Slope group safety risk monitoring classification method and evaluation system

Through the combination of historical data and InSAR data, combined with TOPSIS and AISM algorithms, the risk assessment of slope groups is systematically carried out, which solves the problem of single slope stability classification method in the existing technology, and achieves more accurate risk identification and Improve management efficiency.

CN119004194AInactive Publication Date: 2024-11-22NINGBO LANGDA ENG TECH CO LTD

Patent Information

Application Number
CN202411489496.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-24
Publication Date
2024-11-22
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing slope stability classification method is single and cannot effectively cover multiple scenarios and lithologies, resulting in monitoring points not necessarily corresponding to actual risk points, and lack of systematic risk identification and management.

Method used

Using the initial classification based on historical data, comprehensive evaluation of the clamping process and the refined classification method of InSAR data, combined with TOPSIS and AISM algorithms, the risk monitoring and evaluation of slope groups is monitored and evaluated through a systematic risk assessment model to identify potential risks and Optimize resource configuration.

Benefits of technology

It realizes more accurate risk identification, improves management efficiency and real-time information, supports scientific and reasonable management and repair plans, and ensures the safety and sustainable development of slopes.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119004194A_ABST
    Figure CN119004194A_ABST
Patent Text Reader

Abstract

The invention discloses a slope group safety risk monitoring classification method and evaluation system, and the method comprises the following steps: carrying out the initial classification of a slope group based on the historical data of the slope group; obtaining a comprehensive evaluation rough classification result of each side slope in a manner of solving a clamping process according to the result of the initial classification and the side slope risk influence factors; carrying out fine classification on the unstable slopes in combination with InSAR data of the regional slope group; the evaluation system uses the slope group safety risk monitoring and classifying method. The beneficial effects of the invention are that through systematic classification of the side slope, potential risks can be identified more accurately, measures can be taken in time, and the risks of landslide and collapse can be reduced. After classification, resources and manpower can be reasonably allocated, power is concentrated to solve the key hidden slope problem, and the management efficiency is improved. Dynamic monitoring can be realized based on a classification model, the slope state is updated in time, and the real-time performance and the accuracy of information are ensured.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of slope safety monitoring, and in particular to a slope group safety risk monitoring classification method and assessment system. Background Art

[0002] Slope deformation monitoring technology has been continuously improving, from the initial manual survey to today's advanced automatic monitoring equipment. However, the classification method of slope hazards has not been innovated accordingly. It is still based on the scale of the slope itself, the degree of hazard after the landslide, and the judgment of risk by manual survey, which often results in the monitoring points not necessarily corresponding to the actual risk points.

[0003] At present, the technology of slope stability classification methods mainly focuses on the classification of a single lithology or a single scene. Although these methods have their advantages, they also have certain limitations. First, the scene is single, mostly concentrated in scenes such as open-pit mines, and does not cover other comprehensive scenes; second, the lithology is single, mostly rock slope classification, and does not involve the classification of other lithology slopes. Summary of the invention

[0004] One of the purposes of the present application is to provide a slope group safety risk monitoring classification method that can solve at least one of the defects in the above-mentioned background technology.

[0005] Another object of the present application is to provide a slope group safety risk assessment system that can solve at least one of the defects in the above-mentioned background technology.

[0006] In order to achieve at least one of the above purposes, the technical solution adopted in this application is: a slope group safety risk monitoring classification method, comprising the following steps: S100: Preliminary classification of slope groups based on historical data of slope groups; S200: Based on the results of the initial classification and the factors affecting the slope risk, a rough classification result of the comprehensive evaluation of each slope is obtained by solving the clamping process; the rough classification result of the comprehensive evaluation of each slope is divided into stable slopes, unstable slopes and unstable slopes; S300: Refined classification of the unstable slopes in step S200 is performed in combination with the InSAR data of the regional slope group, and the identification results of the refined classification are divided into key hidden danger slopes and general hidden danger slopes.

[0007] Preferably, in step S100, the initial classification of the slope group includes lithology classification, height classification and slope classification; according to lithology classification, each slope is divided into rock slope and soil slope; according to height classification, each slope is divided into low slope, medium-high slope, high slope and super-high slope; according to slope classification, each slope is divided into gentle slope, medium slope, steep slope, abrupt slope and reverse slope.

