A method and system for assessing the stability of soil and rock masses based on big data.

Abstract

This invention relates to the field of rock and soil stability assessment technology, specifically to a method and system for rock and soil stability assessment based on big data. It includes the following steps: S1, obtaining the geological range of the rock and soil mass to be assessed for stability, and simultaneously collecting relevant big data on the rock and soil mass based on this geological range; S2, dividing the geological range of the rock and soil mass into regions based on the geographical location and geological characteristics corresponding to the relevant big data; predicting rock and soil mass data by selecting factors based on correlation values, and combining the selected factor-predicted rock and soil mass data with time-predicted rock and soil mass data to perform a comprehensive stability score. The scoring process considers multiple dimensions, including basic stability score, time-predicted data weight, weight of changing factors, correlation correction coefficient between regions, and regional area weight, to calculate the final comprehensive stability score, which can objectively and accurately reflect the overall stability status of the rock and soil mass geological range.

Description

Technical Field

[0001] This invention relates to the field of soil and rock stability assessment technology, and more specifically, to a soil and rock stability assessment method and system based on big data. Background Technology

[0002] In the field of geotechnical engineering, geotechnical stability assessment of soil and rock masses is a core link in ensuring the safety of engineering projects such as mining, tunnel construction, and slope treatment. Its purpose is to determine the stability state of soil and rock masses under the influence of natural conditions or human engineering activities, and to provide a basis for engineering design and risk prevention and control.

[0003] Currently, most existing technologies operate in scenarios involving single-factor analysis or localized regional assessments, often focusing on the geological properties of the soil and rock mass itself, such as lithology and thickness, or only addressing individual external environmental factors, such as precipitation or single engineering disturbances. They lack a systematic consideration of the multi-factor coupling effects of soil and rock mass changes and the mutual influence between regions. This limitation makes it difficult for the assessment results to fully reflect the true geological stability of the soil and rock mass. Therefore, a method and system for assessing the stability of soil and rock mass based on big data is proposed. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for assessing the stability of soil and rock masses based on big data, so as to solve the problems mentioned in the background art.

[0005] To address the aforementioned technical problems, one objective of this invention is to provide a method for assessing the stability of soil and rock masses based on big data, comprising the following steps:

[0006] S1. Obtain the geological range of the rock and soil mass that needs to be assessed and stabilized, and collect relevant big data on the rock and soil mass based on the geological range of the rock and soil mass.

[0007] S2. Based on the geographical location and geological characteristics corresponding to the relevant big data of rock and soil, the geological scope of rock and soil is divided into regions, so that the geological scope of rock and soil is composed of multiple regions.

[0008] S3. Combine relevant big data on soil and rock masses with soil and rock mass regions to analyze the range of changes in soil and rock masses. Adjust the soil and rock mass regions based on the analysis results of the range of changes in soil and rock masses. At the same time, conduct a comprehensive analysis of the factors affecting the range of changes in each soil and rock mass.

[0009] S4. Establish a prediction model, and then combine the list of change factors with the rock and soil area corresponding to each range of rock and soil change to make multiple change predictions, and obtain the rock and soil data corresponding to each change factor.

[0010] S5. For the soil and rock mass region at the latest acquisition time, combine the changing factors and time difference to use the prediction model to perform time prediction soil and rock mass data analysis, and at the same time calculate the correlation value between this soil and rock mass region and other soil and rock mass regions.

[0011] S6. Select soil and rock mass data based on the correlation values, and then combine the selected soil and rock mass data based on the time prediction data to perform a comprehensive stability score on the geological range of the soil and rock mass.

[0012] As a further improvement to this technical solution, step S1 involves collecting relevant big data on soil and rock masses based on the geological range of the soil and rock mass. During the collection process, the big data on soil and rock masses is analyzed according to their corresponding geographical locations. Only the big data on soil and rock masses that are relevant to the geological range of the current soil and rock mass are retained. If the coverage area of ​​the big data on soil and rock masses is larger than the geological range of the current soil and rock mass, then the big data on soil and rock masses is combined with the geological range of the current soil and rock mass for correlation extraction. Only the relevant big data on soil and rock masses within the geological range of the current soil and rock mass are retained.

[0013] As a further improvement to this technical solution, step S2 is as follows:

[0014] S2.1 Obtain the geographical location corresponding to each piece of rock and soil data in the relevant big data of rock and soil, and at the same time perform geological characteristic analysis on the rock and soil data to obtain the geological characteristics corresponding to each piece of rock and soil data;

[0015] S2.2 Summarize the geological characteristics corresponding to each soil and rock mass data, and then divide the soil and rock mass regions into areas based on the summarized geological characteristics.

[0016] As a further improvement to this technical solution, when the relevant big data on soil and rock mass cannot cover the geological range of soil and rock mass during the process of dividing the soil and rock mass region, the S2 sends the geographical locations in the geological range of soil and rock mass that do not have corresponding data to the evaluation and management port. The evaluation and management port collects new soil and rock mass data from the sent geographical locations and then feeds the collected data back to S2 for dividing the soil and rock mass region.

