Slope early warning method and system based on multi-coupling data analysis

Abstract

The present invention discloses a slope early warning method and system based on multi-coupling data analysis, which relates to the field of coupled data processing technology, including selecting slopes similar to target slopes from other slopes through geological engineering related data, and comprehensively screening out slopes similar to target slopes by combining stable geological engineering related data and unstable geological engineering related data, so as to provide data support for the coupling state body of the subsequent unstable slope. The stability characteristics and coupling relationship of the slope in an unstable state are taken together as the coupling state body of the unstable slope, which accurately describes the stability characteristics of the slope in an unstable state and the coupling relationship between the characteristics. Based on the coupling state body of the current slope and the coupling state body of the unstable slope, the target slope is given a stability early warning, and the real-time coupling state body of the current slope is compared with the coupling state body of the unstable slope to judge the stability of the slope, thereby improving the timeliness and accuracy of the slope early warning.

Description

Technical Field

[0001] The present invention relates to the technical field of coupled data processing, and in particular to a slope early warning method and system based on multi-coupled data analysis. Background Art

[0002] The background technology of slope early warning scheme stems from the important position of slope engineering in civil engineering, which is directly related to the safety of people's lives and property and the economic development of the country. However, due to the complex geological environment, variable natural factors and the influence of human activities, the stability and safety of slope engineering face severe challenges. Traditional slope engineering monitoring methods have problems such as limited monitoring range, low monitoring frequency and high maintenance cost. Therefore, it is particularly important to develop an intelligent monitoring and early warning system that can monitor slope changes in real time, automatically and remotely, and warn of slope disasters in advance. The system integrates high-precision sensors, data analysis algorithms and remote communication technology to realize real-time monitoring of parameters such as slope displacement, cracks, and groundwater levels, and uses multivariate data fusion technology to conduct comprehensive analysis and early warning of monitoring data, providing scientific and effective decision support for decision makers.

[0003] In the prior art, slope warning is often performed only based on the numerical value of monitoring data, without considering the complex interaction relationship between coupled data, resulting in poor timeliness and accuracy of slope warning, and unable to effectively ensure slope safety.

[0004] Therefore, how to improve the timeliness and accuracy of slope warning is a technical problem that needs to be solved. Summary of the invention

[0005] The purpose of the present invention is to solve the problem of poor timeliness and accuracy of slope warning in the prior art, and propose a slope warning method based on multi-coupling data analysis, which includes:

[0006] Collect the previous multi-source data of the target slope and other slopes, select the geological engineering related data from the multi-source data, and select the slopes similar to the target slope from other slopes through the geological engineering related data, and record them as similar slopes;

[0007] Slope instability data are intercepted from previous multi-source data of target slopes and similar slopes, slope stability features are extracted based on the slope instability data, the coupling relationship between slope stability features is analyzed, and a coupling state body of an unstable slope is generated based on the slope stability features and the coupling relationship between slope stability features;

[0008] Acquire real-time multi-source data of the target slope, extract real-time slope stability features from the real-time multi-source data, analyze the coupling relationship between the real-time slope stability features, and evaluate the coupling relationship, and generate a coupling state body of the current slope according to the real-time slope stability features, the coupling relationship between the real-time slope stability features, and the evaluation result of the coupling relationship;

[0009] Based on the coupled state body of the current slope and the coupled state body of the unstable slope, a stability warning is given to the target slope.

[0010] In some embodiments of the present application, slopes similar to the target slope are selected from other slopes through geological engineering related data and recorded as similar slopes, including:

[0011] Geological engineering related data include geological condition data, engineering type data, slope geometric characteristics data, environmental factor data and monitoring data;

[0012] According to geological condition data, engineering type data, and slope geometric characteristic data, initial other slopes whose three types of data are similar to the target slope are selected from other slopes;

[0013] Calculate the average value range and change rate of each parameter in the environmental factor data and monitoring data of each slope in the initial other slopes and the target slope, and calculate the similarity between the initial other slopes and the target slope through the average value range and change rate of each parameter in the environmental factor data and monitoring data of each slope; ;

[0014] in, For the The similarity between the initial other slopes and the target slope, are the conversion coefficients for the average value range and change rate of each parameter in the environmental factor data and monitoring data, Indicates The number of parameters in the intersection of the set of the average value range of each parameter in the initial other slopes and the set of the average value range of each parameter in the target slope, For the The intersection of the initial other slopes The combined weights of the average values ​​of the parameters, For the The intersection of the initial other slopes and the target slope The overlap of the range of mean values ​​of the parameters, is the environmental factor data under the target slope and the number of parameters in the monitoring data, For the The combined weight of the parameter change rate, For the The initial other slopes and the target slope The absolute value of the difference between the rates of change of the parameters;

[0015] Based on the similarity between the initial other slopes and the target slope, the initial other slopes are screened to obtain slopes similar to the target slope, which are recorded as similar slopes.

