Geological disaster susceptibility assessment method based on risk grade division

Abstract

The invention discloses a geological disaster susceptibility assessment method based on risk grade division, relates to the technical field of geological disaster assessment, and solves the problem of unreasonable risk area division in an existing assessment method. According to the method, the optimal grading method with the highest consistency is selected from multiple grading methods based on the grading result and the regional historical disaster consistency, so that boundaries between different risk areas can be described more accurately, and boundary fuzziness and risk misjudgment caused by a manually set fixed threshold value are avoided; by outputting a geological disaster risk level distribution diagram with intuition, comparability and practicability, the method is suitable for disaster prevention and decision support of key areas such as power transmission channels, traffic engineering and town planning.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of geological disaster assessment, and in particular to a geological disaster susceptibility assessment method based on risk level division. BACKGROUND

[0002] As a major natural disaster type threatening human life and property safety and ecological environment stability, the susceptibility assessment of geological disasters is a core technical link in the disaster prevention and mitigation work system, directly determining the accuracy and effectiveness of disaster warning, engineering management and emergency response. With the intensification of climate change and the increase in the intensity of human engineering activities, the frequency and damage degree of geological disasters such as landslides, collapses and debris flows are showing a significant upward trend, which puts forward higher requirements for the scientificity and practicality of the assessment method.

[0003] The current mainstream geological disaster risk assessment method mostly focuses on the quantitative calculation of disaster occurrence probability, and through the construction of statistical models, machine ensemble learning models, etc., the historical disaster data and disaster-causing factors are fitted and analyzed, and finally a single probability prediction result is output. Then, by setting a threshold, the probability is classified into a certain risk level.

[0004] Such a method does not consider the spatial heterogeneity of geological disaster risk and the difference of disaster-causing factors, which is easy to lead to fuzzy division of high-risk and low-risk areas, and even risk misjudgment.

[0005] Therefore, a geological disaster susceptibility assessment method based on risk level division is needed. SUMMARY

[0006] In view of the unreasonable risk area division in the prior art, the present application provides a geological disaster susceptibility assessment method based on risk level division, which can reasonably divide the risk area. The specific technical solutions are as follows: In the first aspect, the present application provides a geological disaster susceptibility assessment method based on risk level division, comprising: obtaining a geological disaster susceptibility probability prediction result set of a target area, the prediction result set including the disaster occurrence probability of each grid cell in the target area; selecting an optimal grading method from a plurality of grading methods based on the consistency of the grading result and the regional historical disasters; dividing the disaster occurrence probability of each grid cell into a risk level based on the optimal grading method; and outputting a geological disaster risk level distribution map of the target area based on the risk level of the grid cell.

[0007] Preferably, the plurality of grading methods includes a natural breakpoint method; the natural breakpoint method solves an optimal threshold value by an iterative optimization algorithm, the optimal threshold value being used to divide the risk level; and the objective function of the iterative optimization algorithm includes: ; The constraint conditions of the iterative optimization algorithm include: 0= =1; wherein, is the threshold value to be optimized, used to divide the disaster occurrence probability into 5 intervals, each interval corresponding to a risk level; k is the interval index, represents the data set corresponding to interval k after division; is the disaster occurrence probability of the i-th grid unit, is the mean value of the interval k data set.

[0008] Preferably, based on the consistency of the classification results and the regional historical disasters, the optimal classification method is selected from a plurality of classification methods, including: using the plurality of classification methods, respectively classifying the target region based on the prediction result set to obtain a plurality of classification results corresponding; based on the distribution of administrative units, engineering corridors and / or geological disaster hidden danger points in the target region, and the historical disaster events in the target region, the risk classification verification is performed on the plurality of classification results; based on the risk classification verification result, the optimal classification method is determined.

[0009] Preferably, the risk classification verification includes Kappa coefficient verification, and the risk classification verification result includes Kappa coefficient; based on the risk classification verification result, the optimal classification method is determined, including: selecting the classification method corresponding to the classification result with the highest Kappa coefficient as the optimal classification method; the calculation formula of the Kappa coefficient includes: ; wherein, is the consistency rate of the predicted risk level and the historical risk level based on the historical disaster events, is the random consistency probability.

