A Real-time Monitoring System and Device for Formation Deformation Based on Smart Sensors

Abstract

This invention relates to the field of geological deformation measurement technology and discloses a real-time geological deformation monitoring system and device based on intelligent sensors. The method includes: an acquisition module for collecting experimental soil data, historical environmental data, and historical geological deformation data; an analysis and clustering module for determining multiple environmental cluster sets and multiple deformation cluster sets, and determining multiple second environmental deformation cluster sets; a geological characteristic module for determining the geological characteristic data of the stratum to be monitored; and a real-time monitoring module for collecting real-time environmental data and real-time geological deformation data, combining all second environmental deformation cluster sets and geological characteristic data to achieve real-time monitoring of the geological deformation of the stratum to be monitored. This system enables refined monitoring and data processing of the complex relationship between strata, environment, and deformation, identifies geological deformation patterns, achieves accurate prediction and dynamic early warning of deformation, improves monitoring accuracy, monitoring timeliness, and risk prevention capabilities, and reduces the risk of geological disasters.

Description

Technical Field

[0001] This invention relates to the field of formation deformation measurement technology, and in particular to a real-time formation deformation monitoring system and device based on intelligent sensors. Background Technology

[0002] In the fields of geological engineering and geophysics, monitoring of strata deformation is crucial. Early methods relied primarily on manual monitoring, which was inefficient, lacked real-time accuracy, and was highly susceptible to construction conditions and external environmental factors. Later, bedrock markers were developed, using markers placed at different depths underground and their positional changes measured periodically to monitor subsurface rock deformation, but these methods could not monitor surface deformation. Fiber optic strata deformation measurement, as a new technology, while offering advantages such as high sensitivity, is also limited to subsurface rock strata. In recent years, with the development of artificial intelligence and smart sensor technologies, smart sensors have been increasingly applied to strata monitoring, enabling real-time acquisition of environmental data such as temperature and humidity, groundwater level, and soil stress, as well as micro-deformation signals. However, current monitoring methods still struggle to comprehensively and accurately obtain the overall situation of strata deformation, and the integrated analysis of environmental monitoring data and strata deformation data is insufficient.

[0003] Therefore, this invention proposes a real-time monitoring system and device for formation deformation based on intelligent sensors. Summary of the Invention

[0004] This invention provides a real-time monitoring system and device for ground deformation based on intelligent sensors. By analyzing historical environmental data and historical ground deformation data, multiple second-environment deformation clusters are determined. Based on collected experimental soil data and all second-environment deformation clusters, ground characteristic data of the stratum to be monitored are determined. Real-time environmental data and real-time ground deformation data of the stratum to be monitored are collected. By combining all second-environment deformation clusters and ground characteristic data, real-time monitoring of ground deformation of the stratum to be monitored is achieved. This enables refined monitoring and data processing of the complex relationship between strata, environment, and deformation, accurately identifying ground deformation patterns, achieving accurate deformation prediction and dynamic early warning, improving monitoring accuracy, timeliness, and risk prevention capabilities, and reducing the risk of geological disasters.

[0005] This invention provides a real-time monitoring system for formation deformation based on intelligent sensors, comprising:

[0006] Acquisition module: Collects test soil data of the strata to be monitored, and obtains historical environmental data and historical stratum deformation data of the strata to be monitored;

[0007] Clustering Analysis Module: Analyzes historical environmental data and historical stratigraphic deformation data to determine multiple environmental cluster sets and multiple deformation cluster sets. Based on all environmental cluster sets and all deformation cluster sets, it determines multiple second environmental deformation cluster sets.

[0008] Stratigraphic characteristics module: Based on experimental soil data and all second environmental deformation cluster sets, determine the stratigraphic characteristics data of the strata to be monitored;

[0009] Real-time monitoring module: Collects real-time environmental data and real-time formation deformation data of the formation to be monitored. Based on the real-time environmental data, real-time formation deformation data, all second-environment deformation clusters, and formation characteristic data, it realizes real-time monitoring of formation deformation of the formation to be monitored.

[0010] Preferably, a real-time monitoring system for formation deformation based on intelligent sensors includes an acquisition module comprising:

[0011] Key area units: Obtain geological exploration data and engineering design data of the strata to be monitored, and identify multiple key areas of the strata to be monitored based on the geological exploration data;

[0012] Grid division unit: The first grid is used to divide all key areas of the stratum to be monitored, and the second grid is used to divide the other areas of the stratum to be monitored except for all key areas.

[0013] Sampling point unit: Based on the strata to be monitored after the first grid division and the second grid division, multiple sampling points are determined for the strata to be monitored;

[0014] Experimental sub-soil unit: Soil is collected from each sampling point of the stratum to be tested, and the experimental sub-soil and experimental sub-soil data for each sampling point are determined. The experimental sub-soil data includes the sampling point number, sampling time, sampling depth, and sampling location.

[0015] Sample soil data unit: Based on the test sub-soil and test sub-soil data of all sampling points of the stratum to be tested, the sample soil data of the stratum to be tested is determined.

[0016] Preferably, a real-time monitoring system for formation deformation based on intelligent sensors, including a data acquisition module, further comprises:

[0017] Historical environmental data unit: Acquire historical environmental sub-data for each sampling point in the stratum to be monitored, and determine the historical environmental data of the stratum to be monitored based on the historical environmental sub-data of all sampling points in the stratum to be monitored;

[0018] Historical stratigraphic deformation data unit: Acquire historical stratigraphic deformation sub-data for each sampling point in the stratigraphic unit to be monitored, and determine the historical stratigraphic deformation data of the stratigraphic unit to be monitored based on the historical stratigraphic deformation sub-data for all sampling points in the stratigraphic unit to be monitored.

