assessment models are mainly divided into three types: qualitative assessment methods, deterministic models, and statistical analysis models. Qualitative assessment methods mainly include topographic and geomorphic analysis methods and qualitative zoning map overlay methods; however, this method is highly dependent on the knowledge accumulation and cognition of experts themselves and is greatly affected by subjectivity. Common algorithms for deterministic models include the Newmark model. Such models require detailed spatial variable parameters and are very effective and require less workload for individual landslides; however, due to the complexity, uncertainty of the landslide disaster mechanism, and the limitations of human cognitive level, it is very difficult to carry out evaluation work. Among statistical analysis models, the most common ones include information quantity models, weight of evidence models, logistic regression models, artificial intelligence models, fuzzy mathematics models, analytic hierarchy process models, etc. These models are all data-driven and obtain research results through a large amount of data calculation, which is relatively objective and reliable. And the study of earthquake landslide hazards through collecting historical data has a certain overall perspective. However, these models also have the drawbacks that due to different selected earthquake cases and the large span of the regions where earthquakes occur, it is very difficult to collect relatively complete data, and the data coverage is relatively limited.

[0052] The regional earthquake landslide hazard assessment methods in related technologies mainly have the following limitations: the mutual sensitivity between the spatial distribution of earthquake intensity and its background conditions is not considered in regional earthquake landslide hazards; and it is difficult to carry out rapid evaluation. It is necessary to collect ground motion after an earthquake and evaluate the sensitivity of earthquake landslide influencing factors to carry out the evaluation, and a general earthquake landslide hazard assessment model and its parameters cannot be obtained.

[0053] In view of the above problems, the embodiments of the present application propose a method, device, equipment, storage medium and product for rapid evaluation of the risk of regional earthquake-induced landslides, so as to achieve rapid and accurate evaluation of earthquake landslide risk. The following will combine the accompanying drawings to detail a method, device, equipment, storage medium and product for rapid evaluation of the risk of regional earthquake-induced landslides provided by the embodiments of the present application through some embodiments and their application scenarios.

[0054] Referring to Figure 1 , Figure 1 is the step flow chart of the method for rapid evaluation of the risk of regional earthquake-induced landslides proposed in Embodiment 1 of the present application. As Figure 1 shown, the method includes the following steps S11 to S14:

[0055] Step S11: For an emergency earthquake-induced landslide event or a predicted earthquake-induced landslide event, obtain or predict the distribution map of the corresponding earthquake intensity parameters and the background condition data. The distribution map of the earthquake intensity data at least includes: the earthquake intensity distribution map and the earthquake peak acceleration distribution map.

[0056] For an emergency earthquake-induced landslide event or a predicted earthquake-induced landslide event, the distribution map of the earthquake intensity parameters (such as the earthquake intensity distribution map and the earthquake peak acceleration distribution map) and the background condition data corresponding to the earthquake-induced landslide event can be obtained or predicted through a seismic monitoring agency. The earthquake intensity distribution map shows the degree of damage caused by the earthquake to different regions, usually represented by colors or numbers for different intensity levels; the earthquake peak acceleration distribution map describes the distribution of the maximum absolute value of the ground acceleration during the earthquake process, usually represented by different colors or contour lines for regions with different peak acceleration values; the background condition data refers to the factors that can affect the slope stability, such as slope, aspect, elevation, terrain undulation, lithology, etc.

[0057] Step S12: According to the distribution map of the earthquake intensity parameters, divide the earthquake intensity levels of the multiple regions included in the earthquake-induced landslide event.

[0058] According to the obtained distribution map of the earthquake intensity parameters (such as the earthquake intensity distribution map and the earthquake peak acceleration distribution map), the multiple regions included in the earthquake-induced landslide event are uniformly divided. The division can refer to the earthquake intensity from 1 to 10 degrees. The earthquake peak acceleration can be correspondingly divided after being converted according to the earthquake intensity conversion formula. The earthquake intensity level and the magnitude of the earthquake peak acceleration are in one-to-one correspondence. Furthermore, the earthquake intensity levels corresponding to the multiple regions included in the earthquake-induced landslide event are obtained. For example, the earthquake intensity level n corresponding to the earthquake-induced landslide event is divided into: n1, n2, n3... n 10 .