[0010] Preferably, the target region includes a power transmission channel; based on the risk level of the grid unit, the geological disaster risk level distribution map of the target region is output, including: based on the risk level of the grid unit, the risk level along the power transmission channel is calculated; based on the risk level along the power transmission channel and the risk level of the grid unit outside the power transmission channel, the geological disaster risk level distribution map is output; wherein, the calculation formula of the risk level along the power transmission channel includes: ; wherein, represents the risk level along the power transmission channel, is the length of the risk level k along the power transmission channel, is the total length of the power transmission channel.

[0011] Preferably, the geological disaster risk level distribution map is in GeoTIFF format, which is used for visual display and spatial superposition in a geographic information system (GIS) platform; based on the risk level of the grid cell, the method outputs a geological disaster risk level distribution map of the target area, and the method further comprises: receiving a risk assessment instruction, the risk assessment instruction being used to instruct risk level assessment on a sub-region in the target area; based on the risk level of the grid cell of the sub-region, calculating a risk level proportion of the sub-region; the calculation formula of the risk level proportion of the sub-region comprises: ; wherein, represents the proportion of the part with a risk level k in the sub-region, is an indicator function, and |A| is the total number of grids in the sub-region A, represents the risk level k corresponding to the disaster occurrence probability of the grid cell in the sub-region A.

[0012] Preferably, the disaster occurrence probability is calculated based on the prediction results of a plurality of ensemble learning models; the disaster occurrence probability calculation formula comprises: ; wherein, is the prediction probability of the ith grid cell by the tth model, is the weight of the tth model, and ; the plurality of ensemble learning models comprise one or more of a random forest model, an extreme random tree model, an extreme gradient boosting model, a light gradient boosting machine model, a gradient boosting model, and a histogram gradient boosting model.

[0013] In a second aspect, an embodiment of the present application provides a geological disaster proneness assessment system based on risk level division, which is applied to the method of the first aspect, and the system comprises: an acquisition module configured to acquire a set of geological disaster proneness probability prediction results of a target area, the prediction result set comprising a disaster occurrence probability of each grid cell in the target area; a selection module configured to select an optimal grading method from a plurality of grading methods based on the consistency of the grading result and the historical disasters in the region; a division module configured to divide the disaster occurrence probability of each grid cell into a risk level based on the optimal grading method; an output module configured to output a geological disaster risk level distribution map of the target area based on the risk level of the grid cell.

[0014] In a third aspect, an embodiment of the present application provides a computing device, comprising: a memory configured to store a program; and a processor configured to load the program to execute the method according to the first aspect.

[0015] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, comprising a stored program, wherein the program, when executed, controls a device on which the computer-readable storage medium is located to perform the method according to the first aspect.

[0016] Compared with the prior art, the present application has the beneficial effects that: by selecting the optimal grading method with the highest consistency from multiple grading methods based on the consistency of the grading results and the regional historical disasters, the boundaries between different risk areas can be more accurately described, and the boundary ambiguity and risk misjudgment caused by the fixed threshold set by humans can be avoided; by outputting the geological disaster risk grade distribution map with intuitiveness, comparability and practicality, the geological disaster risk grade distribution map is suitable for disaster prevention and decision support in key areas such as power transmission channels, traffic engineering and urban planning. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the drawings needed in the specific embodiments or the prior art description will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual proportions.

[0018] Figure 1 A flowchart of a geological disaster prone area assessment method based on risk grade division provided by an embodiment of the present application; Figure 2 Figure 2 A specific example diagram of a geological disaster risk grade distribution map of a city provided by an embodiment of the present application; Figure 3 A structural diagram of a geological disaster prone area assessment system based on risk grade division provided by an embodiment of the present application; Figure 4 A structural diagram of a computing device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0019] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0020] It should be understood that the terms "comprises" and "comprising," when used in this specification and the following claims, indicate the presence of the described features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0021] It should also be understood that the terms used in the specification of the application are only for the purpose of describing particular embodiments and are not intended to limit the application. As used in the specification and the appended claims of the application, the singular forms "a," "an," and "the" are intended to include the plural forms unless the context clearly indicates otherwise.

[0022] It should be further understood that the term "and / or" used in the specification of the application means one or more of the associated listed items and all possible combinations thereof, and includes these combinations.