[0019] Preferably, a real-time monitoring system for formation deformation based on intelligent sensors includes an analysis and clustering module, comprising:

[0020] Environmental feature vector unit: Extract features from the historical environmental sub-data of each sampling point in the historical environmental data of the stratum to be monitored, and determine the environmental feature vector of each sampling point in the stratum to be monitored;

[0021] First cluster analysis unit: Based on the environmental feature vectors of all sampling points in the stratum to be monitored, a first cluster analysis is performed on all sampling points in the stratum to be monitored. Based on the results of the first cluster analysis, multiple environmental cluster sets are determined. Each environmental cluster set includes multiple sampling points and the environmental feature vector of each sampling point.

[0022] Stratigraphic deformation vector unit: Features are extracted from the historical stratigraphic deformation sub-data of each sampling point in the historical stratigraphic deformation data of the stratum to be monitored to determine the stratigraphic deformation vector of each sampling point in the stratum to be monitored;

[0023] The second cluster analysis unit: Based on the formation deformation vectors of all sampling points in the formation to be monitored, a second cluster analysis is performed on all sampling points in the formation to be monitored. Based on the results of the second cluster analysis, multiple deformation cluster sets are determined. Each deformation cluster set includes multiple sampling points and the formation deformation vector of each sampling point.

[0024] Preferably, a real-time monitoring system for formation deformation based on intelligent sensors, including an analysis and clustering module, further includes:

[0025] Calculation unit: Based on all environment cluster sets and all deformed cluster sets, calculate the first set consistency value and the second set consistency value for each environment cluster set and each deformed cluster set, and calculate the sampling weight for each sampling point;

[0026] Unit determination: Based on the first set consensus value and the second set consensus value of all environmental cluster sets and deformation cluster sets, as well as the sampling weights of all sampling points, determine the important sampling point set and determine multiple first environmental deformation cluster sets, wherein the first environmental deformation cluster set includes multiple sampling points and the environmental feature vector and the formation deformation vector of each sampling point;

[0027] Membership value unit: For each sampling point that is not in any of the first environment deformation cluster sets, calculate the membership value of each sampling point and each environment deformation cluster set;

[0028] Second environmental deformation cluster set unit: Based on the membership values ​​of all sampling points not in any environmental deformation cluster set and all first environmental deformation cluster sets, multiple second environmental deformation cluster sets are determined. The second environmental deformation cluster set includes multiple sampling points and the environmental feature vector and formation deformation vector of each sampling point.

[0029] Non-convergent unit: For all sampling points that are not in any second environmental deformation cluster set, as well as the environmental feature vector and formation deformation vector of each sampling point, determine a second environmental deformation cluster set and mark it as non-convergent.

[0030] Preferably, a real-time formation deformation monitoring system based on intelligent sensors includes a formation characteristic module, comprising:

[0031] Physical property vector unit: Physical property tests are performed on the test sub-soil at each sampling point in the sample soil data to determine the physical property vector of the test sub-soil at each sampling point;

[0032] Mechanical property vector unit: The mechanical properties of the test sub-soil at each sampling point in the soil sample data are tested to determine the mechanical property vector of the test sub-soil at each sampling point;

[0033] Microscopic property vector unit: Perform microstructure analysis on the test sub-soil at each sampling point in the soil sample data to determine the microscopic property vector of the test sub-soil at each sampling point;

[0034] Stratigraphic characteristic sub-vector unit: Based on the physical property vector, mechanical property vector and microscopic property vector of the test sub-soil at each sampling point in the soil sample data, the stratigraphic characteristic sub-vector of the test sub-soil at each sampling point in the soil sample data is determined.

[0035] Converging stratigraphic data unit: Based on all sampling points in each second environmental deformation cluster set without non-convergence labels and the stratigraphic characteristic sub-vector of each sampling point, the convergent stratigraphic data of each second environmental deformation cluster set without non-convergence labels is determined;

[0036] Non-convergent stratigraphic data unit: Based on all sampling points of the second environmental deformation cluster set with non-convergent labels and the stratigraphic characteristic sub-vector of each sampling point, non-convergent stratigraphic data of the second environmental deformation cluster set with non-convergent labels are determined;

[0037] Stratigraphic characteristic data unit: Based on the convergent stratigraphic data of all second environmental deformation clusters without non-convergence labels, and the non-convergence stratigraphic data of second environmental deformation clusters with non-convergence labels, the stratigraphic characteristic data is determined.

[0038] Preferably, a real-time monitoring system for formation deformation based on intelligent sensors includes a real-time monitoring module comprising:

[0039] Formation deformation prediction sub-model unit: The environmental feature vectors of all sampling points in each second environmental deformation cluster set, and the formation characteristic sub-vectors of all sampling points in convergent or non-convergent formation data are used as inputs to the formation deformation prediction sub-model. The formation deformation vectors of all sampling points in each second environmental deformation cluster set are used as outputs to construct the formation deformation prediction sub-model for each second environmental deformation cluster set.

[0040] Stratigraphic Deformation Prediction Model Unit: Based on the stratigraphic deformation prediction sub-model of all second-environment deformation cluster sets, the stratigraphic deformation prediction model of the stratum to be monitored is determined.

[0041] Real-time monitoring unit: Based on intelligent sensor array, it monitors the real-time environmental data and real-time formation deformation data of the formation to be monitored.

[0042] Prediction Unit: Inputs real-time environmental data and real-time stratum deformation data into the stratum deformation prediction model. Based on the output of the stratum deformation prediction model, it determines the predicted deformation data and deformation early warning data of the stratum to be monitored. Based on the predicted deformation data and deformation early warning data, it realizes real-time monitoring of stratum deformation of the stratum to be monitored.