[0008] Preferably, in step S200, the clamping process is solved by jointly solving the TOPSIS algorithm and the AISM algorithm, including the following steps: S210: According to the factors affecting slope risk, the historical data of the slope group is reduced in dimension and reasonable evaluation indicators are selected; S220: collecting performance data of each slope under different evaluation indicators to form an original decision matrix and perform preprocessing; S230: performing standardization processing on the preprocessed original decision matrix to obtain a distance decision matrix; S240: Calculate the similarity index of each slope according to the result of the normalization process and sort them to obtain a closeness matrix; S250: Generate an adversarial topology hierarchy diagram corresponding to each matrix in steps S220 to S240 through the AISM algorithm, and analyze the generated adversarial topology hierarchy diagram to obtain a comprehensive evaluation result of each slope.

[0009] Preferably, the evaluation indicators selected in step S210 include geological conditions, hydrological conditions, slope geometry, human activities and historical records.

[0010] Preferably, the k evaluation indicators selected in step S210 are marked as evaluation indicators 1 to evaluation indicators k in sequence; in step S230, the maximum normalization and minimum normalization of the performance data of the slope under each evaluation indicator are calculated to perform standardization processing; the maximum normalization corresponding to any slope i and minimal normalization The calculation formula is as follows: ; ; Among them, x ij represents the performance data of slope i corresponding to any evaluation index j, represents the optimal value of the evaluation index j, Represents the worst value of evaluation index j.

[0011] Preferably, in step S240, the similarity index C corresponding to any slope i i The calculation formula is as follows: .

[0012] Preferably, step S300 includes the following process: S310: Acquire high-resolution InSAR data of the slope group and perform preprocessing; S320: Compare displacement data at different time points according to the preprocessed data to obtain the deformation trend of the slope; S330: identifying deformation features of the slope by feature extraction according to the obtained deformation trend, where the deformation features include displacement mode, displacement speed and displacement amplitude; S340: performing feature analysis based on the extracted deformation features combined with the geological environment features of the slope; S350: Perform fine classification using a machine learning model based on the results of feature analysis.

[0013] A slope group safety risk assessment system using the above-mentioned slope group safety risk monitoring and classification method includes a data cleaning module, a feature module and a model generation module; the data cleaning module is suitable for cleaning the input slope group historical data; the feature module is suitable for feature extraction and feature selection of the cleaned data; the model generation module is suitable for model training, model testing and debugging and tuning the data output by the feature module to determine the final slope risk assessment model; the slope risk assessment model is suitable for outputting the stability score of the slope based on the input real-time monitoring data of the slope.

[0014] Preferably, the cleaning methods of the data cleaning module include removing fields, filling missing values ​​and re-acquiring data; for fields with low importance and high missing rate in the historical data of the slope group, a cleaning method of removing fields is adopted; for fields with high importance and low missing rate in the historical data of the slope group, a cleaning method of filling missing values ​​is adopted; for fields with both high importance and missing rate in the historical data of the slope group, a cleaning method of re-acquiring data is adopted; for fields with both low importance and missing rate in the historical data of the slope group, no processing is performed or they are simply filled.

[0015] Preferably, the feature extraction of the feature module includes the following processes: calculating the statistics of historical data, including mean error, variance, maximum value and minimum value; extracting trend characteristics, seasonal characteristics and periodic characteristics of historical data, and using Fourier transform to analyze the spectral characteristics of monitoring data to identify potential vibration modes; performing morphological analysis on displacement or stress data to extract feature points; the feature selection of the feature module includes the following process: based on the results of feature extraction, identifying the features that have the greatest impact on slope stability prediction.

[0016] Compared with the prior art, the beneficial effects of this application are: (1) Through the systematic classification of slopes, potential risks can be identified more accurately and timely measures can be taken to reduce the risk of landslides and collapses.

[0017] (2) After classification, resources and manpower can be rationally allocated, and efforts can be concentrated on solving key slope problems with hidden dangers, thereby improving management efficiency.

[0018] (3) The classification-based model can realize dynamic monitoring and timely update of slope status to ensure the real-time and accuracy of information.

[0019] (4) Classification provides data support, which helps to formulate scientific and reasonable management and restoration plans and improve the effectiveness of decision-making.

[0020] (5) Systematic slope classification helps to establish a long-term monitoring and management mechanism to achieve sustainable development. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 This is a schematic diagram of the overall workflow of the slope group safety risk monitoring classification method in this application.