[0016] In some embodiments of the present application, slope instability data is extracted from previous multi-source data of target slope and similar slopes, including:

[0017] According to the timestamp of the slope accident, the previous multi-source data of the target slope and similar slopes are divided into multi-source data under slope accidents and multi-source data under non-slope accidents. The multi-source data under slope accidents are multi-source data between a period of time before the slope accident and a period of time after the slope accident.

[0018] The slope instability data are extracted from the multi-source data under slope accidents and the multi-source data under non-slope accidents.

[0019] In some embodiments of the present application, the slope instability data is extracted from the multi-source data under the slope accident and the multi-source data under the non-slope accident, including:

[0020] For the multi-source data under the slope accident, the multi-source data under the slope accident with significant changes before and after the slope accident is screened by the first preset condition, and the first slope instability data is intercepted;

[0021] For the multi-source data under non-slope accidents, outlier and abnormal trend analysis is performed on the multi-source data under non-slope accidents, data representing slope instability is screened out through a second preset condition, and second slope instability data is intercepted;

[0022] The first slope instability data and the second slope instability data are taken together as slope instability data.

[0023] In some embodiments of the present application, the coupling relationship between slope stability features is analyzed, including:

[0024] Identify and determine the interaction relationship between slope stability features, and combine the slope stability features through the interaction relationship to obtain feature interaction terms;

[0025] The slope stability coupling model is trained according to the characteristic interaction terms, and the coupling relationship between the slope stability characteristics in the unstable state of the slope is described by the unstable state slope stability coupling model.

[0026] In some embodiments of the present application, a coupling state body of an unstable slope is generated according to the slope stability feature and the coupling relationship between the slope stability features, including:

[0027] The coupled state body of the unstable slope includes two parts: the unstable state slope stability characteristics and the unstable state slope stability coupling model. The unstable state slope stability characteristics describe the stability characteristics of each slope under the unstable state of the slope.

[0028] In some embodiments of the present application, the coupling relationship is evaluated, and a coupling state body of the current slope is generated according to the real-time slope stability characteristics, the coupling relationship between the real-time slope stability characteristics, and the evaluation result of the coupling relationship, including:

[0029] The real-time slope stability coupling model describes the coupling relationship between slope stability characteristics under the current state of the slope, simulates different external disturbance conditions, inputs different external disturbance conditions into the real-time slope stability coupling model, obtains the change of output characteristics, analyzes the correlation and response degree between the change of output characteristics and external disturbance, and thus calculates the coupling sensitivity coefficient of the coupling relationship;

[0030] The current slope coupling state body includes three parts: real-time slope stability characteristics, real-time slope stability coupling model and coupling sensitivity coefficient.

[0031] In some embodiments of the present application, a stability warning is performed on a target slope based on a coupling state body of a current slope and a coupling state body of an unstable slope, including:

[0032] The matching degree between the real-time slope stability characteristics and the unstable state slope stability characteristics, and the matching degree between the real-time slope stability coupling model and the unstable state slope stability coupling model are calculated respectively, which are recorded as the first matching degree and the second matching degree respectively, and the overall matching degree is determined by combining the first matching degree, the second matching degree and the coupling sensitivity coefficient; ;

[0033] in, For the overall matching, are the weights of slope stability characteristics and slope stability coupling model, The matching degree between the real-time slope stability characteristics and the unstable state slope stability characteristics, and the matching degree between the real-time slope stability coupling model and the unstable state slope stability coupling model are calculated respectively. is the coupling sensitivity coefficient, is a preset constant;

[0034] If the overall matching degree is greater than the preset overall matching degree threshold, a stability warning is issued for the target slope;

[0035] Otherwise, the evolution probability between the coupling state body of the current slope and the coupling state body of the unstable slope is calculated, and the evolution probability is used to determine whether to issue a stability warning for the target slope.