[0043] The present invention provides a real-time monitoring device for formation deformation based on intelligent sensors, used to execute the real-time monitoring system for formation deformation based on intelligent sensors described in any one of embodiments 1 to 7.

[0044] The beneficial effects of this invention compared to existing technologies are as follows: By analyzing historical environmental data and historical stratigraphic deformation data, multiple second environmental deformation cluster sets are determined. Based on collected experimental soil data and all second environmental deformation cluster sets, stratigraphic characteristic data of the stratum to be monitored are determined. Real-time environmental data and real-time stratigraphic deformation data of the stratum to be monitored are collected. By combining all second environmental deformation cluster sets and stratigraphic characteristic data, real-time monitoring of stratigraphic deformation of the stratum to be monitored is achieved. This enables refined monitoring and data processing of the complex relationship between stratigraphy, environment, and deformation, accurately identifies stratigraphic deformation patterns, achieves accurate deformation prediction and dynamic early warning, improves monitoring accuracy, monitoring timeliness, and risk prevention capabilities, reduces geological disaster risks, and lays a good foundation for the formation of artificial intelligence demonstration application scenarios.

[0045] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in this application.

[0046] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0047] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0048] Figure 1 This is a schematic diagram of a real-time ground deformation monitoring system based on intelligent sensors, according to an embodiment of the present invention. Detailed Implementation

[0049] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention. Example 1:

[0050] This invention provides a real-time monitoring system for formation deformation based on intelligent sensors, with reference to... Figure 1 ,include:

[0051] Acquisition module: Collects test soil data of the strata to be monitored, and obtains historical environmental data and historical stratum deformation data of the strata to be monitored;

[0052] Clustering Analysis Module: Analyzes historical environmental data and historical stratigraphic deformation data to determine multiple environmental cluster sets and multiple deformation cluster sets. Based on all environmental cluster sets and all deformation cluster sets, it determines multiple second environmental deformation cluster sets.

[0053] Stratigraphic characteristics module: Based on experimental soil data and all second environmental deformation cluster sets, determine the stratigraphic characteristics data of the strata to be monitored;

[0054] Real-time monitoring module: Collects real-time environmental data and real-time formation deformation data of the formation to be monitored. Based on the real-time environmental data, real-time formation deformation data, all second-environment deformation clusters, and formation characteristic data, it realizes real-time monitoring of formation deformation of the formation to be monitored.

[0055] In this embodiment, physical and mechanical property data of the strata to be monitored are obtained through professional surveying methods (such as borehole sampling and in-situ testing). Historical environmental information of the area to be monitored is collected, including meteorological data (such as annual precipitation, temperature, and freeze-thaw cycle records), hydrological data (such as interannual variations in groundwater level and fluctuations in pore water pressure), and human activity data (such as the construction sequence of surrounding buildings and historical load changes). For example, in urban areas, it is necessary to collect data on the pile foundation construction time of surrounding high-rise buildings and drainage records of underground pipelines to analyze the long-term impact of human activities on the strata. Historical monitoring data is compiled, including deformation records such as settlement, horizontal displacement, and crack propagation at each sampling point. For example, by reviewing monitoring reports from the past 5 years, the cumulative settlement curve of a strata along a subway line or the displacement trend of a foundation pit slope can be obtained. This data is used to reveal the historical patterns of strata deformation and their correlation with environmental driving factors.

[0056] In this embodiment, feature extraction is performed on historical environmental data, and a clustering algorithm is used to group sampling points with similar environmental conditions into the same set. Feature extraction is also performed on historical stratigraphic deformation data, and a clustering algorithm is used to classify deformation patterns.

[0057] In this embodiment, multi-scale tests are conducted on soil samples at each sampling point, covering physical properties (such as water content and density), mechanical properties (such as compression modulus and shear strength), and microscopic properties (such as particle size distribution and pore structure).

[0058] In this embodiment, the data of all sampling points within each second environmental deformation cluster set are integrated to determine whether the formation data is convergent or non-convergent based on the second environmental deformation cluster set, and the formation characteristic data is determined.

[0059] In this embodiment, real-time environmental data and deformation data are input into the formation deformation prediction model (integrated by various clustering sub-models) to determine the predicted deformation data and deformation early warning data, thereby realizing real-time monitoring of formation deformation of the formation to be monitored.

[0060] The beneficial effects of the above technology are as follows: By analyzing historical environmental data and historical stratigraphic deformation data, multiple second-environment deformation clusters are determined. Based on collected experimental soil data and all second-environment deformation clusters, stratigraphic characteristic data of the stratum to be monitored are determined. Real-time environmental data and real-time stratigraphic deformation data of the stratum to be monitored are collected. By combining all second-environment deformation clusters and stratigraphic characteristic data, real-time monitoring of stratigraphic deformation of the stratum to be monitored can be achieved. This enables refined monitoring and data processing of the complex relationship between stratigraphy, environment, and deformation, accurately identifies stratigraphic deformation patterns, achieves accurate deformation prediction and dynamic early warning, improves monitoring accuracy, monitoring timeliness, and risk prevention and control capabilities, and reduces the risk of geological disasters. Example 2:

[0061] Based on Example 1, a real-time monitoring system for formation deformation based on intelligent sensors includes an acquisition module comprising:

[0062] Key area units: Obtain geological exploration data and engineering design data of the strata to be monitored, and identify multiple key areas of the strata to be monitored based on the geological exploration data;

[0063] Grid division unit: The first grid is used to divide all key areas of the stratum to be monitored, and the second grid is used to divide the other areas of the stratum to be monitored except for all key areas.