[0022] Figure 2 This is a flow chart of the initial classification of historical data for this application.

[0023] Figure 3 This is a schematic diagram of solving the pinching process using the TOPSIS algorithm and the AISM algorithm in this application.

[0024] Figure 4 This is a schematic diagram of the workflow of the slope group safety risk assessment system in this application. DETAILED DESCRIPTION

[0025] Below, the present application is further described in conjunction with specific implementation methods. It should be noted that in the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms should not be understood as necessarily being directed to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification.

[0026] In the description of the present application, it should be noted that directional words, such as the terms "center", "lateral", "longitudinal", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise", etc., indicating directions and positional relationships are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of narrating the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and cannot be understood as limiting the specific scope of protection of the present application.

[0027] It should be noted that the terms "first", "second", etc. in the description and claims of the present application are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0028] In this application, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be connected, detachably connected, or integrated; it can be mechanically connected or electrically connected; it can be directly connected or indirectly connected through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0029] In the present application, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may include that the first and second features are in direct contact, or may include that the first and second features are not in direct contact but are in contact through another feature between them. Moreover, a first feature being "above", "above" and "above" a second feature includes that the first feature is directly above and obliquely above the second feature, or simply indicates that the first feature is higher in level than the second feature. A first feature being "below", "below" and "below" a second feature includes that the first feature is directly below and obliquely below the second feature, or simply indicates that the first feature is lower in level than the second feature.

[0030] The terms "including" and "having" and any variations thereof in the specification and claims of the present application are intended to cover non-exclusive inclusions. For example, a process, method, system, product or apparatus comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products or apparatuses.

[0031] One aspect of the present application provides a slope group safety risk monitoring classification method, such as Figure 1 As shown, one of the preferred embodiments includes the following steps: S100: Preliminary classification of the slope group based on the historical data of the slope group.

[0032] S200: Based on the results of the initial classification and the factors affecting the slope risk, a rough classification result of the comprehensive evaluation of each slope is obtained by solving the clamping process; the rough classification result of the comprehensive evaluation of each slope is divided into stable slopes, unstable slopes and unstable slopes.

[0033] S300: Refined classification of the unstable slopes in step S200 is performed in combination with the InSAR data of the regional slope group, and the identification results of the refined classification are divided into key hidden danger slopes and general hidden danger slopes.

[0034] It is understandable that the current slope classification method is relatively simple and cannot effectively cover the slope groups in the region. However, this embodiment can maximize the use of data and clarify the attributes of the slope by roughly classifying the historical data of the slope groups in the region, and then combining the information accumulated in the early stage. The state of the slope is judged through comprehensive risk assessment, and stable slopes are regularly inspected to maintain their good condition; repair measures are implemented for unstable slopes to ensure that they can restore stability; unstable slopes can be further subdivided into key hidden danger slopes and general hidden danger slopes, and key hidden danger slopes are monitored, and general hidden danger slopes are regularly inspected. Ultimately, a comprehensive classification of slope groups in the region can be achieved to ensure that all types of slopes have corresponding safety protection measures.

[0035] More specifically, this embodiment can more accurately identify potential risks through systematic classification of slopes, take timely measures to reduce the risk of landslides and collapses to provide slope safety. For classified slopes, resources and manpower can be reasonably allocated according to the classification, and efforts can be concentrated on solving key hidden danger slope problems, improving management efficiency to achieve resource optimization. The risk assessment model based on slope classification can realize dynamic monitoring, timely update of slope status, and ensure the real-time and accuracy of information. Through comprehensive classification of slopes, data support can be provided for the formulation of scientific and reasonable management and repair plans, thereby improving the effectiveness of decision-making; and systematic slope classification helps to establish a long-term monitoring and management mechanism to achieve sustainable development.

[0036] In this embodiment, the historical data of the slope group in the region mainly includes slope geology, topographic information, historical landslide event records, monitoring data, rainfall and meteorological data, construction and mining history, maintenance and reinforcement records, and slope management and risk assessment reports. These data are helpful for analyzing the stability trend and potential risks of the slopes, and for the initial classification of the slopes; for ease of understanding, the initial classification of the slope group can be described in detail below.