[0023] To solve the problem of unreasonable risk area division in geological disaster assessment method, the application provides a geological disaster prone area assessment method based on risk level division, which can reasonably divide the risk area.

[0024] Please refer to Figure 1 , Figure 1 A flowchart of a geological disaster prone area assessment method based on risk level division is provided for the embodiments of the application, and the method is applied to a computing device. Figure 1 As shown in the figure, the method comprises the following steps. Step 101, the computing device obtains a geological disaster prone area probability prediction result set of a target area.

[0025] The computing device can be a terminal or a server, specifically, the computing device can be a personal computer, a notebook computer, a smart phone, a tablet computer or other devices with data processing capability.

[0026] The target area refers to an area that needs to be evaluated for geological disaster prone area, specifically, a whole connected area. If there are multiple unconnected areas that need to be evaluated for geological disaster prone area, the multiple areas need to be sequentially taken as the target area to execute the method of the embodiments of the application.

[0027] The geological disaster prone area probability prediction result set prediction result set P= includes the disaster occurrence probability of each grid unit i in the target area ∈[0,1].

[0028] Preferably, the disaster occurrence probability is calculated based on the prediction results of multiple ensemble learning models; the disaster occurrence probability calculation formula comprises: ; in, Let be the prediction probability of the t-th model for the i-th grid cell. The weights of the t-th model are... The multiple ensemble learning models include one or more of the following: random forest model, extreme random tree model, extreme gradient boosting model, lightweight gradient boosting machine model, gradient boosting model, and histogram gradient boosting model.

[0029] The prediction output of the random forest model is: ; Where N is the number of decision trees. This represents the prediction result for the i-th tree.

[0030] The computing device can deploy multiple ensemble learning models on itself to calculate the probability of the disaster based on its own computing power; or it can form a computing cluster with other computing devices, with the multiple ensemble learning models deployed on different devices in the computing cluster; and then, after the computing cluster calculates the probability of the disaster, it returns the calculation result to the computing device.

[0031] Specifically, the computing device can use the dynamic and static factors of the target area as inputs to multiple ensemble learning models. These ensemble learning models then output a high-precision probability of disaster occurrence corresponding to each grid cell through factor optimization, weight assignment, and fusion calculation of dynamic and static factors.

[0032] For example, dynamic factors include monthly precipitation, surface runoff depth, maximum temperature, and minimum temperature.

[0033] For example, static factors include digital elevation model, slope, aspect, plane curvature, profile curvature, soil type, land use type, normalized vegetation index, enhanced vegetation index, and population density.

[0034] Step 102: The computing device selects the optimal classification method from multiple classification methods based on the consistency between the classification results and the historical disasters in the region.

[0035] The computing device can be pre-set with multiple classification methods, which are used to classify risk levels; more specifically, they are used to classify risk levels based on the probability of different disasters occurring.

[0036] Preferably, the computing device can classify the probability results into five risk levels using a grading method: extremely low (L1), low (L2), medium (L3), high (L4), and extremely high (L5), satisfying the following: in This is the upper limit threshold for the k-th level.

[0037] Preferably, the multiple classification methods include the natural breakpoint method; the natural breakpoint method solves for the optimal threshold through an iterative optimization algorithm, and the optimal threshold is used to classify the risk level; the objective function of the iterative optimization algorithm includes: ; The constraints of this iterative optimization algorithm include: 0 = =1; where, The threshold value to be optimized is used to divide the probability of this disaster occurring into 5 intervals, each interval corresponding to a risk level; k is the interval index. This represents the dataset corresponding to interval k after partitioning; It represents the probability of a disaster occurring in the i-th grid cell. It is the mean of the k-interval dataset.

[0038] The objective function of the natural breakpoint method is to minimize the sum of squared differences between the data within an interval and the mean of the interval dataset, which is to minimize the sum of variances within classes from the ClassMean (SDCM). This allows us to find a partitioning method that minimizes the differences within the interval dataset and maximizes the differences between different interval datasets.

[0039] Optionally, the various grading methods include the equal interval method; the formula for calculating the grading intervals in the equal interval method includes: .

[0040] Optionally, the various grading methods include the quantile method; the quantile method determines the grading threshold using the cumulative distribution function. Where F(·) is the probability distribution function, For the k-th grading threshold The corresponding quantiles.