[0064] Sampling point unit: Based on the strata to be monitored after the first grid division and the second grid division, multiple sampling points are determined for the strata to be monitored;

[0065] Experimental sub-soil unit: Soil is collected from each sampling point of the stratum to be tested, and the experimental sub-soil and experimental sub-soil data for each sampling point are determined. The experimental sub-soil data includes the sampling point number, sampling time, sampling depth, and sampling location.

[0066] Sample soil data unit: Based on the test sub-soil and test sub-soil data of all sampling points of the stratum to be tested, the sample soil data of the stratum to be tested is determined.

[0067] In this embodiment, geological exploration data includes detailed information such as lithology, stratigraphic structure, groundwater level, and physical and mechanical properties of the soil. This data is typically obtained through drilling, geophysical exploration, and in-situ testing.

[0068] In this embodiment, the engineering design data involves construction plans and design parameters (such as excavation depth, tunnel cross-sectional dimensions, and building loads) for completed or ongoing projects. These data reflect the potential impact of engineering activities on the strata.

[0069] In this embodiment, based on geological survey data, the heterogeneity of strata, the location of weak layers, faults and other special geological conditions are analyzed to determine key areas in the strata. Combined with engineering design data, areas that are closely related to engineering activities and may be greatly affected, such as foundation pit slopes and tunnel faces, are identified and marked as key areas.

[0070] In this embodiment, key areas are divided into fine grids to ensure that subtle changes in the strata in these areas can be captured. The grid size can be adjusted according to the specific geological conditions and engineering requirements of the key areas, and is usually small (e.g., 1 meter × 1 meter). Fine grid division can improve monitoring accuracy.

[0071] In this embodiment, a coarser grid is used to divide the strata under monitoring, except for key areas. These areas are relatively stable, and the grid size can be larger (e.g., 5m x 5m) to reduce monitoring costs. Although the coarse grid has lower precision, it can cover a larger area.

[0072] In this embodiment, multiple sampling points are determined based on the results of the first and second grid divisions. The sampling points should be evenly distributed across the grid nodes to ensure coverage of the entire monitoring area. In critical areas, the sampling points should be more densely packed to obtain more detailed data; in other areas, the sampling points can be relatively sparse to balance monitoring cost and accuracy.

[0073] In this embodiment, soil samples are collected from each sampling point to obtain experimental sub-soil samples. Detailed sampling information should be recorded during the collection process, including the sampling point number, sampling time, sampling depth, and sampling location. Sampling point number: used to uniquely identify each sampling point; Sampling time: records the specific time of sampling; Sampling depth: records the specific depth of sampling, reflecting the vertical distribution of the strata; Sampling location: records the specific coordinates of the sampling location.

[0074] In this embodiment, the sample soil data of the stratum to be monitored is integrated based on the test sub-soil and its data from all sampling points.

[0075] The beneficial effects of the above technologies are: collecting test soil data from the strata to be monitored can accurately identify key areas, improve monitoring accuracy and targeting, provide more reliable data support for stratum deformation monitoring, and reduce the risk of geological disasters. Example 3:

[0076] Based on Example 2, the acquisition module further includes:

[0077] Historical environmental data unit: Acquire historical environmental sub-data for each sampling point in the stratum to be monitored, and determine the historical environmental data of the stratum to be monitored based on the historical environmental sub-data of all sampling points in the stratum to be monitored;

[0078] Historical stratigraphic deformation data unit: Acquire historical stratigraphic deformation sub-data for each sampling point in the stratigraphic unit to be monitored, and determine the historical stratigraphic deformation data of the stratigraphic unit to be monitored based on the historical stratigraphic deformation sub-data for all sampling points in the stratigraphic unit to be monitored.

[0079] In this embodiment, for each sampling point, its historical environmental sub-data is obtained, including temperature, humidity, rainfall, groundwater level, etc.

[0080] In this embodiment, historical environmental sub-data from all sampling points are integrated to form complete historical environmental data of the stratum to be monitored.

[0081] In this embodiment, for each sampling point, its historical stratigraphic deformation data are obtained, including displacement, strain, dip angle, etc.

[0082] In this embodiment, historical stratigraphic deformation sub-data from all sampling points are integrated to form complete historical stratigraphic deformation data of the strata to be monitored.

[0083] The beneficial effects of the above technologies are: obtaining historical environmental data and historical stratum deformation data of the strata to be monitored can provide more reliable data support for stratum deformation monitoring and reduce the risk of geological disasters. Example 4:

[0084] Based on Example 3, the clustering module is analyzed, including:

[0085] Environmental feature vector unit: Extract features from the historical environmental sub-data of each sampling point in the historical environmental data of the stratum to be monitored, and determine the environmental feature vector of each sampling point in the stratum to be monitored;

[0086] First cluster analysis unit: Based on the environmental feature vectors of all sampling points in the stratum to be monitored, a first cluster analysis is performed on all sampling points in the stratum to be monitored. Based on the results of the first cluster analysis, multiple environmental cluster sets are determined. Each environmental cluster set includes multiple sampling points and the environmental feature vector of each sampling point.

[0087] Stratigraphic deformation vector unit: Features are extracted from the historical stratigraphic deformation sub-data of each sampling point in the historical stratigraphic deformation data of the stratum to be monitored to determine the stratigraphic deformation vector of each sampling point in the stratum to be monitored;

[0088] The second cluster analysis unit: Based on the formation deformation vectors of all sampling points in the formation to be monitored, a second cluster analysis is performed on all sampling points in the formation to be monitored. Based on the results of the second cluster analysis, multiple deformation cluster sets are determined. Each deformation cluster set includes multiple sampling points and the formation deformation vector of each sampling point.