[0037] Specifically, Figure 1 and Figure 2 As shown, in step S100, based on the historical data of the slope group in the region, the initial classification of the slope group mainly includes lithology classification, height classification and slope classification. According to lithology classification, each slope can be divided into rock slope and soil slope; according to height classification, each slope can be divided into low slope, medium-high slope, high slope and super-high slope; according to slope classification, each slope can be divided into gentle slope, medium slope, steep slope, sharp slope and reverse slope.

[0038] It should be known that the specific classification standard for the height classification of the slope is related to the lithology of the slope, and the specific classification standard can be set according to the actual needs of those skilled in the art. For ease of understanding, a detailed description will be given below using a specific example.

[0039] 1. For rock slopes.

[0040] If the height of the slope h≤8 meters, the slope can be identified as a low slope; if the height of the slope h>8 meters & h≤15 meters, the slope can be identified as a medium-high slope; if the height of the slope h>15 meters & h≤30 meters, the slope can be identified as a high slope; if the height of the slope h>30 meters, the slope can be identified as a superhigh slope.

[0041] 2. For soil slopes.

[0042] If the height of the slope h≤5 meters, the slope can be identified as a low slope; if the height of the slope h>5 meters & h≤10 meters, the slope can be identified as a medium-high slope; if the height of the slope h>10 meters & h≤15 meters, the slope can be identified as a high slope; if the height of the slope h>15 meters, the slope can be identified as a superhigh slope.

[0043] It should also be known that the standards for slope classification of rock slopes and soil slopes are consistent, and the specific classification standards can be set according to the actual needs of those skilled in the art. For ease of understanding, a detailed description will be given below using an example.

[0044] If the slope of the slope α≤15°, the slope can be determined as a gentle slope; if the slope of the slope α>15°&α≤30°, the slope can be determined as a medium slope; if the slope of the slope α>30°&α≤60°, the slope can be determined as a steep slope; if the slope of the slope α>60°&α≤90°, the slope can be determined as a steep slope; if the slope of the slope α>90°, the slope can be determined as a reverse slope.

[0045] In this embodiment, after the initial classification of the slope groups in the region is completed, the slopes in the region can be roughly classified in all aspects in combination with the slope risk influencing factors. There are many ways to roughly classify the slope groups in the region. For ease of understanding, Figure 3 As shown, the following will describe in detail how to solve the clamping process by combining the TOPSIS algorithm and the AISM algorithm, which specifically includes the following steps: S210: According to the factors affecting slope risk, the historical data of the slope group are reduced in dimension and reasonable evaluation indicators are selected.

[0046] S220: Collect the performance data of each slope under different evaluation indicators to form an original decision matrix and perform preprocessing.

[0047] S230: Standardize the preprocessed original decision matrix to obtain a distance decision matrix.

[0048] S240: Calculate the similarity index of each slope according to the result of the normalization process and sort them to obtain a closeness matrix.

[0049] S250: Generate an adversarial topology hierarchy diagram corresponding to each matrix in steps S220 to S240 through the AISM algorithm, and analyze the generated adversarial topology hierarchy diagram to obtain a comprehensive evaluation result of each slope.

[0050] It is understandable that the TOPSIS algorithm, also known as the superiority and inferiority solution distance algorithm, is a comprehensive evaluation method based on raw data. The specific implementation process is well known to those skilled in the art, so it is not elaborated in detail here. The TOPSIS algorithm is applied to the present application, and the superiority and inferiority evaluation can be performed according to the specific performance data of the slope group corresponding to each evaluation index in the present embodiment. Specifically, the TOPSIS algorithm can judge the superiority and inferiority of the same evaluation index of each slope, and then combine the superiority and inferiority judgments of each evaluation index to obtain the overall superiority and inferiority of the slope. However, there is a relationship between the evaluation indexes of the slope. If the TOPSIS algorithm is simply used to judge the superiority and inferiority of each evaluation index of the slope, it is difficult to achieve a comprehensive evaluation of all slopes in the slope group. Therefore, the present embodiment integrates the AISM algorithm on the basis of the TOPSIS algorithm. The AISM algorithm is also called the adversarial interpretation decoupling stock model algorithm. On the basis of the classic explanatory structure model, it incorporates the game confrontation idea in the generative adversarial network. The specific implementation method is a known technology for those skilled in the art, so it is not elaborated in detail here. The AISM algorithm can construct a relationship matrix according to the evaluation index corresponding to each slope, and generate a corresponding topological hierarchical diagram to display the hierarchical relationship between the evaluation indexes corresponding to the slope. That is, in this embodiment, the TOPSIS algorithm can judge the quality of each slope according to the evaluation index, and the AISM algorithm can construct a hierarchical relationship between the evaluation indexes corresponding to the slope, thereby realizing a comprehensive evaluation of all slopes in the slope group.