[0036] Correspondingly, the present application also provides a slope warning system based on multi-coupling data analysis, including:

[0037] The screening module is used to collect the previous multi-source data of the target slope and other slopes, screen out the geological engineering related data from the multi-source data, and select the slopes similar to the target slope from other slopes through the geological engineering related data, and record them as similar slopes;

[0038] A construction module is used to extract slope instability data from previous multi-source data of target slopes and similar slopes, extract slope stability features based on the slope instability data, analyze the coupling relationship between slope stability features, and generate a coupling state body of an unstable slope based on the slope stability features and the coupling relationship between slope stability features;

[0039] A real-time module is used to obtain real-time multi-source data of the target slope, extract real-time slope stability features from the real-time multi-source data, analyze the coupling relationship between the real-time slope stability features, and evaluate the coupling relationship, and generate a coupling state body of the current slope based on the real-time slope stability features, the coupling relationship between the real-time slope stability features, and the evaluation result of the coupling relationship;

[0040] The early warning module is used to provide stability early warning for the target slope based on the coupling state body of the current slope and the coupling state body of the unstable slope.

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] 1. Select slopes similar to the target slope from other slopes through geological engineering related data, and comprehensively select slopes similar to the target slope by combining stable geological engineering related data and unstable geological engineering related data, so as to provide sufficient data support for the subsequent construction of the coupling state body of unstable slopes. And generate the coupling state body of unstable slopes according to the slope stability characteristics and the coupling relationship between the slope stability characteristics, and use the slope stability characteristics and coupling relationship under unstable state as the coupling state body of unstable slope, which accurately describes the slope stability characteristics under unstable state and the coupling relationship between the characteristics.

[0043] 2. Based on the coupling state body of the current slope and the coupling state body of the unstable slope, the target slope is given a stability warning. The real-time coupling state body of the current slope is compared with the coupling state body of the unstable slope to judge the stability of the slope, improve the timeliness and accuracy of the slope warning, effectively ensure the safety of the slope, and take preventive measures in time. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a flow chart of the slope early warning method based on multi-coupling data analysis proposed by the present invention;

[0045] Figure 2 This is a schematic diagram of the structure of the slope early warning system based on multi-coupling data analysis proposed by the present invention. DETAILED DESCRIPTION

[0046] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0047] Reference Figure 1 ,The slope early warning method based on multi-coupling data analysis includes the following steps:

[0048] Step S101, collect the previous multi-source data of the target slope and other slopes, screen out the geological engineering related data from the multi-source data, and select the slopes similar to the target slope from other slopes through the geological engineering related data, and record them as similar slopes.

[0049] In this embodiment, the previous multi-source data includes collecting as comprehensive multi-source data as possible by using remote sensing technology (such as satellite images, drone aerial photography), geological exploration reports, meteorological records, historical monitoring data (such as displacement monitoring, groundwater level changes), human activity records, etc. The detailed data sets of the target slope and multiple candidate slopes include but are not limited to topography, geological structure, rock and soil properties, hydrological conditions, historical disaster records, etc.

[0050] It should be noted that in order to accurately and comprehensively construct the situation of the target slope under unstable state, it is necessary to avoid the low accuracy of subsequent model establishment due to the lack of previous multi-source data on the target slope.

[0051] In some embodiments of the present application, slopes similar to the target slope are selected from other slopes through geological engineering related data and recorded as similar slopes, including:

[0052] Geological engineering related data include geological condition data, engineering type data, slope geometric characteristics data, environmental factor data and monitoring data;

[0053] According to geological condition data, engineering type data, and slope geometric characteristic data, initial other slopes whose three types of data are similar to the target slope are selected from other slopes;

[0054] Calculate the average value range and change rate of each parameter in the environmental factor data and monitoring data of each slope in the initial other slopes and the target slope, and calculate the similarity between the initial other slopes and the target slope through the average value range and change rate of each parameter in the environmental factor data and monitoring data of each slope; ;

[0055] in, For the The similarity between the initial other slopes and the target slope, are the conversion coefficients for the average value range and change rate of each parameter in the environmental factor data and monitoring data, Indicates The number of parameters in the intersection of the set of the average value range of each parameter in the initial other slopes and the set of the average value range of each parameter in the target slope, For the The intersection of the initial other slopes The combined weights of the average values ​​of the parameters, For the The intersection of the initial other slopes and the target slope The overlap of the range of mean values ​​of the parameters, is the environmental factor data under the target slope and the number of parameters in the monitoring data, For the The combined weight of the parameter change rate, For the The initial other slopes and the target slope The absolute value of the difference between the rates of change of the parameters;

[0056] Based on the similarity between the initial other slopes and the target slope, the initial other slopes are screened to obtain slopes similar to the target slope, which are recorded as similar slopes.