[0059] Step S13: According to the seismic intensity levels and background condition data of multiple regions, select landslide hazard assessment models corresponding to the seismic intensity levels, and conduct hazard assessments on each region respectively to obtain the landslide hazard assessment results of multiple regions included in the earthquake-induced landslide event;

[0060] According to the seismic intensity levels and background condition data of multiple regions, select landslide hazard assessment models corresponding to each seismic intensity level to conduct hazard assessments on each region. Input the background condition data corresponding to the seismic intensity level into the corresponding landslide hazard assessment model, and output the predicted landslide distribution data (regions where landslides may occur, scales, probabilities, etc.); conduct hazard assessments on each region respectively, and finally obtain the landslide hazard assessment results of multiple regions included in the earthquake-induced landslide event. These results are usually expressed in forms such as probabilities and grades.

[0061] Step S14: Display the landslide hazard assessment results.

[0062] Present the landslide hazard assessment results in an intuitive way so that managers can clearly understand the risk situation of earthquake-induced landslides, thereby formulating effective prevention and response measures. Geographic information system software can be used to make landslide hazard assessment maps, and regions with different hazard levels are represented by different colors or symbols, as Figure 2 shown.

[0063] In this embodiment, by obtaining the distribution map of seismic intensity parameters and background condition data corresponding to the earthquake-induced landslide event; according to the distribution map of the seismic intensity parameters, divide the seismic intensity levels of multiple regions included in the earthquake-induced landslide event; according to the seismic intensity levels and background condition data of multiple regions, select landslide hazard assessment models corresponding to the seismic intensity levels, and conduct hazard assessments on each region respectively to obtain the landslide hazard assessment results of multiple regions included in the earthquake-induced landslide event; display the landslide hazard assessment results. By dividing the seismic intensity levels of multiple regions of the earthquake-induced landslide event, the uneven distribution of seismic intensity in different regions is considered, which helps to improve the accuracy and reliability of landslide hazard assessment; according to different seismic intensity levels and background condition data, select corresponding evaluation models to conduct hazard assessments, without the need to re-collect ground motion data and evaluate the sensitivity of earthquake landslide influencing factors, etc., which can achieve rapid and accurate earthquake landslide hazard assessment, and different models can be applied in different regions, with a wide range of applications.

[0064] In an alternative embodiment, the training steps of the landslide hazard assessment model include:

[0065] Step S101: Obtain the target data of multiple historical earthquake-induced landslide events. The target data includes: earthquake intensity level data, landslide distribution data, and background condition data.

[0066] Through earthquake monitoring agencies, geological survey departments, remote sensing data centers, research institutions, etc., obtain the target data of multiple historical earthquake-induced landslide events. The target data includes at least: earthquake intensity level data, landslide distribution data, and background condition data.

[0067] The earthquake intensity level data refers to the ground motion data that describes earthquake activities and their influence intensities, usually including the depth of the seismogenic fault, magnitude, spatial distribution and energy distribution, post-seismic peak acceleration records, earthquake intensity isoseismic maps, earthquake peak acceleration isoseismic maps, etc.

[0068] The induced landslide distribution data refers to the spatial distribution and characteristics of each earthquake-induced landslide event, usually including the number, scale, depth, slip surface characteristics, spatial location, landslide boundary characteristics, etc. of the landslides.

[0069] The background condition data refers to the factors that can affect slope stability, usually including hydrogeological conditions, NDVI (vegetation index), lithology map, DEM (digital elevation model), landslide slope height, slope angle, slope surface morphology, slope surface curvature, fault distance, fault direction, river distance, stratigraphic lithology, lithology water-rich degree, etc. within the influence area of each earthquake-induced landslide event, as shown in Table 1. The factors included in the background condition M can be respectively represented as 1, 2, 3... m.