[0041] Optionally, the various grading methods include the fuzzy membership function method. This fuzzy membership function method can use sigmoid, linear, or exponential membership functions to calculate the grading threshold. An sigmoid membership function can be as follows: ; in, This is the k-th grading threshold.

[0042] An exponential membership function for calculating the extremely high risk level (L5) can be as follows: Where k is a shape parameter that controls the growth rate of membership degree.

[0043] The computing device can use the above-mentioned multiple classification methods, or more classification methods, to classify based on a preset test set, a partial prediction result set, or a complete prediction result set, respectively; then, combined with the historical disaster events of the corresponding region, the consistency between the classification results and the historical disaster situation of the corresponding region is calculated; and then the classification method with the highest consistency is adopted as the optimal classification method.

[0044] Preferably, the computing device can employ the multiple classification methods to classify the risk of the target area based on the prediction result set, thereby obtaining multiple classification results; based on the distribution of administrative units, engineering corridors and / or geological hazard points in the target area, as well as historical disaster events in the target area, the multiple classification results are verified for risk classification; based on the risk classification verification results, the optimal classification method is determined.

[0045] After classifying the target area into risk levels, in order to verify the reasonableness and accuracy of the classification results, the computing device can combine the classification results with the distribution of the following types of actual geographical or engineering-related elements to verify the risk zoning: Administrative units are regions with clearly defined boundaries and management entities, such as provinces, cities, counties, and townships. Combining risk classification results with administrative units allows us to see whether the distribution of risk levels within the same administrative unit matches the actual situation within that region. For example, can factors such as the geological conditions and disaster history of a particular county support the classification results? On the other hand, it also facilitates administrative departments to carry out risk prevention and control work more efficiently based on the management scope of the administrative unit. If the classification results match the actual situation within the administrative unit to a high degree, it indicates that the classification is reasonable.

[0046] Engineering corridors refer to narrow, elongated areas traversed by engineering facilities such as major transportation lines (railways, highways), power transmission lines, and oil and gas pipelines. These areas often have specific engineering geological conditions and are more sensitive to geological hazards. Combining risk classification results with engineering corridors can verify whether the risk level classification along and around the engineering corridors reflects the actual geological hazard risks faced by the engineering construction. For example, whether mountainous sections through which the engineering corridor passes have been reasonably designated as high-risk areas to ensure the safe planning of engineering construction and operation.

[0047] Geological hazard sites are identified locations where geological disasters may occur. Comparing the risk classification results with the distribution of these hazard sites, if high-risk zones contain more geological hazard sites and low-risk zones contain fewer hazard sites, it indicates that the classification results can better reflect the actual disaster risk distribution and verify the effectiveness of the classification.

[0048] Specifically, the computing device can count the number of geological disaster events occurring in each grid cell of the target area and verify whether the number of historical geological disaster events in each grid cell is proportional to its assigned risk level; then, based on the number of grid cells that pass the verification, the optimal grading method is determined.

[0049] Preferably, the risk grading verification includes Kappa coefficient verification, and the risk grading verification result includes a Kappa coefficient; the computing device can calculate the Kappa coefficient for each grading method; then, the grading method corresponding to the grading result with the highest Kappa coefficient is selected as the optimal grading method; the formula for calculating the Kappa coefficient includes: ; in, To predict the consistency rate between risk levels and historical risk levels derived from historical disaster events, This represents the probability of random consistency.

[0050] Specifically, the computing device can normalize the number of geological disaster events occurring in each grid cell based on the maximum and minimum number of historical disaster events corresponding to each grid cell in the target area; then, based on the normalized values, a historical risk level is obtained. This historical risk level is set in the same way as the predicted risk level.

[0051] Preferably, the computing device can treat the engineering corridor and / or areas with potential geological hazards as key areas, calculate the Kappa coefficient separately, and assign it a higher weight; then, it can combine the two Kappa coefficients of the key areas and the ordinary areas for weighted calculation to obtain the final Kappa coefficient used to evaluate the grading method.

[0052] Step 103: The computing device, based on the optimal classification method, classifies the probability of disaster occurrence for each grid cell into risk levels.

[0053] If the computing device has already classified the risk level of each grid cell in the target area based on the full set of prediction results using the optimal classification method in step 102, then the computing device can directly use the classification result to execute step 104; otherwise, the probability of disaster occurrence of each grid cell is classified into risk levels using the optimal classification method.