[0089] In this embodiment, feature extraction is performed on the historical environmental sub-data of each sampling point in the historical environmental data of the stratum to be monitored. This data includes environmental parameters such as temperature, humidity, rainfall, and groundwater level.

[0090] In this embodiment, environmental data at each sampling point is transformed into a feature vector through feature extraction.

[0091] In this embodiment, a first cluster analysis is performed based on the environmental feature vectors of all sampling points in the stratum to be monitored. Clustering algorithms such as K-means or DBSCAN can be used to divide the sampling points into multiple environmental cluster sets according to the cluster analysis results. Each environmental cluster set includes multiple sampling points and their corresponding environmental feature vectors.

[0092] In this embodiment, feature extraction is performed on the historical stratigraphic deformation sub-data of each sampling point in the historical stratigraphic deformation data of the stratum to be monitored. These data include deformation parameters such as displacement, strain, and dip angle.

[0093] In this embodiment, feature extraction is used to transform the deformed data of each sampling point into a feature vector.

[0094] In this embodiment, a second cluster analysis is performed based on the formation deformation vectors of all sampling points in the formation to be monitored. Similarly, clustering algorithms such as K-means or DBSCAN can be used to determine deformation cluster sets: based on the cluster analysis results, the sampling points are divided into multiple deformation cluster sets. Each deformation cluster set includes multiple sampling points and their corresponding formation deformation vectors.

[0095] The beneficial effects of the above technologies are as follows: analyzing historical environmental data and historical stratigraphic deformation data, identifying multiple environmental cluster sets and multiple deformation cluster sets can improve data support for identifying multiple secondary environmental deformation cluster sets, provide a scientific basis for accurate monitoring and prediction of stratigraphic deformation, improve monitoring efficiency, and effectively reduce the risk of geological disasters. Example 5:

[0096] Based on Example 4, the clustering analysis module further includes:

[0097] Calculation unit: Based on all environment cluster sets and all deformed cluster sets, calculate the first set consistency value and the second set consistency value for each environment cluster set and each deformed cluster set, and calculate the sampling weight for each sampling point;

[0098] Unit determination: Based on the first set consensus value and the second set consensus value of all environmental cluster sets and deformation cluster sets, as well as the sampling weights of all sampling points, determine the important sampling point set and determine multiple first environmental deformation cluster sets, wherein the first environmental deformation cluster set includes multiple sampling points and the environmental feature vector and the formation deformation vector of each sampling point;

[0099] Membership value unit: For each sampling point that is not in any of the first environment deformation cluster sets, calculate the membership value of each sampling point and each environment deformation cluster set;

[0100] Second environmental deformation cluster set unit: Based on the membership values ​​of all sampling points not in any environmental deformation cluster set and all first environmental deformation cluster sets, multiple second environmental deformation cluster sets are determined. The second environmental deformation cluster set includes multiple sampling points and the environmental feature vector and formation deformation vector of each sampling point.

[0101] Non-convergent unit: For all sampling points that are not in any second environmental deformation cluster set, as well as the environmental feature vector and formation deformation vector of each sampling point, determine a second environmental deformation cluster set and mark it as non-convergent.

[0102] In this embodiment, the calculation unit calculates the first set consistency value and the second set consistency value for each environment cluster set and each deformed cluster set based on all environment cluster sets and all deformed cluster sets, and the formula for calculating the sampling weight of each sampling point can be expressed as follows:

[0103] ;

[0104] ;

[0105] ;

[0106] in, This indicates the first set consistency value between the i-th environment cluster set and the j-th deformed cluster set. This represents the i-th environment cluster set. Let j represent the j-th deformed cluster set. This indicates the second set consistency value between the i-th environment cluster set and the j-th deformed cluster set. This represents the k-th sampling point. The value is 1 when the k-th sampling point belongs to the intersection of the i-th environment cluster set and the j-th deformed cluster set. The value is 1 when the k-th sampling point belongs to the i-th environment cluster set. The value is 1 when the k-th sampling point belongs to the intersection of the j-th deformed cluster set, and N1 represents the number of sampling points. This represents the sampling weight of the k-th sampling point. This represents the environmental feature vector of the k-th sampling point. This represents the average environmental feature vector of all sampling points in the environmental cluster set to which the k-th sampling point belongs. This represents the standard deviation of the environmental feature vectors of all sampling points in the environmental cluster set to which the k-th sampling point belongs. Let α represent the formation deformation vector at the k-th sampling point, where α represents the environmental weight and β represents the deformation weight.

[0107] In this embodiment, This represents the magnitude of the formation deformation vector at the k-th sampling point.

[0108] In this embodiment, The Euclidean distance between the environmental feature vector of the k-th sampling point and the average environmental feature vector of all sampling points in the environmental cluster set to which the k-th sampling point belongs is expressed.

[0109] In this embodiment, the determining unit determines the set of important sampling points based on the first set consensus value and the second set consensus value of all environmental cluster sets and deformed cluster sets, as well as the sampling weights of all sampling points. The calculation formula for determining multiple first environmental deformed cluster sets can be expressed as follows:

[0110] ;

[0111] :

[0112] ;

[0113] Where IS represents the set of important sampling points, TH1 represents the consensus threshold of the first set, and TH2 represents the consensus threshold of the second set. Let DT represent the first environmental deformation cluster set based on the i-th environmental cluster set and the j-th deformation cluster set, and let DT represent the maximum deformation of the formation to be monitored.