[0070] Table 1. Background conditions

[0071] Serial number Data Serial number Data 1 Elevation 9 Soil moisture 2 Slope 10 Slope structure 3 Aspect 11 Fault distribution 4 Terrain undulation 12 Road distribution 5 Water system distribution 13 Normalized difference vegetation index (NDVI) 6 Land use map 14 Rainfall distribution 7 Surface roughness 15 Lithology distribution 8 Slope curvature 16 Vegetation type map

[0072] Step S102: Establish a database based on the target data.

[0073] Based on the collected target data of multiple earthquake-induced landslide events, establish a safe, reliable, and easily accessible database (such as MySQL, PostgreSQL). Some earthquake-induced landslide attribute data in the database is as Figure 3 shown. Provide comprehensive and accurate data support for subsequent model establishment.

[0074] Step S103: Perform data processing and grading on the target data of multiple historical earthquake-induced landslide events in the database to obtain the background condition data and landslide distribution data corresponding to each earthquake intensity level.

[0075] For all target data in the database, data processing is performed, and the processed data is transformed into a unified projected coordinate system through coordinate system transformation and projection transformation to facilitate subsequent spatial analysis and processing. Multiple historical earthquake-induced landslide events can be divided into different earthquake intensity levels according to the earthquake intensity levels in the target data. Each earthquake intensity level involves multiple historical earthquake-induced landslide events. Based on the projected coordinate system, for each earthquake intensity level, the background condition data and landslide distribution data within the spatial range of all earthquake-induced landslide events at that earthquake intensity level are extracted, so as to obtain the background condition data and landslide distribution data corresponding to each earthquake intensity level. This helps to analyze the distribution characteristics of landslides at different earthquake intensity levels to establish an earthquake-induced landslide hazard assessment model corresponding to each earthquake intensity level.

[0076] Step S104: Based on the background condition data and landslide distribution data corresponding to each earthquake intensity level, establish the earthquake-induced landslide hazard assessment model corresponding to each earthquake intensity level.

[0077] For each earthquake intensity level, using its background condition data as input and landslide distribution data as output, landslide hazard assessment is carried out through multiple models (such as information value model, weight of evidence model, logistic regression model, artificial intelligence model, fuzzy mathematics model, analytic hierarchy process model, random forest model, etc.). By comparing the performance of different models, the optimal model corresponding to each earthquake intensity level is selected as the earthquake-induced landslide hazard assessment model. Considering the mutual sensitivity between the spatial distribution of earthquake intensity and its background conditions, and establishing corresponding models for each earthquake intensity level can improve the accuracy and reliability of landslide hazard assessment, and different models can be applied in different regions, with a wide range of applications.

[0078] In an alternative embodiment, data processing and grading are performed on the target data of multiple historical earthquake-induced landslide events in the database to obtain the background condition data and landslide distribution data corresponding to each earthquake intensity level, including:

[0079] Vectorize the target data in the database to obtain vector data;

[0080] Convert the vector data into raster data, and through coordinate system transformation and projection transformation, obtain a unified projected coordinate system;

[0081] Divide each earthquake-induced landslide event to obtain the areas included in multiple earthquake-induced landslide events corresponding to each earthquake intensity level;

[0082] Based on the projection coordinate system and the areas included in multiple earthquake-induced landslide events corresponding to each earthquake intensity level, determine the spatial positions corresponding to each earthquake intensity level, and obtain the corresponding spatial ranges according to the spatial positions;

[0083] Based on the projection coordinate system, according to the spatial ranges corresponding to each earthquake intensity level, obtain the background condition data and landslide distribution data corresponding to each earthquake intensity level.

[0084] Vectorize all target data in the database, convert the generated vector data into raster data, and perform coordinate system transformation and projection transformation on the raster data to obtain a unified projection coordinate system. The earthquake-induced landslide events can be classified according to the earthquake intensity level in the target data to obtain the areas included in multiple earthquake-induced landslide events corresponding to each earthquake intensity level. Based on the projection coordinate system and the areas included in multiple earthquake-induced landslide events corresponding to each earthquake intensity level, the spatial positions corresponding to each earthquake intensity level can be determined, and the corresponding spatial range information can be extracted according to the spatial positions of each earthquake intensity level. By spatially corresponding the background conditions and landslide distribution data within the spatial range corresponding to each earthquake intensity level, the background condition data and landslide distribution data corresponding to each earthquake intensity level can be obtained.