[0051] It should be known that in step S210, the factors affecting slope risk mainly include hydrological conditions, human activities and historical records, etc. Hydrological conditions include groundwater level and rainfall, etc. Human activities include construction and mining conditions, etc. Historical records include landslide records, monitoring records and assessment records, etc. Then, combined with the initial classification results of the historical data of the slope group and the factors affecting slope risk, the evaluation indicators in step S210 can be selected to include geological conditions, hydrological conditions, slope geometry, human activities and historical records. Among them, the geological conditions can be obtained through the lithology classification results of the slope group, and the slope geometry can be obtained through the height classification and slope classification results of the slope group.

[0052] Specifically, in step S220, according to the above selected evaluation index, after collecting the performance data of each slope under different evaluation indicators, a parameter table can be constructed first to facilitate the establishment of the subsequent decision matrix. When constructing the parameter table, the data of all slopes in the slope group can be listed, or the slopes with the same or similar performance data can be classified according to the above evaluation index, that is, the number of slopes in each category is at least one. In order to reduce the amount of calculation, the classification statistics method is preferably adopted in this embodiment, then all slopes in the slope group can be divided into m categories, and the slopes of category m can be marked as slope 1, slope 2, ..., slope m in turn; then the parameter table used to construct the original decision matrix is ​​as follows:

[0053] Among them, the k evaluation indicators can be marked as evaluation indicator 1 to evaluation indicator k in sequence, and the value of k is 5; then the representation data of any evaluation indicator j corresponding to any slope i is x ij ; The value of i is {1, 2, ..., m}, and the value of j is {1, 2, ..., k}.

[0054] It should be known that, according to the above parameter table, an original decision matrix with m rows and 5 columns can be constructed; after obtaining the original decision matrix, the original decision matrix needs to be preprocessed to ensure that each column of the original decision matrix must be a positive indicator or a negative indicator before standardization. After preprocessing the original decision matrix, a normalized matrix can be obtained.

[0055] In this embodiment, the normalization matrix is ​​normalized by calculating the maximum normalization and minimum normalization of the performance data of the slope under each evaluation index to obtain the required distance decision matrix. There are many specific calculation formulas for the maximum normalization and the minimum normalization. For the convenience of understanding, a detailed description will be given below through a specific example.

[0056] Specifically, the maximum normalized value corresponding to any slope i is and minimal normalization The calculation formula is as follows: .

[0057] .

[0058] in, represents the optimal value of the evaluation index j, Represents the worst value of evaluation index j.

[0059] It should be known that the dimensions of the performance data corresponding to each evaluation index of the slope are different. By standardizing the normalized matrix, a dimensionless distance decision matrix can be obtained. There are two distance decision matrices, corresponding to the maximum normalization and the minimum normalization respectively; each distance decision matrix is ​​an m-row single-column matrix.

[0060] In this embodiment, after obtaining the distance decision matrix based on maximum normalization and minimum normalization, step S240 obtains the corresponding closeness matrix by calculating the similarity index of each evaluation index of the slope. There are many specific calculation methods for the similarity index. For the convenience of understanding, a detailed description will be given below through one of the specific examples.

[0061] Specifically, the similarity index C corresponding to any slope i is i The calculation formula is as follows: .

[0062] It should be known that the closeness matrix is ​​a fusion of two distance decision matrices, so the number of closeness matrices is one, and it is still an m-row single-column matrix.