[0057] In this embodiment, the geological condition data, engineering type data, and slope geometric feature data are relatively stable, and the change range may be low. Geological conditions, including rock and soil types, rock layer structures, and geological structures, are usually formed on a long time scale and are therefore relatively stable. These characteristics usually do not change significantly in a short period of time (such as a few years or decades). Engineering type data, such as excavation methods, support structures, and construction methods, once determined and implemented during the construction process, usually remain relatively stable. These data reflect the design and construction plan of the slope project and will not change frequently. The geometric characteristics of the slope, such as height, slope, and shape, usually remain relatively stable in the natural state. Unless affected by external forces (such as weathering, erosion, earthquakes) or human activities (such as excavation and filling), these characteristics will not change significantly. Environmental factor data and monitoring data (such as displacement rate, settlement, etc.) vary greatly, and environmental factors, such as rainfall, temperature, and seismic activity, are highly variable. These factors are affected by climate, seasons, geographical location, and geodynamic processes and often change.

[0058] In this embodiment, other slopes are preliminarily screened based on three relatively stable data, namely geological condition data, engineering type data, and slope geometric characteristic data, and then a secondary screening is performed on the results of the preliminary screening based on unstable environmental factor data and monitoring data.

[0059] In this embodiment, the similarity between the initial other slopes and the target slope is calculated by the average value range and change rate of each parameter in the environmental factor data of each slope and the monitoring data, and the similarity between the initial other slopes and the target slope is jointly described by combining the overlap degree of the average value range of the parameters and the difference in the change rate.

[0060] Step S102, extracting slope instability data from previous multi-source data of the target slope and similar slopes, extracting slope stability features based on the slope instability data, analyzing the coupling relationship between the slope stability features, and generating a coupling state body of the unstable slope based on the slope stability features and the coupling relationship between the slope stability features.

[0061] In this embodiment, characteristic data of unstable state are extracted from historical data of similar slopes and target slopes. The data of unstable state are defined as data before and after the slope accident and unstable data at other times.

[0062] In some embodiments of the present application, slope instability data is extracted from previous multi-source data of target slope and similar slopes, including:

[0063] According to the timestamp of the slope accident, the previous multi-source data of the target slope and similar slopes are divided into multi-source data under slope accidents and multi-source data under non-slope accidents. The multi-source data under slope accidents are multi-source data between a period of time before the slope accident and a period of time after the slope accident.

[0064] The slope instability data are extracted from the multi-source data under slope accidents and the multi-source data under non-slope accidents.

[0065] In some embodiments of the present application, the slope instability data is extracted from the multi-source data under the slope accident and the multi-source data under the non-slope accident, including:

[0066] For the multi-source data under the slope accident, the multi-source data under the slope accident with significant changes before and after the slope accident is screened by the first preset condition, and the first slope instability data is intercepted;

[0067] For the multi-source data under non-slope accidents, outlier and abnormal trend analysis is performed on the multi-source data under non-slope accidents, data representing slope instability is screened out through a second preset condition, and second slope instability data is intercepted;

[0068] The first slope instability data and the second slope instability data are taken together as slope instability data.

[0069] In this embodiment, for multi-source data under slope accidents, the focus is on identifying and screening out characteristic data directly related to slope instability, such as significant changes in parameters such as displacement, stress, and groundwater level. The collected data is cleaned and sorted to remove outliers and noise to ensure the accuracy and reliability of the data. The change trend of the data before and after the slope accident is analyzed, especially the parameters directly related to the stability of the slope. By drawing time series graphs, scatter plots and other charts, the change of the data is intuitively displayed. The first preset condition is the threshold of the difference corresponding to each parameter type, so as to judge whether the change is significant. This threshold can be set accordingly according to the type of slope accident. For multi-source data under non-slope accidents, the focus is on identifying and screening out characteristic data that may indicate slope instability, that is, those data sequences that also show abnormal or unstable trends during non-accident periods. Outlier detection: Statistical methods (such as box plots, Z-score, etc.) or machine learning algorithms (such as isolated forests, LOF, etc.) are used to detect outliers or abnormal sequences in the data. These anomalies may indicate changes in slope stability. Trend analysis: Use time series analysis methods (such as ARIMA model, exponential smoothing method, etc.) to analyze the long-term trend and cyclical changes of data. Pay special attention to data series that show unstable trends or cyclical fluctuations. The second preset condition is the unstable threshold corresponding to each parameter type.