[0085] In an alternative embodiment, dividing each earthquake-induced landslide event to obtain the areas included in multiple earthquake-induced landslide events corresponding to each earthquake intensity level includes:

[0086] According to the division standard of earthquake intensity in the earthquake intensity level data and the conversion formula between earthquake acceleration and earthquake intensity, obtain the division standard for earthquake-induced landslide events;

[0087] According to the division standard, divide each earthquake-induced landslide event to obtain the earthquake intensity level of each of the multiple areas included in each earthquake-induced landslide event;

[0088] According to the earthquake intensity levels of each of the multiple areas included in each earthquake-induced landslide event, obtain the areas included in multiple earthquake-induced landslide events corresponding to each earthquake intensity level.

[0089] Uniformly divide each earthquake-induced landslide event (corresponding to events N = 1, 2, 3... n events respectively) according to the earthquake intensity level, the peak ground acceleration, etc. The division can refer to earthquake intensities from 1 to 10 degrees. The peak ground acceleration can be correspondingly divided after being converted according to the earthquake intensity conversion formula. The earthquake intensity level and the peak ground acceleration are in one-to-one correspondence. Thus, the earthquake intensity levels of each of the multiple areas included in each earthquake-induced landslide event are n: n1, n2, n3... n 10For each earthquake intensity level, based on the earthquake intensity levels of the respective multiple regions included in each earthquake-induced landslide event, the regions included in multiple earthquake-induced landslide events corresponding to each earthquake intensity level can be obtained.

[0090] In an alternative implementation, after obtaining the background condition data and landslide distribution data corresponding to each earthquake intensity level, the method further includes:

[0091] Classify and divide the background condition data corresponding to each earthquake intensity level to obtain the classified background condition data.

[0092] According to relevant literature and standards, classify and divide the background condition data M of earthquake intensity level n to obtain the classified background condition data. For example, data such as slope, elevation, terrain undulation, distance from water system, etc. are divided according to Table 2 below, and lithology classification and grading are divided according to Table 3 below, etc. The data after classification and division is more conducive to correlation analysis with the landslide distribution data, and helps to construct and optimize the landslide hazard assessment model.

[0093] Table 2. Classification of background condition levels

[0094] Division level Elevation / m Slope / ° Terrain undulation / m Distance to water system / m 1 720 0 320 0 2 818 3.3 408 104 3 916 6.6 496 208 4 1014 9.9 584 312 5 1112 13.2 672 416 6 1210 16.5 760 520 7 1308 19.8 848 624 8 1406 23.1 936 728 9 1504 26.4 1024 832 10 1602 29.7 1112 936 11 1700 33 1200 1040 12 1798 36.3 1288 1144 13 1896 39.6 1376 1248 14 1994 42.9 1464 1352 15 2092 46.2 1552 1456 16 2190 49.5 1640 1560 17 2288 52.8 1728 1664 18 2386 56.1 1816 1768 19 2484 59.4 1904 1872 20 2582 62.7 1992 1976 21 2680 66 2080 2080 22 2778 69.3 2168 2184 23 2876 72.6 2256 2288 24 2974 75.9 2344 2392 25 3072 79.2 2432 2496

[0095] Table 3. Lithology classification and grading

[0096]

[0097]

[0098] In an alternative implementation, based on the background condition data and landslide distribution data corresponding to each earthquake intensity level, establish the earthquake-induced landslide hazard assessment model corresponding to each earthquake intensity level, including:

[0099] Perform multicollinearity analysis and importance analysis on the landslide distribution data and background condition data in the database to obtain the target landslide influencing factors;

[0100] Select the classified background condition data and landslide distribution data corresponding to each earthquake intensity level according to the target landslide influencing factors to obtain the target background condition data and target landslide distribution data corresponding to each earthquake intensity level;

[0101] For each earthquake intensity level in the database, using the corresponding target background condition data as the input and the target landslide distribution data as the output, perform hazard assessment through multiple models to obtain the AUC index of the ROC curve of the evaluation results of multiple models;

[0102] For each earthquake intensity level, select the model corresponding to the maximum AUC value as the landslide hazard assessment model for that earthquake intensity level.