[0114] In this embodiment, the membership value unit: for each sampling point that is not in any of the first environmental deformation cluster sets, the formula for calculating the membership value of each sampling point and each environmental deformation cluster set can be expressed as:

[0115] ;

[0116] in, This represents the membership value of the a-th sampling point that is not in any environmental deformation cluster set, and γ represents the distance attenuation coefficient. This represents the environmental feature vector of the a-th sampling point that is not in any environmental deformation cluster set. Let represent the environmental feature vector of the b-th sampling point belonging to the environmental deformable cluster set based on the i-th environmental cluster set and the j-th deformable cluster set. Nu represents the number of environmental cluster sets and the number of deformable cluster sets, and ijN2 represents the number of sampling points in the environmental deformable cluster set based on the i-th environmental cluster set and the j-th deformable cluster set.

[0117] In this embodiment, Let represent the Euclidean distance between the environmental feature vector of the a-th sampling point that is not in any environmental deformation cluster set and the environmental feature vector of the b-th sampling point that belongs to the environmental deformation cluster set based on the i-th environmental cluster set and the j-th deformation cluster set.

[0118] In this embodiment, Let Euclidean distance be the environmental feature vector of the k-th sampling point and the environmental feature vector of the a-th sampling point that is not in any environmental deformation cluster set.

[0119] In this embodiment, the calculation formula for determining multiple second environmental deformation cluster sets based on the membership values ​​of all sampling points not in any environmental deformation cluster set and all first environmental deformation cluster sets can be expressed as follows:

[0120] ;

[0121] in, This represents the second environmental deformable cluster set based on the i-th environmental cluster set and the j-th deformable cluster set. This represents the a-th sampling point. MD1 represents the sub-membership value of the a-th sampling point, which is not in any environmental deformation cluster set, based on the i-th environmental cluster set and the j-th deformation cluster set. MD2 represents the membership value threshold and the sub-membership value threshold.

[0122] In this embodiment, for sampling points that cannot be reasonably assigned to any environmental deformation cluster set, they are separately divided into a new second environmental deformation cluster set and marked as non-convergent. For example, if two sampling points cannot be assigned to any second environmental deformation cluster set, these two sampling points are combined into a new second environmental deformation cluster set and marked as non-convergent.

[0123] The beneficial effects of the above technologies are as follows: Based on all environmental cluster sets and all deformation cluster sets, multiple second environmental deformation cluster sets are determined. Through multi-level cluster optimization, it can be ensured that all sampling points are reasonably classified, realize refined data processing of complex relationships between strata, environment and deformation, improve monitoring accuracy, and effectively reduce the risk of geological disasters. Example 6:

[0124] Based on Example 1, the formation characteristics module includes:

[0125] Physical property vector unit: Physical property tests are performed on the test sub-soil at each sampling point in the sample soil data to determine the physical property vector of the test sub-soil at each sampling point;

[0126] Mechanical property vector unit: The mechanical properties of the test sub-soil at each sampling point in the soil sample data are tested to determine the mechanical property vector of the test sub-soil at each sampling point;

[0127] Microscopic property vector unit: Perform microstructure analysis on the test sub-soil at each sampling point in the soil sample data to determine the microscopic property vector of the test sub-soil at each sampling point;

[0128] Stratigraphic characteristic sub-vector unit: Based on the physical property vector, mechanical property vector and microscopic property vector of the test sub-soil at each sampling point in the soil sample data, the stratigraphic characteristic sub-vector of the test sub-soil at each sampling point in the soil sample data is determined.

[0129] Converging stratigraphic data unit: Based on all sampling points in each second environmental deformation cluster set without non-convergence labels and the stratigraphic characteristic sub-vector of each sampling point, the convergent stratigraphic data of each second environmental deformation cluster set without non-convergence labels is determined;

[0130] Non-convergent stratigraphic data unit: Based on all sampling points of the second environmental deformation cluster set with non-convergent labels and the stratigraphic characteristic sub-vector of each sampling point, non-convergent stratigraphic data of the second environmental deformation cluster set with non-convergent labels are determined;

[0131] Stratigraphic characteristic data unit: Based on the convergent stratigraphic data of all second environmental deformation clusters without non-convergence labels, and the non-convergence stratigraphic data of second environmental deformation clusters with non-convergence labels, the stratigraphic characteristic data is determined.

[0132] In this embodiment, physical properties of the test soil at each sampling point in the soil sample data are tested, including but not limited to moisture content, density, and porosity. For example, moisture content is tested using the drying method, density is tested using the ring sampler method, and porosity is tested by calculating the ratio of soil pore volume to total volume.

[0133] In this embodiment, the mechanical properties of the test sub-soil at each sampling point in the soil sample data are tested, including but not limited to compression modulus, cohesion, and internal friction angle. For example, the compression modulus is obtained through a compression test, the cohesion is obtained through a direct shear test, and the internal friction angle is also calculated through a direct shear test.

[0134] In this embodiment, the microstructure of the test sub-soil at each sampling point in the soil sample data is analyzed, including soil particle size distribution and pore structure. For example, the soil particle size distribution is measured using a laser particle size analyzer, and the pore structure is observed using a scanning electron microscope (SEM).

[0135] In this embodiment, the physical, mechanical, and microscopic characteristic vectors of each sampling point are concatenated into a unified formation characteristic sub-vector to determine the formation characteristic sub-vector of each sampling point.

[0136] In this embodiment, for cluster sets without convergence labels, all sampling points and their stratigraphic characteristic sub-vectors in the second environmental deformation cluster set are integrated to determine the convergent stratigraphic data for each cluster set.

[0137] In this embodiment, for a second environmental deformation cluster set with non-convergence labels, all sampling points and their formation characteristic sub-vectors in the second environmental deformation cluster set are integrated to determine the non-convergence formation data of the cluster set.

[0138] In this embodiment, converged stratigraphic data of all second environmental deformation cluster sets without non-convergence labels and non-convergence stratigraphic data of second environmental deformation cluster sets with non-convergence labels are summarized to form complete stratigraphic characteristic data.