[0054] Step 104: The calculation device outputs a geological hazard risk level distribution map of the target area based on the risk level of the grid cell.

[0055] The geological hazard risk level distribution map is stored in raster or vector format.

[0056] Preferably, the target area includes a power transmission channel; the computing device can calculate the risk level along the power transmission channel based on the risk level of the grid cell; and output the geological hazard risk level distribution map based on the risk level along the power transmission channel and the risk level of the grid cells outside the power transmission channel.

[0057] The formula for calculating the risk level along the transmission line includes: ; in, Indicates the risk level along the power transmission corridor. Let risk level k be the length along the transmission line. This represents the total length of the channel.

[0058] Specifically, the computing device can first filter the grid cells that intersect with the transmission channel, and then calculate the length of the transmission channel within the grid cell; then, for each risk level k, it can count all the grid cells marked with risk level k that the transmission channel passes through, and calculate the sum of the lengths of the line segments of the transmission channel within these grid cells; then, based on the above calculation formula, it can calculate the risk level along the transmission channel by weighting.

[0059] By performing a weighted calculation of the overall risk level along the power transmission corridor, we can quickly understand the overall geological disaster risk faced by the entire corridor from a macro perspective. We can clearly determine whether the power transmission corridor is at a high, medium, or low risk level, which will help relevant management departments to have an intuitive and comprehensive understanding of the safety status of the power transmission corridor from a global perspective, and provide a macro basis for overall resource allocation and safety decision-making.

[0060] By calculating the overall risk level using a weighted average, the impact of different risk level areas along the corridor on the entire corridor can be comprehensively considered.

[0061] Preferably, the geological hazard risk level distribution map is in GeoTIFF format, used for visualization on a Geographic Information System (GIS) platform, and spatially overlaid with a Digital Elevation Model (DEM), administrative divisions, or power transmission channel locations to form an intuitive risk distribution map. After outputting the geological hazard risk level distribution map of the target area based on the risk level of the raster unit, the computing device can receive a risk assessment instruction, which instructs the risk level assessment of sub-regions within the target area. Based on the risk level of the raster units of the sub-region, the risk level percentage of the sub-region is calculated.

[0062] After obtaining the geological disaster risk level distribution map, the computing device can visualize it in conjunction with the geographic information system platform and provide it to the user for browsing; when the user selects a sub-region composed of grid units, or selects a pre-set sub-region, the computing device can calculate the risk level percentage of the corresponding sub-region and return it to the front end for display.

[0063] The formula for calculating the risk level percentage of this sub-region includes: ; in, This indicates the proportion of the sub-region with risk level k. Let |A| be the indicator function, and |A| be the total number of grid cells in sub-region A. Represents the grid cells within sub-region A The risk level corresponding to the probability of a disaster occurring is k.

[0064] In this embodiment, by selecting the optimal classification method with the highest consistency from multiple classification methods based on the consistency of classification results and regional historical disasters, the boundaries between different risk zones can be described more accurately, avoiding boundary ambiguity and risk misjudgment caused by artificially set fixed thresholds. By outputting a geological disaster risk level distribution map that is intuitive, comparable, and practical, it is suitable for disaster prevention and decision support in key areas such as power transmission channels, transportation projects, and urban planning.

[0065] The method portion of this application's embodiment will be further described below through a specific example. Please refer to [link / reference needed]. Figure 2 , Figure 2 This is a specific example diagram of the geological disaster risk level distribution map of a certain city provided in the embodiments of this application.

[0066] by Figure 2 The 220kV transmission network of a certain city shown is the target area. This target area encompasses diverse landforms such as mountains, basins, and river valleys, with an average annual rainfall of 1200mm. Geological hazards are mainly rainfall-induced landslides. Based on the susceptibility probability prediction results, this embodiment details the application process of risk level classification: 1. Data Acquisition and Preprocessing Probabilistic prediction results: A GeoTIFF format susceptibility probability grid based on a dynamic-static factor fusion model was used, with a resolution of 30m and a study area of ​​approximately 1500km². The probability values ​​ranged from 0 to 1, with approximately 1.67 million grids within the transmission channel buffer zone (5km range).