[0103] Based on the landslide distribution data and background condition data in the database, use data analysis methods to conduct multicollinearity analysis and importance evaluation of various factors. To detect the multicollinearity of the selected landslide influencing factors, common test indicators include the variance inflation factor (VIF) and tolerance (TOL), which are reciprocals of each other, and the calculation formulas are shown in (1) and (2):

[0104] T i =1 - R i 2 (1)

[0105]

[0106] Where: T i is the tolerance value; VIF i is the variance inflation factor; is the determination coefficient between the i-th influencing factor and the remaining influencing factors. When the correlation degree between the i-th influencing factor and the remaining influencing factors is higher, tends to be close to 1; the larger the value of the variance inflation factor VIF i is, the greater the possibility of collinearity between influencing factors. Generally, if the tolerance is less than 0.1 or the variance inflation factor exceeds 10, the regression model has serious multicollinearity. The calculation results of the tolerance values and variance inflation factors of different factors are shown in Table 4.

[0107] Table 4. Tolerance and variance inflation factors of different factors

[0108] Landslide hazard factors Tolerance value Variance inflation factor (VIF) Slope 0.1264 7.91 Aspect 0.3806 2.6276 Elevation 0.2413 4.1438 Surface undulation 0.3341 2.9934 Lithology 0.1092 9.1557 Fault distance 0.1162 8.603 Land use 0.1956 5.1037 Vegetation index 0.4961 2.0156 River distance 0.4455 2.2445 Road distance 0.1164 8.593

[0109] After conducting multicollinearity analysis on the factors, it is also necessary to conduct importance analysis on multiple factors, and eliminate the factors with less effect on network division according to the scores. GI (Gini Gain) is used to quantify the contribution of factors to landslide occurrence. It is defined as the average of the total reduction of node impurities of all trees. The higher the GI value, the greater the contribution of the factor to the model prediction result. The calculation of GI is based on Gini impurity, given by Equation (3):

[0110]

[0111] Where, N is the number of classes, and p i is the proportion of class i. The calculated GI values can be used, such as Figure 4As shown, sort each factor and set a threshold according to the sorting result to screen important factors.

[0112] By performing multiple collinearity analysis and importance analysis on the landslide distribution data and background condition data (multiple factors) in the database, determine the background condition data that has a significant impact on the landslide distribution, that is, the target landslide influencing factors, which provide key inputs for subsequent model training to improve the stability and accuracy of the established model.

[0113] According to the target landslide influencing factors, screen out the relevant data from the classified background condition data corresponding to each earthquake intensity level as the target background condition data; at the same time, screen out the landslide data corresponding to the earthquake intensity level from the landslide distribution data as the target landslide distribution data.

[0114] For a single earthquake intensity level (any one of n1, n2, n3... n 10 in the database), using its target background condition data as input and the target landslide distribution data as output, conduct landslide hazard assessment through various models (such as information amount model, weight of evidence model, logistic regression model, artificial intelligence model, fuzzy mathematics model, analytic hierarchy process model, random forest model, etc.).

[0115] Calculate the AUC index of the ROC curve of the evaluation results of various models. The AUC value is a quantification of the overall performance of the model, and the model corresponding to the largest AUC value is the optimal evaluation model corresponding to the earthquake intensity level. As Figure 5 shown, in a certain earthquake intensity level, the comparison of the ROC curve and AUC value between the deep learning model (CNN) and the random forest model (RF). The AUC value of the deep learning model is 0.8696145742, and the AUC value of the random forest model is 0.8628145606, so the deep learning model has higher accuracy.