[0070] In some embodiments of the present application, the coupling relationship between slope stability features is analyzed, including:

[0071] Identify and determine the interaction relationship between slope stability features, and combine the slope stability features through the interaction relationship to obtain feature interaction terms;

[0072] The slope stability coupling model is trained according to the characteristic interaction terms, and the coupling relationship between the slope stability characteristics in the unstable state of the slope is described by the unstable state slope stability coupling model.

[0073] In this example, first, we need to identify which characteristics are key factors in slope stability. These characteristics may include soil physical properties (such as internal friction angle, cohesion, water content, etc.), topography (such as slope, slope height, slope shape, etc.), hydrological conditions (such as groundwater level, rainfall intensity, etc.), and external loads (such as earthquake force, human activity impact, etc.).

[0074] The interactions between slope stability features include linear and nonlinear relationships, which can be identified and determined in the following ways:

[0075] Scatter plot: Plotting a scatter plot between features allows you to visually observe the relationship between them. If the data points are roughly distributed along a straight line, then there may be a linear relationship between the two features. Conversely, if the data points present a curve or other non-linear shape, there may be a non-linear relationship.

[0076] Residual plot: After fitting a linear regression model, analyze a graph of the residuals (the differences between the predicted and actual values). If the points in the residual plot are randomly distributed, then the linear model fits the data; if there is an obvious pattern (such as a curve or trend), then a nonlinear relationship may exist.

[0077] Feature interaction terms: In machine learning models, feature interaction terms usually refer to the product or other forms of combination between two or more features to capture the nonlinear relationship between them.

[0078] After considering the interactions between features, we can select a suitable model to establish a slope stability prediction model. As mentioned earlier, machine learning algorithms such as neural networks, support vector machines, and decision trees, or mathematical models based on physical mechanisms (such as finite element method, discrete element method, etc.) can be used. When selecting a model, it is necessary to consider the model's complexity, interpretability, computational efficiency, and ability to handle interactions between features.

[0079] Improve model accuracy:

[0080] By capturing the interactions between features, combined features and feature interaction terms can provide richer and more accurate information to the model, thereby improving the model's prediction accuracy.

[0081] Enhance model generalization ability:

[0082] Combining features and feature interactions helps the model learn deep patterns in the data, which enables the model to make better predictions when faced with new data, that is, it enhances the model's generalization ability.

[0083] Reduce model complexity:

[0084] In some cases, by combining features or constructing feature interactions, we can reduce the number of original features that need to be considered in the model, thereby simplifying the model structure and reducing the model complexity.

[0085] Step S103, acquiring real-time multi-source data of the target slope, extracting real-time slope stability features from the real-time multi-source data, analyzing the coupling relationship between the real-time slope stability features, and evaluating the coupling relationship, and generating a coupling state body of the current slope based on the real-time slope stability features, the coupling relationship between the real-time slope stability features, and the evaluation result of the coupling relationship.

[0086] In this embodiment, the coupling relationship is evaluated. The coupling sensitivity coefficient refers to the response degree of the slope stability characteristics to external disturbances. A high-sensitivity feature combination may cause the slope to change significantly under relatively small external disturbances, thereby increasing the risk of instability.

[0087] In some embodiments of the present application, a coupling state body of an unstable slope is generated according to the slope stability feature and the coupling relationship between the slope stability features, including:

[0088] The coupled state body of the unstable slope includes two parts: the unstable state slope stability characteristics and the unstable state slope stability coupling model. The unstable state slope stability characteristics describe the stability characteristics of each slope under the unstable state of the slope.