[0017] As a further improvement to this technical solution, step S3 is as follows:

[0018] S3.1 Set a change threshold and sort the relevant soil and rock mass big data according to the collection time. Combine the sorted relevant soil and rock mass big data with the soil and rock mass region and the change range threshold to analyze the change range of soil and rock mass. When the change rate of the soil and rock mass position in the soil and rock mass region is greater than the change threshold, the position is determined to be the change range of soil and rock mass. Conversely, when the change rate of the soil and rock mass position in the soil and rock mass region is less than the change threshold, no adjustment is made.

[0019] S3.2 Adjust the soil and rock mass region based on the analysis results of the soil and rock mass variation range;

[0020] S3.3. Summarize the variation range of each soil and rock mass and combine it with relevant big data of soil and rock masses to analyze the variation factors and obtain a list of variation factors corresponding to the variation range of each soil and rock mass.

[0021] As a further improvement to this technical solution, step S4 is as follows:

[0022] S4.1 Utilize machine learning to build predictive models;

[0023] S4.2. Input the soil and rock mass region corresponding to the range of change of each soil and rock mass, along with the list of change factors and the time difference between real time and acquisition time, into the prediction model. The prediction model will then predict the data for each change factor and output the prediction data corresponding to each change factor. Finally, the output prediction data will be used as the factor prediction soil and rock mass data.

[0024] As a further improvement to this technical solution, step S5 is as follows:

[0025] S5.1. The relevant big data of the soil and rock mass corresponding to the soil and rock mass area at the latest collection time, combined with the changing factors and the time difference between the real time and the collection time, are input into the prediction model for prediction. The prediction model outputs the prediction data, and the prediction data output this time is used as the time prediction soil and rock mass data.

[0026] S5.2 Calculate the correlation value between the latest collected rock and soil area and other rock and soil areas based on the relevant big data of rock and soil, so as to obtain the correlation value between the latest collected rock and soil area and each other rock and soil area.

[0027] As a further improvement to this technical solution, step S6 is as follows:

[0028] S6.1 Set the correlation threshold, and then compare the correlation value of each other soil and rock mass region obtained in S5.2 with the correlation threshold, and only retain other soil and rock mass regions whose correlation value is greater than the correlation threshold;

[0029] S6.2, combine the time-predicted soil and rock mass data with the factor-predicted soil and rock mass data corresponding to other soil and rock mass areas for matching value analysis, and select the corresponding factor-predicted soil and rock mass data for other soil and rock mass areas based on the matching value;

[0030] S6.3 When there are soil and rock areas where soil and rock mass data for which no selection factors have been selected are not available, extract the data again in the unselected soil and rock areas based on the latest acquisition time. Repeat steps S5.1, S5.2, S6.1 and S6.2 until all soil and rock areas have corresponding time-predicted soil and rock mass data and factor-predicted soil and rock mass data.

[0031] S6.4 Integrate the time-predicted rock and soil data and factor-predicted rock and soil data corresponding to all rock and soil regions to perform a comprehensive stability score on the geological range of the rock and soil, thereby obtaining a comprehensive stability score for the geological range of the rock and soil.

[0032] The second objective of this invention is to provide a big data-based stability assessment system for soil and rock masses, including any one of the big data-based stability assessment methods for soil and rock masses described above, comprising a big data collection module, a change factor analysis module, and a comprehensive scoring module;

[0033] The big data collection module is used to obtain the geological range of the rock and soil body that needs to be evaluated for stability. At the same time, it collects relevant big data of the rock and soil body based on the geological range of the rock and soil body. Based on the geographical location and geological characteristics corresponding to the relevant big data of the rock and soil body, it divides the geological range of the rock and soil body into regions, so that the geological range of the rock and soil body consists of multiple rock and soil body regions.

[0034] The change factor analysis module is used to combine relevant big data of soil and rock masses with soil and rock mass regions to analyze the range of change of soil and rock masses, adjust the soil and rock mass regions according to the analysis results of the range of change of soil and rock masses, and conduct a comprehensive analysis of the change factors list for each range of soil and rock masses.

[0035] The comprehensive scoring module is used to establish a prediction model, and then perform multiple change predictions for the soil and rock mass regions corresponding to each range of change, combined with a list of change factors. It obtains the predicted soil and rock mass data corresponding to each change factor, and performs time-predicted soil and rock mass data analysis for the soil and rock mass regions with the latest collection time, combined with change factors and time difference, using the prediction model. At the same time, it calculates the correlation value between the soil and rock mass regions and other soil and rock mass regions, selects the predicted soil and rock mass data based on the correlation value, and then combines the selected predicted soil and rock mass data with the time-predicted soil and rock mass data to perform a comprehensive stability score for the geological range of the soil and rock mass.