[0067] Auxiliary data: 30m resolution DEM, administrative division vector map, transmission line vector data (including 12 lines with a total length of 350km), and historical landslide point data (89 in total from 2015 to 2023).

[0068] 2. Selection of grading method and parameter settings Method Comparison: Four methods were used to pre-classify the probability data, and the results are shown in Table 1. Table 1. Comparison of grading results from various grading methods Optimal method selection: Based on the Kappa coefficient verification of historical landslide points, the natural breakpoint method was found to have the highest Kappa value (specifically 0.76), so it was selected as the final classification method, with thresholds of T1=0.12, T2=0.27, T3=0.45, and T4=0.68.

[0069] 3. Risk Level Classification and GIS Mapping Grading: According to the formula The risk classification was implemented, resulting in five risk levels, with the extremely high-risk area (L5) accounting for 12.3% of the total area, mainly distributed in the northeastern mountainous region.

[0070] Then, the computing device overlays the risk classification data of each grid cell onto the power transmission channel vector in ArcGIS software to generate a distribution map of geological disaster risk levels along the power transmission channel. High-risk areas and extremely high-risk areas are distributed in a "point + strip" pattern along the channel.

[0071] Then, the computing device automatically generates a hierarchical statistical table: within the channel buffer range, L1 accounts for 28.5%, L2 accounts for 25.2%, L3 accounts for 21.8%, L4 accounts for 12.2%, and L5 accounts for 12.3%.

[0072] Then, generate a risk level legend and color scale (very low: light blue, low: blue, medium: yellow, high: orange, very high: red).

[0073] 4. Results Output and Validation Spatial consistency: Of the 89 historical landslide sites, 76 (85.4%) are located in high-risk areas (L4) and extremely high-risk areas (L5), indicating that the classification results are reasonable.

[0074] Partition accuracy: Kappa coefficient = 0.76, accuracy rate = 83.1%, which meets the requirements of engineering applications.

[0075] Risk section identification: The total length of L5 section along the corridor is 42km, concentrated in the K20-K35, K80-K95 and K150-K170 sections of the three lines, and prevention and control projects should be carried out in these areas first.

[0076] Prevention and control recommendations: For section L5, adopt the measures of "interception and drainage system + slope reinforcement" and for section L4, adopt the measures of "automated monitoring + regular inspection". It is expected to reduce the disaster risk by more than 60%.

[0077] Application of results: The output GeoTIFF format risk map has been integrated into the city's power GIS platform, supporting dynamic risk updates and visual management.

[0078] As can be seen from this embodiment, the method of this application can achieve reasonable classification and visual expression of the probability results of geological disaster susceptibility.

[0079] The method provided in the embodiments of this application has been described above. The system provided in the embodiments of this application will be described below.

[0080] Please see Figure 3 , Figure 3 A schematic diagram of a geological hazard susceptibility assessment system based on risk level classification is provided for an embodiment of this application, as shown below. Figure 3 As shown, the system 30 includes: The acquisition module 301 is used to acquire a set of geological hazard susceptibility probability prediction results for a target area. The prediction result set includes the probability of hazard occurrence for each grid cell in the target area. Selection module 302 is used to select the optimal classification method from multiple classification methods based on the consistency between the classification results and the historical disasters in the region; The partitioning module 303 is used to classify the probability of disaster occurrence of each grid cell into risk levels based on the optimal classification method; Output module 304 is used to output a geological hazard risk level distribution map of the target area based on the risk level of the grid cell.

[0081] Preferably, the multiple classification methods include the natural breakpoint method; the natural breakpoint method solves for the optimal threshold through an iterative optimization algorithm, and the optimal threshold is used to classify the risk level; the objective function of the iterative optimization algorithm includes: ; The constraints of this iterative optimization algorithm include: 0 = =1; where, The threshold value to be optimized is used to divide the probability of this disaster occurring into 5 intervals, each interval corresponding to a risk level; k is the interval index. This represents the dataset corresponding to interval k after partitioning; It represents the probability of a disaster occurring in the i-th grid cell. It is the mean of the k-interval dataset.