[0089] In some embodiments of the present application, the coupling relationship is evaluated, and a coupling state body of the current slope is generated according to the real-time slope stability characteristics, the coupling relationship between the real-time slope stability characteristics, and the evaluation result of the coupling relationship, including:

[0090] The real-time slope stability coupling model describes the coupling relationship between slope stability characteristics under the current state of the slope, simulates different external disturbance conditions, inputs different external disturbance conditions into the real-time slope stability coupling model, obtains the change of output characteristics, analyzes the correlation and response degree between the change of output characteristics and external disturbance, and thus calculates the coupling sensitivity coefficient of the coupling relationship;

[0091] The current slope coupling state body includes three parts: real-time slope stability characteristics, real-time slope stability coupling model and coupling sensitivity coefficient.

[0092] In this embodiment, the coupling sensitivity coefficient will affect the comparison result between the coupling state body of the current slope and the coupling state body of the unstable slope, so the coupling sensitivity coefficient is calculated to adaptively adjust the comparison result.

[0093] Step S104: providing a stability warning for the target slope based on the coupling state body of the current slope and the coupling state body of the unstable slope.

[0094] In this embodiment, the matching degree (similar situation) of the coupling state body of the current slope and the coupling state body of the unstable slope is compared to provide a stability warning for the target slope.

[0095] In some embodiments of the present application, a stability warning is performed on a target slope based on a coupling state body of a current slope and a coupling state body of an unstable slope, including:

[0096] The matching degree between the real-time slope stability characteristics and the unstable state slope stability characteristics, and the matching degree between the real-time slope stability coupling model and the unstable state slope stability coupling model are calculated respectively, which are recorded as the first matching degree and the second matching degree respectively, and the overall matching degree is determined by combining the first matching degree, the second matching degree and the coupling sensitivity coefficient; ;

[0097] in, For the overall matching, are the weights of slope stability characteristics and slope stability coupling model, The matching degree between the real-time slope stability characteristics and the unstable state slope stability characteristics, and the matching degree between the real-time slope stability coupling model and the unstable state slope stability coupling model are calculated respectively. is the coupling sensitivity coefficient, is a preset constant;

[0098] If the overall matching degree is greater than the preset overall matching degree threshold, a stability warning is issued for the target slope;

[0099] Otherwise, the evolution probability between the coupling state body of the current slope and the coupling state body of the unstable slope is calculated, and the evolution probability is used to determine whether to issue a stability warning for the target slope.

[0100] In this embodiment, It indicates that the matching between the slope stability characteristics and the slope stability coupling model is corrected by the coupling sensitivity coefficient of the current coupling relationship. The matching degree is determined by calculating the similarity between the slope stability characteristics and the slope stability coupling model.

[0101] In this embodiment, if the overall matching degree is not greater than the threshold, the evolution probability between the coupling state body of the current slope and the coupling state body of the unstable slope is calculated. This evolution probability describes the state transition probability from the coupling state body of the current slope to the coupling state body of the unstable slope, which can be achieved through a Markov chain.

[0102] Compared with the prior art, the present invention has the following beneficial effects:

[0103] 1. Select slopes similar to the target slope from other slopes through geological engineering related data, and comprehensively select slopes similar to the target slope by combining stable geological engineering related data and unstable geological engineering related data, so as to provide sufficient data support for the subsequent construction of the coupling state body of unstable slopes. And generate the coupling state body of unstable slopes according to the slope stability characteristics and the coupling relationship between the slope stability characteristics, and use the slope stability characteristics and coupling relationship under unstable state as the coupling state body of unstable slope, which accurately describes the slope stability characteristics under unstable state and the coupling relationship between the characteristics.

[0104] 2. Based on the coupling state body of the current slope and the coupling state body of the unstable slope, the target slope is given a stability warning. The real-time coupling state body of the current slope is compared with the coupling state body of the unstable slope to judge the stability of the slope, improve the timeliness and accuracy of the slope warning, effectively ensure the safety of the slope, and take preventive measures in time.

[0105] Correspondingly, the present application also provides a slope warning system based on multi-coupling data analysis, such as Figure 2 As shown, including,

[0106] The screening module is used to collect the previous multi-source data of the target slope and other slopes, screen out the geological engineering related data from the multi-source data, and select the slopes similar to the target slope from other slopes through the geological engineering related data, and record them as similar slopes;

[0107] A construction module is used to extract slope instability data from previous multi-source data of target slopes and similar slopes, extract slope stability features based on the slope instability data, analyze the coupling relationship between slope stability features, and generate a coupling state body of an unstable slope based on the slope stability features and the coupling relationship between slope stability features;

[0108] A real-time module is used to obtain real-time multi-source data of the target slope, extract real-time slope stability features from the real-time multi-source data, analyze the coupling relationship between the real-time slope stability features, and evaluate the coupling relationship, and generate a coupling state body of the current slope based on the real-time slope stability features, the coupling relationship between the real-time slope stability features, and the evaluation result of the coupling relationship;

[0109] The early warning module is used to provide stability early warning for the target slope based on the coupling state body of the current slope and the coupling state body of the unstable slope.