[0036] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0037] 1. A method and system for assessing the stability of soil and rock masses based on big data. This method selects factors based on correlation values ​​to predict soil and rock mass data, and combines the selected factor-predicted soil and rock mass data with time-predicted soil and rock mass data to perform a comprehensive stability score. The scoring process considers multiple dimensions such as basic stability score, time-predicted data weight, weight of changing factors, correlation correction coefficient between regions, and regional area weight. The final comprehensive stability score is calculated, which can objectively and accurately reflect the overall stability of the geological range of soil and rock masses, and provides a highly valuable decision-making basis for geological engineering design, construction, and safety management.

[0038] 2. A method and system for assessing the stability of soil and rock masses based on big data. After establishing a prediction model, the method performs multiple change predictions for each soil and rock mass region corresponding to the range of change, combined with a list of change factors. It obtains the predicted soil and rock mass data corresponding to each change factor, and calculates the correlation value between soil and rock mass regions and other soil and rock mass regions. It considers key factors such as spatial, geological, physical, and temporal synchronicity, and calculates the final correlation value through weighted summation to reflect the comprehensive correlation degree between regions. This method of multi-factor prediction and regional correlation analysis comprehensively considers the internal and external factors affecting the geological stability of soil and rock masses, as well as the interaction between different regions, making the assessment more systematic and comprehensive.

[0039] 3. A method and system for assessing the stability of rock and soil masses based on big data. After analyzing the range of changes in rock and soil masses, the boundary of the rock and soil mass region is adjusted according to the analysis results. For example, the range of changes in rock and soil masses at the junction of the edges of two rock and soil mass regions is allocated to the rock and soil mass region with a larger proportion of changes. This allows the division of rock and soil mass regions to be dynamically optimized as the rock and soil masses change, always maintaining a high degree of consistency with the actual geological conditions, and further improving the scientificity and accuracy of the assessment. Attached Figure Description

[0040] Figure 1 This is an overall flowchart of the present invention;

[0041] Figure 2 This invention provides a flowchart for obtaining the geographical location corresponding to each piece of soil and rock data in the relevant soil and rock big data.

[0042] Figure 3 A flowchart illustrating the process of setting a variable threshold for this invention;

[0043] Figure 4 This is a flowchart illustrating the process of building a predictive model using machine learning in this invention.

[0044] Figure 5 This is a flowchart illustrating the process of obtaining the correlation values ​​between the latest data collection time of the soil and rock mass region and each other soil and rock mass region in this invention.

[0045] Figure 6 This is a flowchart illustrating the process of selecting corresponding factors to predict soil and rock mass data based on matching values ​​for other soil and rock mass regions in this invention. Detailed Implementation

[0046] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0047] like Figures 1-6 As shown, one of the objectives of this invention is to provide a method for assessing the stability of soil and rock masses based on big data, comprising the following steps:

[0048] S1. Obtain the geological range of the rock and soil mass that needs to be assessed and stabilized, and collect relevant big data on the rock and soil mass based on the geological range of the rock and soil mass.

[0049] S1 collects relevant big data on soil and rock masses based on their geological extent. During the collection process, it performs geographic location analysis on the big data, retaining only the big data related to the geological extent of the current soil and rock mass. If the coverage area of ​​the big data is larger than the current geological extent, it performs correlation extraction on the big data and retains only the relevant big data within the current geological extent. The specific steps are as follows:

[0050] Data collection scope definition: Clearly define the geological scope of the soil and rock mass to be assessed as the spatial basis for data screening, determine which areas of soil and rock mass big data to collect, and capture soil and rock mass big data covering the scope and surrounding areas from data sources (geological databases, monitoring systems, etc.), including geological attributes, monitoring indicators and other information;

[0051] Geographic location analysis and preliminary screening: For each big data set of soil and rock mass, extract its geographic location information (latitude and longitude, regional boundaries, etc.), mark the spatial coverage area corresponding to the data, compare the data coverage area with the target geological area, and only retain big data set of soil and rock mass whose spatial range is completely within the target geological area, while filtering out obviously irrelevant data;

[0052] Correlation extraction of large-scale data: For big data on soil and rock masses with a coverage area larger than the target geological area, GIS spatial intersection calculation is used to determine the overlapping area between the data and the target geological area. The big data content of soil and rock masses in the overlapping area is extracted, including geological attributes, monitoring values, etc. Irrelevant data that exceeds the target range is discarded, and the part that is relevant to the target range is retained.

[0053] S2. Based on the geographical location and geological characteristics corresponding to the relevant big data of rock and soil, the geological scope of rock and soil is divided into regions, so that the geological scope of rock and soil is composed of multiple regions.