[0082] Preferably, the selection module 302 is specifically used to employ the multiple classification methods to classify the risk of the target area based on the prediction result set, thereby obtaining multiple classification results; to verify the risk classification of the multiple classification results based on the distribution of administrative units, engineering corridors and / or geological disaster hazard points in the target area, as well as historical disaster events in the target area; and to determine the optimal classification method based on the risk classification verification results.

[0083] Preferably, the risk grading verification includes Kappa coefficient verification, and the risk grading verification result includes a Kappa coefficient; the selection module 302 is specifically used to select the grading method corresponding to the grading result with the highest Kappa coefficient as the optimal grading method; the formula for calculating the Kappa coefficient includes: ; in, To predict the consistency rate between risk levels and historical risk levels derived from historical disaster events, This represents the probability of random consistency.

[0084] Preferably, the target area includes a power transmission corridor; the process of outputting a geological hazard risk level distribution map of the target area based on the risk level of the grid cells includes: calculating the risk level along the power transmission corridor based on the risk level of the grid cells; and outputting the geological hazard risk level distribution map based on the risk level along the power transmission corridor and the risk levels of grid cells outside the power transmission corridor; wherein the formula for calculating the risk level along the power transmission corridor includes: ; in, Indicates the risk level along the power transmission corridor. Let risk level k be the length along the transmission line. Total length of the transmission channel Preferably, the geological hazard risk level distribution map is in GeoTIFF format for visualization and spatial overlay on a Geographic Information System (GIS) platform. The system 30 also includes: a regional assessment module 305, used to receive risk assessment instructions, which instruct the assessment of risk levels in sub-regions within the target area; and to calculate the risk level percentage of the sub-region based on the risk level of its raster cells; the formula for calculating the risk level percentage of the sub-region includes: ; in, This indicates the proportion of the sub-region with risk level k. Let |A| be the indicator function, and |A| be the total number of grid cells in sub-region A. Represents the grid cells within sub-region A The risk level corresponding to the probability of a disaster occurring is k.

[0085] Preferably, the probability of disaster occurrence is calculated based on the prediction results of multiple ensemble learning models; the formula for calculating the probability of disaster occurrence includes: ; in, Let be the prediction probability of the t-th model for the i-th grid cell. The weights of the t-th model are... The multiple ensemble learning models include one or more of the following: random forest model, extreme random tree model, extreme gradient boosting model, lightweight gradient boosting machine model, gradient boosting model, and histogram gradient boosting model.

[0086] The geological hazard susceptibility assessment system based on risk level classification provided in this application can be understood by referring to the relevant content in the foregoing method embodiment section, and will not be repeated here.

[0087] like Figure 4 As shown, Figure 4 This is a schematic diagram of a possible logical structure of a computing device provided in an embodiment of this application. The computing device 40 includes a processor 401, a communication interface 402, a memory 403, and a bus 404. The processor 401, the communication interface 402, and the memory 403 are interconnected via the bus 404. In an embodiment of this application, the processor 401 is used to control and manage the operation of the computing device 40. For example, the processor 401 is used to execute... Figure 1 The steps in the embodiments and / or other processes used in the techniques described herein. Communication interface 402 is used to support communication by computing device 40. Memory 403 is used to store program code and data of computing device 400.

[0088] The processor 401 can be a central processing unit, a general-purpose processor, a digital signal processor, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. The processor can also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a digital signal processor and a microprocessor, etc. The bus 404 can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0089] In another embodiment of this application, a computer-readable storage medium is also provided, the computer-readable storage medium including instructions that, when executed on a computer, cause the computer to perform the above-described... Figure 1 The method described in the embodiments.

[0090] Those skilled in the art will recognize that the units of the various examples described in connection with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0091] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0092] In the embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, 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 couplings, direct couplings, or communication connections may be through some interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0093] 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 network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0094] 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.

[0095] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they 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.

[0096] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the claims and specification of the present invention.

Claims

1. A method for assessing geological hazard susceptibility based on risk level classification, characterized in that, include: Obtain a set of geological hazard susceptibility probability prediction results for a target area, wherein the prediction result set includes the probability of hazard occurrence for each grid cell in the target area; Based on the consistency between the classification results and the region's historical disasters, the optimal classification method is selected from multiple classification methods; Based on the optimal classification method, the probability of disaster occurrence for each grid cell is divided into risk levels; Based on the risk level of the grid cells, a geological hazard risk level distribution map of the target area is output.