[0139] The beneficial effects of the above technologies are as follows: based on experimental soil data and all second environmental deformation cluster sets, the stratigraphic characteristic data of the strata to be monitored can be determined, thereby achieving a refined and structured characterization of stratigraphic characteristics, providing a scientific basis for stratigraphic deformation monitoring, and improving monitoring accuracy. Example 7:

[0140] Based on Example 1, the real-time monitoring module includes:

[0141] Formation deformation prediction sub-model unit: The environmental feature vectors of all sampling points in each second environmental deformation cluster set, and the formation characteristic sub-vectors of all sampling points in convergent or non-convergent formation data are used as inputs to the formation deformation prediction sub-model. The formation deformation vectors of all sampling points in each second environmental deformation cluster set are used as outputs to construct the formation deformation prediction sub-model for each second environmental deformation cluster set.

[0142] Stratigraphic Deformation Prediction Model Unit: Based on the stratigraphic deformation prediction sub-model of all second-environment deformation cluster sets, the stratigraphic deformation prediction model of the stratum to be monitored is determined.

[0143] Real-time monitoring unit: Based on intelligent sensor array, it monitors real-time environmental data and real-time formation deformation data of the formation to be monitored.

[0144] Prediction Unit: Inputs real-time environmental data and real-time stratum deformation data into the stratum deformation prediction model. Based on the output of the stratum deformation prediction model, it determines the predicted deformation data and deformation early warning data of the stratum to be monitored. Based on the predicted deformation data and deformation early warning data, it realizes real-time monitoring of stratum deformation of the stratum to be monitored.

[0145] In this embodiment, the intelligent sensor group includes displacement sensors, humidity sensors, tilt sensors, etc., deployed on the stratum to be monitored. The type, deployment method, and data transmission method of the intelligent sensors can be determined and implemented by those skilled in the art according to actual needs.

[0146] In this embodiment, for each second environmental deformation cluster set, the environmental feature vector and formation characteristic sub-vector of all sampling points are used as input, and the formation deformation vector is used as output to construct a formation deformation prediction sub-model. The formation deformation prediction sub-model can be a physical guided neural network model, a spatiotemporal graph convolutional network model, etc.

[0147] In this embodiment, each second environmental deformation cluster set corresponds to a sub-model, enabling refined prediction of different geological units.

[0148] In this embodiment, a comprehensive formation deformation prediction model for the formation to be monitored is constructed based on the formation deformation prediction sub-models of all second-environment deformation clusters. A final formation deformation prediction model is determined by analyzing the sub-models of all second-environment deformation clusters. Specifically, by integrating the outputs of each sub-model and combining their predictive capabilities for formation deformation, an overall model is constructed. This model comprehensively considers the characteristics and deformation patterns of different clusters, providing a comprehensive prediction of the deformation of the formation to be monitored. This comprehensive model can fuse prediction information from different clusters, making the prediction more comprehensive and highly reliable, especially in handling complex formation environments, where it can better capture the combined influence of multiple factors on formation deformation. For example, suppose there are three second-environment deformation clusters in the formation, each with its own prediction sub-model. The algorithm integrates these sub-models to form a comprehensive prediction model that reflects the overall deformation pattern of the formation.

[0149] In this embodiment, a smart sensor array is used to monitor real-time environmental data (such as temperature, humidity, rainfall, and groundwater level) and real-time stratum deformation data (such as displacement, strain, and dip angle) of the stratum to be monitored. The arrangement and acquisition frequency of the smart sensor array are adjusted according to the specific needs of the monitoring target of the stratum to be monitored, so as to ensure that sufficiently accurate and timely stratum information can be monitored.

[0150] In this embodiment, the intelligent sensor group transmits the collected data in real time through a wireless communication module.

[0151] In this embodiment, the prediction unit utilizes real-time environmental data and real-time formation deformation data to predict formation deformation and potential deformation warnings through a formation deformation prediction model. During this process, the prediction unit inputs the real-time monitored environmental data and formation deformation data into the formation deformation prediction model. This input data is compared and calculated with historical data already trained in the model, outputting predicted deformation data for the formation to be monitored. This predicted data includes not only the amount of formation deformation (such as settlement and displacement) but also deformation-related warning data, helping engineers or monitoring personnel identify potential risks.

[0152] In this embodiment, based on the predicted deformation data and deformation early warning data, the monitoring system can provide early warning information in real time, indicating possible dangerous stratum deformation.

[0153] In this embodiment, the predicted deformation data includes at least the following: deformation amount prediction: including horizontal displacement prediction value and vertical displacement prediction value, specifically indicating the magnitude of the displacement of the strata in the horizontal and vertical directions within a specific time period in the future; deformation rate prediction: predicting the rate of change of strata deformation over time, such as predicting that the subsidence rate of strata in a certain area will increase from the current 1 mm / day to 3 mm / day in the next three days; deformation trend prediction: analyzing and predicting the long-term trend of strata deformation, such as predicting that in the next year or several years, with the continuous decline of groundwater level.

[0154] In this embodiment, the deformation early warning data includes at least the early warning level: it can be divided into three levels: Level 1 (yellow warning), Level 2 (orange warning), and Level 3 (red warning); early warning time: clearly defining the time when the early warning information is released and the predicted time when the dangerous situation may occur; early warning area: accurately defining the affected area and clearly indicating which areas or buildings are within the early warning range through geographical coordinates, area names, etc.; response suggestions: providing specific response suggestions and measures.