[0054] The steps for S2 are as follows:

[0055] S2.1 Obtain the geographical location corresponding to each piece of rock and soil data in the relevant big data of rock and soil, and at the same time perform geological characteristic analysis on the rock and soil data to obtain the geological characteristics corresponding to each piece of rock and soil data;

[0056] Traverse each collected big data set of soil and rock mass data, extract the geographical location information of the corresponding soil and rock mass data from the data element information or attribute fields, such as latitude and longitude coordinates, name of the mining area / geological block, etc., to clarify the spatial location of the data;

[0057] Using geological analysis methods (such as lithology identification, structural analysis, physical and mechanical testing, etc.), geological characteristics are analyzed for each piece of rock and soil data, and geological characteristic indicators such as lithology (sandstone, shale, etc.), rock and soil thickness, compressive strength, and degree of joint development are identified and recorded.

[0058] S2.2 Summarize the geological characteristics corresponding to each soil and rock mass data, and then divide the soil and rock mass regions into areas based on the summarized geological characteristics.

[0059] First, the regions are divided according to the differences in geological characteristics. Then, a second division is made according to the geographical location corresponding to each rock and soil body data, so that the geological range of the rock and soil body is divided into multiple rock and soil body regions.

[0060] During the process of delineating soil and rock mass regions, when the relevant big data on soil and rock mass cannot cover the geological range of the soil and rock mass, S2 sends the geographical locations within the geological range for which there is no corresponding data to the assessment and management port. The assessment and management port then collects new soil and rock mass data from the sent geographical locations and feeds the collected data back to S2 for soil and rock mass region delineation. The specific steps are as follows:

[0061] Based on differences in geological characteristic indicators (such as different lithology, strength threshold distinction, etc.), cluster analysis (such as K-Means clustering, grouping according to characteristic similarity) and expert experience classification are used to initially divide the geological range of rock and soil into several sub-regions, so that the geological characteristics of rock and soil within the same sub-region are similar.

[0062] For each sub-region obtained in the initial division, the geographical location information of all soil and rock mass data in the region is extracted, and the spatial distribution pattern is analyzed (such as whether it is continuous and whether there are geographical barriers). According to the spatial correlation of geographical location (such as adjacent latitude and longitude belonging to the same region, or being divided by geographical boundaries such as mountains / rivers), the initially divided sub-regions are further subdivided or merged. Finally, the entire geological range of soil and rock mass is divided into multiple soil and rock mass regions with clear boundaries and relatively homogeneous characteristics.

[0063] S3. Combine relevant big data on soil and rock masses with soil and rock mass regions to analyze the range of changes in soil and rock masses. Adjust the soil and rock mass regions based on the analysis results of the range of changes in soil and rock masses. At the same time, conduct a comprehensive analysis of the factors affecting the range of changes in each soil and rock mass.

[0064] The steps for S3 are as follows:

[0065] S3.1 Set a change threshold and sort the relevant soil and rock mass big data according to the collection time. Combine the sorted relevant soil and rock mass big data with the soil and rock mass region and the change range threshold to analyze the change range of soil and rock mass. When the change rate of the soil and rock mass position in the soil and rock mass region is greater than the change threshold, the position is determined to be the change range of soil and rock mass. Conversely, when the change rate of the soil and rock mass position in the soil and rock mass region is less than the change threshold, no adjustment is made.

[0066] S3.2 Adjust the soil and rock mass region based on the analysis results of the soil and rock mass variation range;

[0067] When the boundary between two soil and rock mass regions is within the range of soil and rock variation, the range of soil and rock variation at the boundary is allocated to the soil and rock mass region with a larger proportion of soil and rock variation, thereby reducing the size of the other soil and rock mass region. The specific steps are as follows:

[0068] Data preprocessing and threshold setting: Based on geological characteristics (such as lithology, structure, etc.) and engineering experience, a critical value (such as 5 mm / month) for the rate of change of the location of the soil and rock mass is pre-set as a standard for judging whether the soil and rock mass has undergone significant changes. At the same time, the relevant big data of soil and rock mass collected are arranged in chronological order of collection time to form time series data, which facilitates the analysis of the changes of soil and rock mass at different times.

[0069] Analysis of the range of changes in soil and rock mass: The sorted big data of soil and rock mass is spatially matched with the divided soil and rock mass regions to determine the soil and rock mass region to which each data point belongs. For monitoring points within each soil and rock mass region, the rate of change of position between adjacent time points is calculated. The calculated rate of change is compared with a preset threshold, as shown in the following formula:

[0070]

[0071] Among them, Ratei Let P be the rate of change of the position of monitoring point i, ||·|| be the Euclidean distance, and P be the Euclidean distance. i (t) represents the location coordinates of the monitoring point within the soil and rock mass area at that time, and Δt represents the time interval between two adjacent data acquisitions;

[0072] Rock and soil mass region boundary adjustment: Traverse all rock and soil mass regions and check whether the edge connection of adjacent regions is marked as the rock and soil mass change range. For the change range at the edge connection, calculate its spatial proportion with the change range of the two adjacent rock and soil mass regions, allocate the change range to the rock and soil mass region with the larger proportion, and adjust the boundary of the other region accordingly.