[0116] Based on the same inventive concept, the embodiment of the present application discloses a device for rapid assessment of regional earthquake-induced landslide hazards, including:

[0117] An acquisition module, for an earthquake-induced landslide event caused by a sudden earthquake or a predicted earthquake-induced landslide event, acquires or predicts the distribution map of the corresponding earthquake intensity parameters and the background condition data. The distribution map of the earthquake intensity data at least includes: earthquake intensity distribution map, earthquake peak acceleration distribution map;

[0118] A division module, used to divide the earthquake intensity levels of each of the multiple regions included in the earthquake-induced landslide event according to the distribution map of the earthquake intensity parameters;

[0119] An evaluation module, configured to select a landslide hazard evaluation model corresponding to the earthquake intensity level according to the earthquake intensity levels and background condition data of multiple regions, and perform hazard evaluations on each region respectively to obtain the landslide hazard evaluation results of multiple regions included in the earthquake-induced landslide event;

[0120] A display module, configured to display the landslide hazard evaluation results.

[0121] In an optional implementation manner, the evaluation module is specifically configured to:

[0122] Obtain the target data of multiple historical earthquake-induced landslide events, where the target data includes: earthquake intensity level data, landslide distribution data, and background condition data;

[0123] Establish a database according to the target data;

[0124] Perform data processing and grading on the target data of multiple historical earthquake-induced landslide events in the database to obtain the background condition data and landslide distribution data corresponding to each earthquake intensity level;

[0125] Establish the earthquake-induced landslide hazard evaluation model corresponding to each earthquake intensity level according to the background condition data and landslide distribution data corresponding to each earthquake intensity level.

[0126] In an optional implementation manner, the evaluation module is specifically configured to:

[0127] Vectorize the target data in the database to obtain vector data;

[0128] Convert the vector data into raster data, and through coordinate system conversion and projection conversion, obtain a unified projection coordinate system;

[0129] Divide each earthquake-induced landslide event to obtain the regions included in multiple earthquake-induced landslide events corresponding to each earthquake intensity level;

[0130] Based on the projection coordinate system and the regions included in multiple earthquake-induced landslide events corresponding to each earthquake intensity level, determine the spatial positions corresponding to each earthquake intensity level, and according to the spatial positions, obtain the corresponding spatial ranges;

[0131] Based on the projection coordinate system, according to the spatial ranges corresponding to each earthquake intensity level, obtain the background condition data and landslide distribution data corresponding to each earthquake intensity level.

[0132] In an optional implementation manner, the evaluation module is specifically configured to:

[0133] According to the division standard of seismic intensity in the seismic intensity level data and the conversion formula between seismic acceleration and seismic intensity, the grading standard for earthquake-induced landslide events is obtained;

[0134] According to the grading standard, each earthquake-induced landslide event is divided to obtain the seismic intensity level of each of the multiple regions included in each earthquake-induced landslide event;

[0135] According to the seismic intensity levels of the multiple regions included in each earthquake-induced landslide event, the regions included in multiple earthquake-induced landslide events corresponding to each seismic intensity level are obtained.

[0136] In an alternative embodiment, the evaluation module is specifically configured to:

[0137] Perform grading on the background condition data corresponding to each seismic intensity level to obtain the graded background condition data.

[0138] In an alternative embodiment, the processing module is specifically configured to:

[0139] Perform multicollinearity analysis and importance analysis on the landslide distribution data and background condition data in the database to obtain the target landslide influencing factors;

[0140] According to the target landslide influencing factors, select the graded background condition data and landslide distribution data corresponding to each seismic intensity level to obtain the target background condition data and target landslide distribution data corresponding to each seismic intensity level;

[0141] For each seismic intensity level in the database, using the corresponding target background condition data as the input and the target landslide distribution data as the output, perform hazard assessment through multiple models to obtain the AUC index of the ROC curve of the evaluation results of multiple models;

[0142] For each seismic intensity level, select the model corresponding to the maximum AUC value as the landslide hazard assessment model for that seismic intensity level.