[0110] Through the description of the above implementation methods, those skilled in the art can clearly understand that the present invention can be implemented by hardware, or by software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each implementation scenario of the present invention.

[0111] Those skilled in the art will appreciate that the accompanying drawings are merely schematic diagrams of a preferred implementation scenario, and the modules or processes in the accompanying drawings are not necessarily required for implementing the present invention.

[0112] Those skilled in the art will appreciate that the modules in the system in the implementation scenario can be distributed in the system of the implementation scenario according to the implementation scenario description, or can be changed accordingly and located in one or more systems different from the implementation scenario. The modules in the above implementation scenario can be combined into one module, or can be further split into multiple submodules.

[0113] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. The slope early warning method based on multi-coupling data analysis is characterized by: include, Collect the previous multi-source data of the target slope and other slopes, select the geological engineering related data from the multi-source data, and select the slopes similar to the target slope from other slopes through the geological engineering related data, and record them as similar slopes; Slope instability data are intercepted from previous multi-source data of target slopes and similar slopes, slope stability features are extracted based on the slope instability data, the coupling relationship between slope stability features is analyzed, and a coupling state body of an unstable slope is generated based on the slope stability features and the coupling relationship between slope stability features; Acquire real-time multi-source data of the target slope, extract real-time slope stability features from the real-time multi-source data, analyze the coupling relationship between the real-time slope stability features, and evaluate the coupling relationship, and generate a coupling state body of the current slope according to the real-time slope stability features, the coupling relationship between the real-time slope stability features, and the evaluation result of the coupling relationship; Based on the coupled state body of the current slope and the coupled state body of the unstable slope, a stability warning is provided for the target slope; in, Through the geological engineering related data, the slopes similar to the target slope are selected from other slopes and recorded as similar slopes, including: Geological engineering related data include geological condition data, engineering type data, slope geometric characteristics data, environmental factor data and monitoring data; According to geological condition data, engineering type data, and slope geometric characteristic data, initial other slopes whose three types of data are similar to the target slope are selected from other slopes; Calculate the average value range and change rate of each parameter in the environmental factor data and monitoring data of each slope in the initial other slopes and the target slope, and calculate the similarity between the initial other slopes and the target slope through the average value range and change rate of each parameter in the environmental factor data and monitoring data of each slope; ; in, For the The similarity between the initial other slopes and the target slope, are the conversion coefficients for the average value range and change rate of each parameter in the environmental factor data and monitoring data, Indicates The number of parameters in the intersection of the set of the average value range of each parameter in the initial other slopes and the set of the average value range of each parameter in the target slope, For the The intersection of the initial other slopes The combined weights of the average values ​​of the parameters, For the The intersection of the initial other slopes and the target slope The overlap of the range of mean values ​​of the parameters, is the environmental factor data under the target slope and the number of parameters in the monitoring data, For the The combined weight of the parameter change rate, For the The initial other slopes and the target slope The absolute value of the difference between the rates of change of the parameters; Based on the similarity between the initial other slopes and the target slope, the initial other slopes are screened to obtain slopes similar to the target slope, which are recorded as similar slopes.

2. The slope early warning method based on multi-coupling data analysis according to claim 1 is characterized in that: Extract slope instability data from previous multi-source data of the target slope and similar slopes, including: According to the timestamp of the slope accident, the previous multi-source data of the target slope and similar slopes are divided into multi-source data under slope accidents and multi-source data under non-slope accidents. The multi-source data under slope accidents are multi-source data between a period of time before the slope accident and a period of time after the slope accident. The slope instability data are extracted from the multi-source data under slope accidents and the multi-source data under non-slope accidents.