[0073] S3.3. Summarize the variation range of each soil and rock mass and analyze the variation factors by combining relevant big data of soil and rock masses to obtain a list of variation factors corresponding to the variation range of each soil and rock mass. The specific steps are as follows:

[0074] Factor extraction and screening: From the perspective of geological characteristics, analyze the properties of the rock and soil mass itself, such as lithology, rock and soil mass thickness, degree of joint and fracture development, etc., to determine whether the changes are caused by differences in its own characteristics;

[0075] From the perspective of the external environment, we can extract factors that may have an impact, such as precipitation, seismic activity, human engineering activities, and temperature changes.

[0076] By drawing trend diagrams of factors and changes in soil and rock mass, and judging by professional experience, factors that are related to the range of changes in soil and rock mass in time and space are screened out, and obviously irrelevant factors are eliminated (such as earthquakes far away from the region having no impact on local soil and rock mass changes).

[0077] In-depth analysis and list construction: If historical data is available, use data analysis methods (correlation analysis, calculate the correlation coefficient between factors and soil and rock mass change indicators) to quantify the degree of influence of factors on soil and rock mass changes;

[0078] If sufficient quantitative data is unavailable, qualitative analysis should be conducted based on geological theories and engineering cases to determine the mechanism of action of the factors.

[0079] Factors that have been analyzed, verified, and identified as having an impact on the range of change in the rock and soil mass are categorized and organized according to their degree of influence and type of action (inherent geological factors and external environmental factors) to form a list of change factors corresponding to the range of change in the rock and soil mass, clearly presenting the key influencing factors.

[0080] S4. Establish a prediction model, and then combine the list of change factors with the rock and soil area corresponding to each range of rock and soil change to make multiple change predictions, and obtain the rock and soil data corresponding to each change factor.

[0081] The steps for S4 are as follows:

[0082] S4.1 Utilize machine learning to build predictive models;

[0083] S4.2. Input the soil and rock mass region corresponding to each soil and rock mass variation range, along with the list of variation factors and the time difference between real-time and acquisition time, into the prediction model. The prediction model will then predict the data for each variation factor and output the predicted data for each variation factor. Finally, the output predicted data will be used as the factor prediction soil and rock mass data. The specific steps are as follows:

[0084] Based on the characteristics of the data, a machine learning algorithm can be selected. Random forest can be chosen to build a prediction model, followed by model training:

[0085] Initialize the random forest model, train the model using the training set, use historical data of changing factors as the prediction target (supervised learning), and fit the mapping relationship between features and changing factors by integrating multiple decision trees;

[0086] Use the test set to evaluate model performance, analyze feature importance, and screen key features that have a significant impact on changing factors;

[0087] The regional data of the soil and rock mass corresponding to the range of changes to be predicted (including a list of changing factors and the time difference between real-time and acquisition time) are organized and converted according to the input format during model training, and then input into the trained model. The model then calculates the state of the soil and rock mass for each input changing factor, combined with the time difference information, and simulates the development trend of the soil and rock mass under the action of different factors. By outputting the prediction data corresponding to each changing factor, which covers the state indicators of the soil and rock mass under the influence of the factor (such as displacement, stress, etc.), these data are integrated as the factor prediction soil and rock mass data for subsequent comprehensive geological stability assessment.

[0088] S5. For the soil and rock mass region at the latest acquisition time, combine the changing factors and time difference to use the prediction model to perform time prediction soil and rock mass data analysis, and at the same time calculate the correlation value between this soil and rock mass region and other soil and rock mass regions.

[0089] The steps for S5 are as follows:

[0090] S5.1. The relevant big data of the soil and rock mass corresponding to the soil and rock mass area at the latest collection time, combined with the changing factors and the time difference between the real time and the collection time, are input into the prediction model for prediction. The prediction model outputs the prediction data, and the prediction data output this time is used as the time prediction soil and rock mass data.

[0091] Similar to the method in S4, the prediction model built using random forest is also used to predict the output time of the soil and rock mass data, as shown in the following formula:

[0092]

[0093] in, The output of the model is the time-predicted soil and rock mass data vector, where Model is the trained random forest model, and G is the time-predicted soil and rock mass data vector. new F represents the geotechnical mass region data vector at the latest acquisition time. s For the real-time changing factor vector, Δt s For time difference.