[0063] It is understandable that the process from the original decision matrix to the closeness matrix is ​​a dimensionality reduction process. The AISM algorithm can generate a corresponding adversarial topology hierarchy diagram in each dimensionality reduction process. Specifically, Figure 3As shown, for the original decision matrix O, the AISM algorithm can use the partial order to generate the relationship matrix A1 of the original decision matrix O, and then generate the corresponding adversarial topology hierarchical graph based on the obtained relationship matrix A1. For the normalized matrix K, the AISM algorithm can use the partial order to generate the relationship matrix A2 of the normalized matrix K, and then generate the corresponding adversarial topology hierarchical graph based on the obtained relationship matrix A2. For the distance decision matrix D, the AISM algorithm can use the partial order to generate the relationship matrix A3 of the distance decision matrix D, and then generate the corresponding adversarial topology hierarchical graph based on the obtained relationship matrix A3. For the proximity matrix C, the AISM algorithm can use the partial order to generate the relationship matrix A4 of the proximity matrix C, and then generate the corresponding adversarial topology hierarchical graph based on the obtained relationship matrix A4. Finally, the stability of the slope is comprehensively evaluated based on the four obtained adversarial topology hierarchical graphs.

[0064] In this embodiment, the comprehensive evaluation result of slope stability obtained by jointly solving the clamping process through the TOPSIS algorithm and the AISM algorithm is the coarse classification result of the slope groups in the region. After completing the coarse classification, it is necessary to carry out refined classification of the slope groups in the region for quantitative classification. Under the premise of coarse classification, the slopes are divided into stable slopes, unstable slopes and unstable slopes; through refined classification, the previous coarse classification can be checked for omissions and deficiencies, and further, the refined classification can be used to subdivide unstable slopes into key hidden danger slopes and general hidden danger slopes. Specifically, Figure 1 As shown, step S300 includes the following process: S310: Obtain high-resolution InSAR data of the slope group and perform preprocessing.

[0065] S320: Compare the displacement data at different time points based on the preprocessed data to obtain the deformation trend of the slope.

[0066] S330: According to the obtained deformation trend, deformation characteristics of the slope are identified by feature extraction, where the deformation characteristics include displacement mode, displacement speed and displacement amplitude.

[0067] S340: Perform feature analysis based on the extracted deformation features and the geological environment characteristics of the slope.

[0068] S350: Perform fine classification using a machine learning model based on the results of feature analysis.

[0069] It should be known that InSAR data is also called synthetic aperture radar data. The preprocessing process of the acquired synthetic aperture radar data includes registration and noise removal. Registration is to align images acquired at different times in time and space to ensure the accuracy of subsequent analysis; noise removal is to eliminate the noise caused by atmospheric interference and other non-surface deformation through filtering technology, such as multi-scale filtering.

[0070] Another aspect of the present application provides a slope group safety risk assessment system, which can use the above-mentioned slope group safety risk monitoring classification method to perform a dynamic assessment of slope safety in a region. Figure 1 and Figure 4 As shown, one preferred embodiment includes a data cleaning module, a feature module and a model generation module. The data cleaning module can clean the input slope group historical data. The feature module can perform feature extraction and feature selection on the cleaned data. The model generation module can perform model training, model testing and debugging and tuning on the data output by the feature module to determine the final slope risk assessment model. The slope risk assessment model can output the stability score of the slope based on the input slope real-time monitoring data.

[0071] It is understandable that in order to ensure the working safety of the slope groups in the area, it is necessary to conduct periodic dynamic assessments of the slope groups in the area. Specifically, it is to periodically monitor and process the InSAR data of the slope groups in the area, and establish a slope risk assessment model MSD{[m, n]} based on the aforementioned slope group safety risk monitoring classification method, where n represents the number of layers of the model. After obtaining the slope risk assessment model, the slope risk assessment model is updated using the latest monitoring results. As the monitoring data accumulates, the slope risk assessment model is regularly adjusted and optimized to provide timely feedback on the safety status of the slope. The model parameters are corrected based on the latest InSAR data, and the comprehensive risk value is recalculated to ensure rapid feedback when the risk changes, thereby making an effective response.

[0072] Regular risk assessment and updating of slopes can not only detect potential problems in a timely manner, but also provide a scientific basis for subsequent management decisions. Through dynamic monitoring and evaluation, managers can better formulate emergency plans to ensure that appropriate preventive measures are taken when the risk of slope instability increases to protect the safety of personnel and property. In summary, combining slope classification, periodic InSAR monitoring and the establishment and updating of slope risk assessment models can form a scientific slope management system, improve monitoring efficiency and accuracy, and provide strong technical support for regional safety prevention.