2. The method according to claim 1, characterized in that, The various grading methods include the natural breakpoint method; The natural breakpoint method solves for the optimal threshold through an iterative optimization algorithm, and the optimal threshold is used to classify the risk level. The objective function of the iterative optimization algorithm includes: ； The constraints of the iterative optimization algorithm include: 0 = =1; in, The threshold to be optimized is used to divide the probability of the disaster occurrence into 5 intervals, each interval corresponding to a risk level; k is the interval index. This represents the dataset corresponding to interval k after partitioning; It represents the probability of a disaster occurring in the i-th grid cell. It is the mean of the k-interval dataset.

3. The method according to claim 1, characterized in that, Based on the consistency between the grading results and historical disaster data in the region, the optimal grading method is selected from multiple grading methods, including: Using the aforementioned multiple classification methods, the target area is classified into risk levels based on the prediction result set, resulting in multiple corresponding classification results; Based on the distribution of administrative units, engineering corridors and / or geological hazard points in the target area, as well as historical disaster events in the target area, the risk classification of the multiple classification results is verified. Based on the risk grading verification results, the optimal grading method is determined.

4. The method according to claim 3, characterized in that, The risk grading verification includes Kappa coefficient verification, and the risk grading verification result includes the Kappa coefficient; determining the optimal grading method based on the risk grading verification result includes: The grading method corresponding to the grading result with the highest Kappa coefficient is selected as the optimal grading method. The formula for calculating the Kappa coefficient includes: ； in, To predict the consistency rate between risk levels and historical risk levels derived from historical disaster events, This represents the probability of random consistency.

5. The method according to claim 1, characterized in that, The target area includes power transmission channels; the step of outputting a geological hazard risk level distribution map of the target area based on the risk level of the grid cells includes: Based on the risk level of the grid cells, the risk level along the power transmission channel is calculated. Based on the risk level along the power transmission channel and the risk level of the grid cells outside the power transmission channel, the geological hazard risk level distribution map is output; The formula for calculating the risk level along the power transmission corridor includes: ； in, Indicates the risk level along the power transmission corridor. Let risk level k be the length along the transmission line. This represents the total length of the power transmission channel.

6. The method according to claim 1, characterized in that, The geological hazard risk level distribution map is in GeoTIFF format and is used for visualization and spatial overlay in the geographic information system (GIS) platform. After outputting the geological hazard risk level distribution map of the target area based on the risk level of the grid cells, the method further includes: Receive a risk assessment instruction, which is used to instruct a risk level assessment of a sub-region within the target region; Based on the risk level of the grid cells in the sub-region, calculate the risk level percentage of the sub-region; The formula for calculating the risk level percentage of the sub-regions includes: ； in, This indicates the proportion of the sub-region with risk level k. Let |A| be the indicator function, and |A| be the total number of grid cells in sub-region A. Represents the grid cells within sub-region A The risk level corresponding to the probability of a disaster occurring is k.

7. The method according to claim 1, characterized in that, The probability of disaster occurrence is calculated based on the prediction results of multiple ensemble learning models; the formula for calculating the probability of disaster occurrence includes: ； in, Let be the prediction probability of the t-th model for the i-th grid cell. The weights of the t-th model are... ; The multiple ensemble learning models include one or more of the following: random forest model, extreme random tree model, extreme gradient boosting model, lightweight gradient boosting machine model, gradient boosting model, and histogram gradient boosting model.

8. A geological hazard susceptibility assessment system based on risk level classification, characterized in that, The system, applied to the method of any one of claims 1-7, comprises: The acquisition module is used to acquire a set of geological hazard susceptibility probability prediction results for a target area, wherein the prediction result set includes the probability of hazard occurrence for each grid cell in the target area; The selection module is used to select the optimal classification method from multiple classification methods based on the consistency between the classification results and the historical disasters in the region. The partitioning module is used to classify the probability of disaster occurrence of each grid cell into risk levels based on the optimal classification method. The output module is used to output a geological hazard risk level distribution map of the target area based on the risk level of the grid cells.

9. A computing device, characterized in that, include: Memory, used to store programs; A processor for loading the program to perform the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein, when the program is executed, it controls the device on which the computer-readable storage medium is located to perform the method of any one of claims 1-7.

Images