[0094] S5.2 Calculate the correlation value between the latest soil and rock mass region and other soil and rock mass regions based on the relevant soil and rock mass big data, so as to obtain the correlation value between the latest soil and rock mass region and each other soil and rock mass region;

[0095] The key factors influencing the inter-regional correlation of soil and rock masses were identified as spatial (inter-regional distance, boundary adjacency), geological (lithological similarity, structural continuity), physical (groundwater seepage, stress transmission), and temporal synchronicity (whether the time intervals for monitoring data collection are consistent, and whether there is a concurrent trend). Then, the correlation values ​​for each key factor were calculated, and these values ​​were normalized. Based on geological expertise, weights were assigned to each dimension (spatial distance weight 0.4, geological attribute weight 0.3, physical effects weight 0.2, and time series weight 0.1). The final correlation value, reflecting the comprehensive correlation between the two regions, was calculated through a weighted summation, as shown in the following formula:

[0096] C(A,B i ) = w d ·R d +w geo ·R geo +w phys ·R phys +w time ·R time ;

[0097] Wherein, C(A, B) i ) represents the comprehensive correlation value between regions A and Bi, ranging from 0 to 1. A larger value indicates a stronger correlation. d w geo w phys w time These are the weights for each dimension, with a combined value of 1, R. d R represents spatial distance. geo For geological attribute similarity, R phys R is the intensity of the physical action. time This represents the time series correlation coefficient.

[0098] S6. Select soil and rock mass data based on the correlation values, and then combine the selected soil and rock mass data based on the time prediction data to perform a comprehensive stability score on the geological range of the soil and rock mass.

[0099] The steps for S6 are as follows:

[0100] S6.1 Set an association threshold (e.g., 0.6, used to determine whether the association between soil and rock regions is significant), and then compare the association value of each other soil and rock region obtained in S5.2 with the association threshold, retaining only other soil and rock regions whose association value is greater than the association value threshold.

[0101] The correlation values ​​of the target soil and rock mass region at the latest acquisition time with those of all other soil and rock mass regions are compared one by one with the correlation threshold.

[0102] If the correlation value of a certain region is greater than the threshold, the region is retained and is considered to have a significant correlation with the target region.

[0103] If the correlation value is less than or equal to the threshold, the region is removed and will not be included in subsequent analysis.

[0104] S6.2, combine the time-predicted soil and rock mass data with the factor-predicted soil and rock mass data corresponding to other soil and rock mass areas for matching value analysis, and select the corresponding factor-predicted soil and rock mass data for other soil and rock mass areas based on the matching value;

[0105] For the retained associated soil and rock mass areas, analyze the degree of matching between the time-predicted soil and rock mass data (reflecting the current state) and the factor-predicted soil and rock mass data (reflecting the predicted state under the influence of various changing factors). The matching dimensions include time synchronization (whether the data acquisition time is consistent), spatial correlation (whether the regional boundaries are adjacent), and physical and logical consistency (such as whether the stress transmission direction matches the predicted deformation).

[0106] Based on the matching value (e.g., determined by a comprehensive score), the most suitable factor prediction data for each retained associated region is selected. For example, factor prediction data that are continuous with the geological structure of the target region and are monitored in sync with the time frame are given priority.

[0107] S6.3 When there are soil and rock areas where soil and rock mass data for which no selection factors have been selected are not available, extract the data again in the unselected soil and rock areas based on the latest acquisition time. Repeat steps S5.1, S5.2, S6.1 and S6.2 until all soil and rock areas have corresponding time-predicted soil and rock mass data and factor-predicted soil and rock mass data.

[0108] Check if there are any soil and rock areas where no factor prediction data has been selected. If so, start from the latest data collection time of that area and repeat the correlation value calculation → threshold filtering → data matching until suitable factor prediction data is found for that area. Repeat this process until all soil and rock areas are matched with the corresponding time prediction data and factor prediction data.

[0109] S6.4. Integrate the time-predicted and factor-predicted rock and soil data corresponding to all rock and soil regions to perform a comprehensive stability score on the geological extent of the rock and soil mass, thereby obtaining a comprehensive stability score for the geological extent of the rock and soil mass. The formula is as follows:

[0110]

[0111] in, β is the basic stability score for the i-th region. time Weighting time-predicted data Here, m represents the standardized value of the time-predicted data for the i-th region, m is the total number of variable factors, and β is the standard value. factor,j Let j be the weight of the j-th changing factor. The standardized value of the predicted data for the j-th change factor in the i-th region;

[0112]

[0113] in, C represents the corrected stability score for the i-th region, reflecting the influence of neighboring regions. i This is the comprehensive correlation correction coefficient between the i-th region and other regions;

[0114]

[0115] Among them, R overall This is a comprehensive stability score for the geological extent of the rock and soil mass; a higher value indicates better stability, A. i is the area weight of the region, and n is the total number of soil and rock regions.

[0116] The second objective of this invention is to provide a big data-based stability assessment system for soil and rock masses, including any one of the above-mentioned big data-based stability assessment methods for soil and rock masses, comprising a big data collection module, a change factor analysis module, and a comprehensive scoring module;

[0117] The big data collection module is used to obtain the geological range of rock and soil bodies that need to be evaluated and stabilized. At the same time, it collects relevant big data on rock and soil bodies based on the geological range of rock and soil bodies. Based on the geographical location and geological characteristics corresponding to the relevant big data on rock and soil bodies, it divides the geological range of rock and soil bodies into regions, so that the geological range of rock and soil bodies consists of multiple rock and soil body regions.