[0073] In this embodiment, the cleaning methods of the data cleaning module include removing fields, filling missing values, and reacquiring data. For fields with low importance and high missing rates in the historical data of the slope group, a cleaning method of removing fields is adopted. For fields with high importance and low missing rates in the historical data of the slope group, a cleaning method of filling missing values ​​is adopted. For fields with high importance and missing rates in the historical data of the slope group, a cleaning method of reacquiring data is adopted. For fields with low importance and missing rates in the historical data of the slope group, no processing is performed or they are simply filled.

[0074] In this embodiment, feature extraction of the feature module includes the following processes: calculating the statistics of historical data, including mean error, variance, maximum value and minimum value; extracting trend characteristics, seasonal characteristics and periodic characteristics of historical data, and using Fourier transform to analyze the spectral characteristics of monitoring data to identify potential vibration modes; performing morphological analysis on displacement or stress data to extract feature points; feature selection of the feature module includes the following process: based on the results of feature extraction, identifying the features that have the greatest impact on slope stability prediction.

[0075] In the present embodiment, the process of generating a slope risk assessment model by a model generation module is a well-known technology for those skilled in the art. For ease of understanding, the model generation process will be briefly described below. Specifically, a decision tree algorithm is selected for model training, and the selected features are input into the model for training to generate a preliminary model. Model performance is evaluated using methods such as cross-validation, and the performance of the model on the training set and the validation set is checked. According to the evaluation results, the deficiencies of the model are identified to adjust the hyperparameters, and different parameter combinations are tried to optimize the model performance. Hyperparameter optimization can be performed using methods such as grid search or random search. According to the evaluation results and the tuned parameters, the above steps are repeated until a satisfactory model is obtained, and the results of each experiment are recorded to compare the performance of different models. The final model and its parameters are determined, and necessary verification is performed.

[0076] In this embodiment, the slope group safety risk assessment system also includes a display platform and an early warning module. The slope group safety risk assessment system can draw a GIS map of the slope groups in the area, and establish a corresponding data file for each slope in the slope group; the display platform can display the slope data, and the displayed content includes real-time slope data, historical data and risk assessment results. The slope group safety risk assessment system can set thresholds for various risk indicators based on historical data and expert advice, and different early warning mechanisms should be set for different risk levels. When the real-time data of the slope exceeds the set threshold, the early warning module automatically triggers an alarm. Notify relevant management personnel through SMS, email or APP push; set up sound and light alarms on site, and work with the early warning module to promptly remind passers-by and vehicles.

[0077] The above describes the basic principles, main features and advantages of the present application. Those skilled in the art should understand that the present application is not limited by the above embodiments, and the above embodiments and the specification only describe the principles of the present application. The present application may have various changes and improvements without departing from the spirit and scope of the present application, and these changes and improvements fall within the scope of the present application for which protection is sought. The scope of protection claimed by the present application is defined by the attached claims and their equivalents.

Claims

1. A slope group safety risk monitoring classification method, characterized in that: The steps include: S100: Preliminary classification of slope groups based on historical data of slope groups; S200: Based on the results of the initial classification and the factors affecting the slope risk, a rough classification result of the comprehensive evaluation of each slope is obtained by solving the clamping process; the rough classification result of the comprehensive evaluation of each slope is divided into stable slopes, unstable slopes and unstable slopes; S300: Refined classification of the unstable slopes in step S200 is performed in combination with the InSAR data of the regional slope group, and the identification results of the refined classification are divided into key hidden danger slopes and general hidden danger slopes.

2. The slope group safety risk monitoring classification method according to claim 1 is characterized in that: In step S100, the initial classification of the slope group includes lithology classification, height classification and slope classification; According to the lithology classification, each slope is divided into rock slope and soil slope; According to the height classification, each slope is divided into low slope, medium-high slope, high slope and super-high slope; According to the slope classification, each slope is divided into gentle slope, medium slope, steep slope, sharp slope and reverse slope.

3. The slope group safety risk monitoring classification method according to claim 1 is characterized in that: In step S200, the clamping process is solved by jointly solving the TOPSIS algorithm and the AISM algorithm, including the following steps: S210: According to the factors affecting slope risk, the historical data of the slope group is reduced in dimension and reasonable evaluation indicators are selected; S220: collecting performance data of each slope under different evaluation indicators to form an original decision matrix and perform preprocessing; S230: performing standardization processing on the preprocessed original decision matrix to obtain a distance decision matrix; S240: Calculate the similarity index of each slope according to the result of the normalization process and sort them to obtain a closeness matrix; S250: Generate an adversarial topology hierarchy diagram corresponding to each matrix in steps S220 to S240 through the AISM algorithm, and analyze the generated adversarial topology hierarchy diagram to obtain a comprehensive evaluation result of each slope.