3. The slope early warning method based on multi-coupling data analysis according to claim 2 is characterized in that: Extract slope instability data from multi-source data under slope accidents and multi-source data under non-slope accidents, including: For the multi-source data under the slope accident, the multi-source data under the slope accident with significant changes before and after the slope accident is screened by the first preset condition, and the first slope instability data is intercepted; For the multi-source data under non-slope accidents, outlier and abnormal trend analysis is performed on the multi-source data under non-slope accidents, data representing slope instability is screened out through a second preset condition, and second slope instability data is intercepted; The first slope instability data and the second slope instability data are taken together as slope instability data.

4. The slope early warning method based on multi-coupling data analysis according to claim 1 is characterized in that: Analyze the coupling relationship between slope stability characteristics, including, Identify and determine the interaction relationship between slope stability features, and combine the slope stability features through the interaction relationship to obtain feature interaction terms; The slope stability coupling model is trained according to the characteristic interaction terms, and the coupling relationship between the slope stability characteristics in the unstable state of the slope is described by the unstable state slope stability coupling model.

5. The slope early warning method based on multi-coupling data analysis according to claim 4 is characterized in that: And according to the slope stability characteristics and the coupling relationship between the slope stability characteristics, the coupling state body of the unstable slope is generated, including, The coupled state body of the unstable slope includes two parts: the unstable state slope stability characteristics and the unstable state slope stability coupling model. The unstable state slope stability characteristics describe the stability characteristics of each slope under the unstable state of the slope.

6. The slope early warning method based on multi-coupling data analysis according to claim 1 is characterized in that: The coupling relationship is evaluated, and the coupling state body of the current slope is generated according to the real-time slope stability characteristics, the coupling relationship between the real-time slope stability characteristics and the evaluation results of the coupling relationship, including: The real-time slope stability coupling model describes the coupling relationship between slope stability characteristics under the current state of the slope, simulates different external disturbance conditions, inputs different external disturbance conditions into the real-time slope stability coupling model, obtains the change of output characteristics, analyzes the correlation and response degree between the change of output characteristics and external disturbance, and thus calculates the coupling sensitivity coefficient of the coupling relationship; The current slope coupling state body includes three parts: real-time slope stability characteristics, real-time slope stability coupling model and coupling sensitivity coefficient.

7. The slope early warning method based on multi-coupling data analysis according to claim 5 or 6 is characterized in that: Based on the coupled state body of the current slope and the coupled state body of the unstable slope, the stability warning of the target slope is carried out, including: The matching degree between the real-time slope stability characteristics and the unstable state slope stability characteristics, and the matching degree between the real-time slope stability coupling model and the unstable state slope stability coupling model are calculated respectively, which are recorded as the first matching degree and the second matching degree respectively, and the overall matching degree is determined by combining the first matching degree, the second matching degree and the coupling sensitivity coefficient; ; in, For the overall matching, are the weights of slope stability characteristics and slope stability coupling model, The matching degree between the real-time slope stability characteristics and the unstable state slope stability characteristics, and the matching degree between the real-time slope stability coupling model and the unstable state slope stability coupling model are calculated respectively. is the coupling sensitivity coefficient, is a preset constant; If the overall matching degree is greater than the preset overall matching degree threshold, a stability warning is issued for the target slope; Otherwise, the evolution probability between the coupling state body of the current slope and the coupling state body of the unstable slope is calculated, and the evolution probability is used to determine whether to issue a stability warning for the target slope.

8. The slope warning system based on multi-coupling data analysis is characterized by: For implementing the slope early warning method based on multi-coupling data analysis as described in any one of claims 1 to 7, the system comprises: The screening module is used to collect the previous multi-source data of the target slope and other slopes, screen out the geological engineering related data from the multi-source data, and select the slopes similar to the target slope from other slopes through the geological engineering related data, and record them as similar slopes; A construction module is used to extract slope instability data from previous multi-source data of target slopes and similar slopes, extract slope stability features based on the slope instability data, analyze the coupling relationship between slope stability features, and generate a coupling state body of an unstable slope based on the slope stability features and the coupling relationship between slope stability features; A real-time module is used to obtain real-time multi-source data of the target slope, extract real-time slope stability features from the real-time multi-source data, analyze the coupling relationship between the real-time slope stability features, and evaluate the coupling relationship, and generate a coupling state body of the current slope based on the real-time slope stability features, the coupling relationship between the real-time slope stability features, and the evaluation result of the coupling relationship; The early warning module is used to provide stability early warning for the target slope based on the coupling state body of the current slope and the coupling state body of the unstable slope.

Images