[0143] Based on the same inventive concept, another embodiment of the present application provides an electronic device. Refer to Figure 6 , Figure 6 is a schematic diagram of an electronic device shown in an embodiment of the present application. As Figure 6 shown, the electronic device 100 includes: a memory 110 and a processor 120. The memory 110 is communicatively connected to the processor 120 through a bus. A computer program is stored in the memory 110, and the computer program can run on the processor 120 to implement the steps in the method for rapid assessment of regional earthquake-induced landslide hazards disclosed in the embodiments of the present disclosure.

[0144] Based on the same inventive concept, embodiments of the present disclosure also provide a computer-readable storage medium. When instructions in the computer-readable storage medium are executed by a processor of a computer device, the computer device is enabled to execute the steps in the method for rapid assessment of regional earthquake-induced landslide hazards described in any of the above embodiments of the present application.

[0145] Based on the same inventive concept, embodiments of the present disclosure also provide a computer program product, including a computer program. When the computer program is executed by a processor of a computer device, it is capable of executing the steps in the method for rapid assessment of regional earthquake-induced landslide hazards disclosed in the embodiments of the present disclosure.

[0146] Each embodiment in this specification is described in a progressive manner. The key point of each embodiment is to illustrate the differences from other embodiments. For the same or similar parts among the embodiments, reference can be made to each other.

[0147] Embodiments of the present application are described with reference to the flowcharts and / or block diagrams of methods, devices, electronic devices, and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal devices generate a device for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0148] These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing terminal devices to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured product including an instruction device, and the instruction device implements the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0149] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal devices, so that a series of operation steps are executed on the computer or other programmable terminal devices to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable terminal devices provide steps for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0150] Although the preferred embodiments of the embodiments of the present invention have been described, those skilled in the art can make additional changes and modifications to these embodiments once they learn the basic creative concept. Therefore, the appended claims are intended to be construed as including the preferred embodiments and all changes and modifications falling within the scope of the embodiments of the present invention.

[0151] Finally, it should also be noted that in this text, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or terminal device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or terminal device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the existence of additional identical elements in the process, method, article or terminal device comprising the element.

[0152] The above has introduced in detail a method, device, equipment, storage medium and product for rapid assessment of the risk of regional earthquake-induced landslides provided by this application. Specific examples are used in this text to elaborate on the principle and implementation manner of this application. The description of the above embodiments is only used to help understand the method and its core idea of this application; at the same time, for those of ordinary skill in the art, based on the idea of this application, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to this application.

Claims

1. A method for rapid assessment of regional earthquake-induced landslide hazard, characterized in that: include: For a sudden earthquake-induced landslide event or a predicted earthquake-induced landslide event, a distribution map of corresponding earthquake intensity parameters and background condition data are obtained or predicted, wherein the distribution map of the earthquake intensity data at least includes: a distribution map of earthquake intensity and a distribution map of earthquake peak acceleration; According to the distribution diagram of the earthquake intensity parameter, the earthquake intensity levels of the multiple regions included in the earthquake-induced landslide event are divided and obtained; According to the earthquake intensity levels and background condition data of the multiple regions, a landslide hazard assessment model corresponding to the earthquake intensity level is selected, and the hazard assessment of each region is performed respectively to obtain the landslide hazard assessment results of the multiple regions included in the earthquake-induced landslide event; Display the landslide hazard assessment results.

2. The method for rapid assessment of regional earthquake-induced landslide hazard according to claim 1, characterized in that: The training steps of the landslide hazard assessment model include: Acquire target data of multiple historical earthquake-induced landslide events, the target data including: earthquake intensity level data, landslide distribution data, and background condition data; Establishing a database according to the target data; Processing and grading the target data of multiple historical earthquake-induced landslide events in the database to obtain background condition data and landslide distribution data corresponding to each earthquake intensity level; According to the background condition data and landslide distribution data corresponding to each earthquake intensity level, the earthquake-induced landslide hazard assessment model corresponding to each earthquake intensity level is established.