[0118] The variation factor analysis module is used to combine relevant big data of soil and rock masses with soil and rock mass regions to analyze the variation range of soil and rock masses. Based on the analysis results of the variation range of soil and rock masses, the soil and rock mass regions are adjusted. At the same time, the variation range of each soil and rock mass is comprehensively analyzed by listing the variation factors.

[0119] The comprehensive scoring module is used to establish a prediction model. Then, for each soil and rock mass variation range, it performs multiple variation predictions based on the list of variation factors, and obtains the soil and rock mass data corresponding to each variation factor. For the soil and rock mass region with the latest collection time, it performs time prediction data analysis based on the variation factors and time difference using the prediction model. At the same time, it calculates the correlation value between this soil and rock mass region and other soil and rock mass regions. Based on the correlation value, it selects the soil and rock mass data for variation prediction. Then, it combines the selected soil and rock mass data for variation prediction with the time prediction data to perform a comprehensive stability score for the geological range of the soil and rock mass.

[0120] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely preferred examples and are not intended to limit the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for geotechnical stability assessment based on big data, characterized in that: It comprises the following steps: S1, obtain the geological range of the rock-soil body to be evaluated, and collect relevant rock-soil body big data according to the geological range of the rock-soil body; S2, divide the rock-soil body region according to the geographical location and geological characteristics corresponding to the relevant rock-soil body big data, so that the rock-soil body geological range is composed of multiple rock-soil body regions; S3, analyze the rock-soil body change range by combining the relevant rock-soil body big data with the rock-soil body region, adjust the rock-soil body region according to the analysis result of the rock-soil body change range, and comprehensively analyze the change factor list of each rock-soil body change range; S4, establish a prediction model, then make multiple change predictions for the rock-soil body region corresponding to each rock-soil body change range combined with the change factor list, and obtain the factor prediction rock-soil body data corresponding to each change factor respectively; S5, analyze the time prediction rock-soil body data by using the prediction model for the rock-soil body region with the latest collection time combined with the change factor and time difference, and calculate the correlation value of the rock-soil body region with other rock-soil body regions; S6, select the factor prediction rock-soil body data according to the correlation value, then combine the selected factor prediction rock-soil body data with the time prediction rock-soil body data to comprehensively score the stability of the rock-soil body geological range; The steps of S3 are as follows: S3.1, set a change threshold, sort the relevant rock-soil body big data according to the collection time, and analyze the rock-soil body change range by combining the sorted relevant rock-soil body big data with the rock-soil body region and the change range threshold, if the change rate of the rock-soil body position in the rock-soil body region is greater than the change threshold, the position is determined as the rock-soil body change range, otherwise, if the change rate of the rock-soil body position in the rock-soil body region is less than the change threshold, the rock-soil body region is not adjusted; S3.2, adjust the rock-soil body region according to the analysis result of the rock-soil body change range; S3.3, analyze the change factors by combining the relevant rock-soil body big data for each rock-soil body change range, and obtain the change factor list corresponding to each rock-soil body change range; The steps of S5 are as follows: S5.1, input the relevant rock-soil body big data corresponding to the rock-soil body region with the latest collection time combined with the change factor and the time difference between the real-time time and the collection time into the prediction model for prediction, and output the prediction data from the prediction model, which is used as the time prediction rock-soil body data; S5.2, calculate the correlation value of the rock-soil body region with the latest collection time and other rock-soil body regions according to the relevant rock-soil body big data, so as to obtain the correlation value of the rock-soil body region with each other rock-soil body region with the latest collection time; The calculation formula of the correlation value is as follows: ； Wherein, C(A, B i ) is the integrated correlation value of area A and B i , the value is 0~1, w d , w geo , w phys , w time are the weights of each dimension, the integrated value is 1, R d is the spatial distance, R geo is the geological attribute similarity, R phys is the physical action intensity, and R time is the time series correlation coefficient. 2.The method of claim 1, wherein: The S1 collects the relevant rock-soil body big data according to the rock-soil body geological range, analyzes the corresponding geographical position of the rock-soil body big data during the collection process, and only retains the rock-soil body big data related to the rock-soil body geological range, if the rock-soil body big data corresponding to the coverage range is greater than the rock-soil body geological range, the relevant rock-soil body big data is extracted by combining the rock-soil body big data with the rock-soil body geological range, and only the relevant rock-soil body big data of the rock-soil body geological range is retained.

3. The method of claim 1, wherein: The steps of S2 are as follows: S2.1, acquire the geographical position corresponding to each geotechnical data in the relevant geotechnical big data, and simultaneously analyze the geological characteristics of the geotechnical data to acquire the geological characteristics corresponding to each geotechnical data; S2.2, summarize the geological characteristics corresponding to each geotechnical data, and then divide the geotechnical region to be evaluated this time into geotechnical areas according to the summarized geological characteristics.