4. The slope group safety risk monitoring classification method according to claim 3 is characterized in that: The evaluation indicators selected in step S210 include geological conditions, hydrological conditions, slope geometry, human activities and historical records.

5. The slope group safety risk monitoring and classification method according to claim 3 is characterized in that: The k evaluation indicators selected in step S210 are marked as evaluation indicator 1 to evaluation indicator k in sequence; In step S230, the maximum normalization and minimum normalization of the performance data of the slope under each evaluation index are calculated to perform standardization processing; the maximum normalization corresponding to any slope i and minimal normalization The calculation formula is as follows: ; ; Among them, x ij represents the performance data of slope i corresponding to any evaluation index j, represents the optimal value of the evaluation index j, Represents the worst value of evaluation index j.

6. The slope group safety risk monitoring and classification method according to claim 5 is characterized in that: In step S240, the similarity index C corresponding to any slope i i The calculation formula is as follows: 。 7. The slope group safety risk monitoring and classification method according to any one of claims 1 to 6, characterized in that: Step S300 includes the following process: S310: Acquire high-resolution InSAR data of the slope group and perform preprocessing; S320: Compare displacement data at different time points according to the preprocessed data to obtain the deformation trend of the slope; S330: identifying deformation features of the slope by feature extraction according to the obtained deformation trend, where the deformation features include displacement mode, displacement speed and displacement amplitude; S340: performing feature analysis based on the extracted deformation features combined with the geological environment features of the slope; S350: Perform fine classification using a machine learning model based on the results of feature analysis.

8. A slope group safety risk assessment system using the slope group safety risk monitoring and classification method according to any one of claims 1 to 7, characterized in that: include: Data cleaning module; the data cleaning module is suitable for cleaning the inputted historical data of the slope group; Feature module; The feature module is suitable for performing feature extraction and feature selection on the cleaned data; as well as Model generation module; the model generation module is suitable for performing model training, model testing and debugging and tuning on the data output by the feature module to determine the final slope risk assessment model; The slope risk assessment model is suitable for outputting a slope stability score based on input real-time slope monitoring data.

9. The slope group safety risk assessment system according to claim 8, characterized in that: The cleaning methods of the data cleaning module include removing fields, filling missing values ​​and reacquiring data; For the fields with low importance and high missing rate in the historical data of the slope group, the cleaning method of removing fields is adopted; For the fields with high importance and low missing rate in the historical data of the slope group, the cleaning method of filling missing values ​​is adopted; For the fields with high importance and missing rate in the historical data of the slope group, a cleaning method of re-acquiring data is adopted; The fields with low importance and missing rate in the historical data of slope groups are not processed or simply filled.

10. The slope group safety risk assessment system according to claim 8, characterized in that: The feature extraction of the feature module includes the following processes: calculating the statistics of historical data, including mean error, variance, maximum value and minimum value; extracting the trend characteristics, seasonal characteristics and periodic characteristics of historical data, and using Fourier transform to analyze the spectrum characteristics of monitoring data to identify potential vibration modes; performing morphological analysis on displacement or stress data to extract feature points; The feature selection of the feature module includes the following process: based on the result of feature selection, identifying the features that have the greatest impact on slope stability prediction.

Citation Information

Patent Citations

  • Slope stability monitoring and early warning method and system and medium

    CN112526104A

  • Comprehensive evaluation method and system of improved TOPSIS algorithm based on cloud edge cooperative driving

    CN115630739A

  • High and steep slope geological disaster early warning system

    CN118095869A

  • Slope group safety risk monitoring and evaluating method and management system

    CN118503813A

Cited By

  • Large cantilever structure construction safety risk assessment system

    CN119180507A

  • Slope group stability evaluation method, system and equipment and storage medium

    CN120338489A

  • A method, system, device, and storage medium for assessing the stability of slope groups.

    CN120338489B

  • Method, system and equipment for intelligently transferring side slope monitoring resources in side slope group and medium

    CN120450326A

  • Point-shaped landslide and debris flow potential degree self-adaptive monitoring and early warning method

    CN120808577A