3. The method for rapid assessment of regional earthquake-induced landslide hazard according to claim 2, characterized in that: The target data of multiple historical earthquake-induced landslide events in the database are processed and classified to obtain background condition data and landslide distribution data corresponding to each earthquake intensity level, including: Vectorizing the target data in the database to obtain vector data; Convert the vector data into raster data, and obtain a unified projection coordinate system through coordinate system conversion and projection conversion; Divide each earthquake-induced landslide event to obtain the area covered by multiple earthquake-induced landslide events corresponding to each earthquake intensity level; Based on the projection coordinate system and the area covered by multiple earthquake-induced landslide events corresponding to each earthquake intensity level, determine the spatial position corresponding to each earthquake intensity level, and obtain the corresponding spatial range according to the spatial position; Based on the projection coordinate system, the background condition data and landslide distribution data corresponding to each earthquake intensity level are obtained according to the spatial range corresponding to each earthquake intensity level.

4. The method for rapid assessment of regional earthquake-induced landslide hazard according to claim 3, characterized in that: Each earthquake-induced landslide event is divided to obtain the areas covered by multiple earthquake-induced landslide events corresponding to each earthquake intensity level, including: According to the classification standard of earthquake intensity in earthquake intensity level data and the conversion formula between earthquake acceleration and earthquake intensity, the classification standard of earthquake-induced landslide events is obtained; According to the classification standard, each earthquake-induced landslide event is divided to obtain the seismic intensity levels of the multiple regions included in each earthquake-induced landslide event; According to the respective earthquake intensity levels of the multiple regions included in each earthquake-induced landslide event, the regions included in the multiple earthquake-induced landslide events corresponding to each earthquake intensity level are obtained.

5. The method for rapid assessment of regional earthquake-induced landslide hazard according to claim 3, characterized in that: After obtaining the background condition data and landslide distribution data corresponding to each earthquake intensity level, the method further includes: The background condition data corresponding to each earthquake intensity level are classified and divided to obtain the classified background condition data.

6. The method for rapid assessment of regional earthquake-induced landslide hazard according to claim 5, characterized in that: According to the background condition data and landslide distribution data corresponding to each earthquake intensity level, the earthquake-induced landslide hazard assessment model corresponding to each earthquake intensity level is established, including: Multicollinearity analysis and importance analysis were performed on the landslide distribution data and background condition data in the database to obtain the target landslide influencing factors; According to the target landslide impact factor, the classified background condition data and landslide distribution data corresponding to each earthquake intensity level are selected to obtain the target background condition data and target landslide distribution data corresponding to each earthquake intensity level; For each earthquake intensity level in the database, the corresponding target background condition data is used as input and the target landslide distribution data is used as output. The hazard evaluation is performed through multiple models to obtain the AUC index of the ROC curve of the evaluation results of multiple models. For each earthquake intensity level, the model corresponding to the largest AUC value is selected as the landslide hazard assessment model for that earthquake intensity level.

7. A rapid assessment device for regional earthquake-induced landslide hazard, characterized in that: include: An acquisition module, for a sudden earthquake-induced landslide event or a predicted earthquake-induced landslide event, acquires or predicts a distribution map of corresponding earthquake intensity parameters and background condition data, wherein the distribution map of the earthquake intensity data at least includes: a distribution map of earthquake intensity and a distribution map of earthquake peak acceleration; A division module, used for dividing and obtaining the respective earthquake intensity levels of a plurality of regions included in the earthquake-induced landslide event according to the distribution diagram of the earthquake intensity parameter; An evaluation module is used to select a landslide hazard evaluation model corresponding to the earthquake intensity level according to the earthquake intensity level and background condition data of the multiple regions, and to perform hazard evaluation on each region respectively, so as to obtain landslide hazard evaluation results of the multiple regions included in the earthquake-induced landslide event; The display module is used to display the landslide hazard assessment result.

8. An electronic device, characterized in that: include: A processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method for rapid assessment of regional earthquake-induced landslide hazard according to any one of claims 1 to 6 when executing the computer program.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the method for rapid assessment of regional earthquake-induced landslide hazard according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for rapid assessment of regional earthquake-induced landslide hazard as claimed in any one of claims 1 to 6 is implemented.

Images