4. The method of claim 1, wherein: In the process of dividing the geotechnical area, when the relevant geotechnical big data cannot cover the geotechnical region, the S2 sends the geographical position without corresponding data in the geotechnical region to the evaluation management port, collects new geotechnical data from the geographical position sent by the evaluation management port, and then feeds back the collected data to the S2 for geotechnical area division.

5. The method for geotechnical stability assessment based on big data according to claim 1, characterized in that: The steps of the S4 are as follows: S4.1, use machine learning to establish a prediction model; S4.2, input the geotechnical area corresponding to each geotechnical change range, the change factor list, and the time difference between the real-time time and the collection time into the prediction model, predict each change factor by the prediction model, output the prediction data corresponding to each change factor, and then use the output prediction data as the factor prediction geotechnical data.

6. The method of claim 1, wherein: The steps of the S6 are as follows: S6.1, set a correlation threshold, compare the correlation value of each other geotechnical area obtained by S5.2 with the correlation threshold, and only keep the other geotechnical areas with a correlation value greater than the correlation value threshold; S6.2, analyze the matching value of the time prediction geotechnical data combined with the factor prediction geotechnical data corresponding to the retained other geotechnical areas, and select the corresponding factor prediction geotechnical data for the retained other geotechnical areas according to the matching value; S6.3, when there is a geotechnical area without selected factor prediction geotechnical data, extract again in the unselected geotechnical area according to the latest collection time, repeat the steps of S5.1, S5.2, S6.1, and S6.2, and until all geotechnical areas have corresponding time prediction geotechnical data and factor prediction geotechnical data; S6.4, integrate the time prediction geotechnical data and the factor prediction geotechnical data corresponding to all geotechnical areas to comprehensively score the stability of the geotechnical region, so as to comprehensively obtain the stability score of the geotechnical region.

7. A big data based rock mass stability evaluation system for implementing the big data based rock mass stability evaluation method of any one of claims 1-6. It comprises a big data collection module, a change factor analysis module, and a comprehensive scoring module; The big data collection module is used to acquire the geotechnical region that needs to be evaluated and stabilized, collect relevant geotechnical big data according to the geotechnical region, divide the geotechnical region into geotechnical areas according to the geographical position and geological characteristics corresponding to the relevant geotechnical big data, so that the geotechnical region is composed of multiple geotechnical areas; The change factor analysis module is used to analyze the geotechnical change range of the relevant geotechnical big data combined with the geotechnical area, adjust the geotechnical area according to the analysis result of the geotechnical change range, and comprehensively analyze the change factor list of each geotechnical change range; The comprehensive scoring module is used to establish a prediction model, then make a variety of change predictions on the rock-soil body area corresponding to each rock-soil body change range in combination with the change factor list, obtain the factor prediction rock-soil body data respectively corresponding to each change factor, analyze the time prediction rock-soil body data of the rock-soil body area in combination with the change factor and the time difference at the latest collection time by using the prediction model, calculate the correlation value of the rock-soil body area with other rock-soil body areas at the same time, select the factor prediction rock-soil body data according to the correlation value, and then combine the selected factor prediction rock-soil body data with the time prediction rock-soil body data to make a comprehensive stability score on the rock-soil body geological range; The steps of the change factor analysis module are as follows: S3.1, set a change threshold value, sort the relevant rock-soil body big data according to the collection time, combine the sorted relevant rock-soil body big data with the rock-soil body area and the change range threshold value to analyze the rock-soil body change range, and when the change rate of the rock-soil body position in the rock-soil body area is greater than the change threshold value, the position is determined as the rock-soil body change range, otherwise, when the change rate of the rock-soil body position in the rock-soil body area is less than the change threshold value, no adjustment is made; S3.2, adjust the rock-soil body area according to the analysis result of the rock-soil body change range; S3.3, analyze the change factors by combining the relevant rock-soil body big data for each rock-soil body change range, and obtain the change factor list corresponding to each rock-soil body change range; The steps of the comprehensive scoring module are as follows: S5.1, input the relevant rock-soil body big data of the rock-soil body area at the latest collection time in combination with the change factor and the time difference between the real-time time and the collection time into the prediction model for prediction, output the prediction data from the prediction model, and take the output prediction data this time as the time prediction rock-soil body data; S5.2, calculate the correlation value of the rock-soil body area at the latest collection time with other rock-soil body areas according to the relevant rock-soil body big data, so as to obtain the correlation value of the rock-soil body area at the latest collection time with each other rock-soil body area; The calculation formula of the correlation value is as follows: ； Wherein, C(A, B i ) is the integrated correlation value of area A and B i , the value is 0~1, w d , w geo , w phys , w time are the weights of each dimension, the integrated value is 1, R d is the spatial distance, R geo is the geological attribute similarity, R phys is the physical action intensity, and R time is the time series correlation coefficient.

Images