[0155] The beneficial effects of the above technologies are: real-time acquisition of real-time environmental data and real-time formation deformation data of the strata to be monitored; and real-time monitoring of formation deformation based on real-time environmental data, real-time formation deformation data, all second-environment deformation clusters, and formation characteristic data. This enables precise matching of complex geological and environmental conditions, improves the accuracy of formation deformation prediction and risk prevention capabilities, and enhances the system's adaptability and reliability. Example 8:

[0156] The present invention provides a real-time monitoring device for formation deformation based on intelligent sensors, which is used to execute any one of the real-time monitoring systems for formation deformation based on intelligent sensors in Examples 1 to 7.

[0157] The beneficial effects of the above technology are as follows: By analyzing historical environmental data and historical stratigraphic deformation data, multiple second-environment deformation clusters are determined. Based on collected experimental soil data and all second-environment deformation clusters, stratigraphic characteristic data of the stratum to be monitored are determined. Real-time environmental data and real-time stratigraphic deformation data of the stratum to be monitored are collected. By combining all second-environment deformation clusters and stratigraphic characteristic data, real-time monitoring of stratigraphic deformation of the stratum to be monitored can be achieved. This enables refined monitoring and data processing of the complex relationship between stratigraphy, environment, and deformation, accurately identifies stratigraphic deformation patterns, achieves accurate deformation prediction and dynamic early warning, improves monitoring accuracy, monitoring timeliness, and risk prevention and control capabilities, and reduces the risk of geological disasters.

[0158] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A real-time monitoring system for formation deformation based on intelligent sensors, characterized in that, include: Acquisition module: Collects test soil data of the strata to be monitored, and obtains historical environmental data and historical stratum deformation data of the strata to be monitored; Clustering Analysis Module: Analyzes historical environmental data and historical stratigraphic deformation data to determine multiple environmental cluster sets and multiple deformation cluster sets. Based on all environmental cluster sets and all deformation cluster sets, it determines multiple second environmental deformation cluster sets. Stratigraphic characteristics module: Based on experimental soil data and all second environmental deformation cluster sets, determine the stratigraphic characteristics data of the strata to be monitored; Real-time monitoring module: Real-time acquisition of real-time environmental data and real-time formation deformation data of the formation to be monitored. Based on real-time environmental data, real-time formation deformation data, all second environmental deformation cluster sets and formation characteristic data, real-time monitoring of formation deformation of the formation to be monitored is realized. The data acquisition module includes: Key area units: Obtain geological exploration data and engineering design data of the strata to be monitored, and identify multiple key areas of the strata to be monitored based on the geological exploration data; Grid division unit: The first grid is used to divide all key areas of the stratum to be monitored, and the second grid is used to divide the other areas of the stratum to be monitored except for all key areas. Sampling point unit: Based on the strata to be monitored after the first grid division and the second grid division, multiple sampling points are determined for the strata to be monitored; Experimental sub-soil unit: Soil is collected from each sampling point of the stratum to be tested, and the experimental sub-soil and experimental sub-soil data for each sampling point are determined. The experimental sub-soil data includes the sampling point number, sampling time, sampling depth, and sampling location. Sample soil data unit: Based on the test sub-soil and test sub-soil data of all sampling points of the stratum to be tested, the sample soil data of the stratum to be tested is determined.

2. The real-time monitoring system for formation deformation based on intelligent sensors according to claim 1, characterized in that, The data acquisition module also includes: Historical environmental data unit: Acquire historical environmental sub-data for each sampling point in the stratum to be monitored, and determine the historical environmental data of the stratum to be monitored based on the historical environmental sub-data of all sampling points in the stratum to be monitored; Historical stratigraphic deformation data unit: Acquire historical stratigraphic deformation sub-data for each sampling point in the stratigraphic unit to be monitored, and determine the historical stratigraphic deformation data of the stratigraphic unit to be monitored based on the historical stratigraphic deformation sub-data for all sampling points in the stratigraphic unit to be monitored.

3. The real-time monitoring system for formation deformation based on intelligent sensors according to claim 2, characterized in that, Analysis of clustering module, include: Environmental feature vector unit: Extract features from the historical environmental sub-data of each sampling point in the historical environmental data of the stratum to be monitored, and determine the environmental feature vector of each sampling point in the stratum to be monitored; First cluster analysis unit: Based on the environmental feature vectors of all sampling points in the stratum to be monitored, a first cluster analysis is performed on all sampling points in the stratum to be monitored. Based on the results of the first cluster analysis, multiple environmental cluster sets are determined. Each environmental cluster set includes multiple sampling points and the environmental feature vector of each sampling point. Stratigraphic deformation vector unit: Features are extracted from the historical stratigraphic deformation sub-data of each sampling point in the historical stratigraphic deformation data of the stratum to be monitored to determine the stratigraphic deformation vector of each sampling point in the stratum to be monitored; The second cluster analysis unit: Based on the formation deformation vectors of all sampling points in the formation to be monitored, a second cluster analysis is performed on all sampling points in the formation to be monitored. Based on the results of the second cluster analysis, multiple deformation cluster sets are determined. Each deformation cluster set includes multiple sampling points and the formation deformation vector of each sampling point.

4. The real-time monitoring system for formation deformation based on intelligent sensors according to claim 3, characterized in that, The clustering analysis module also includes: Calculation unit: Based on all environment cluster sets and all deformed cluster sets, calculate the first set consistency value and the second set consistency value for each environment cluster set and each deformed cluster set, and calculate the sampling weight for each sampling point; Unit determination: Based on the first set consensus value and the second set consensus value of all environmental cluster sets and deformation cluster sets, as well as the sampling weights of all sampling points, determine the important sampling point set and determine multiple first environmental deformation cluster sets, wherein the first environmental deformation cluster set includes multiple sampling points and the environmental feature vector and the formation deformation vector of each sampling point; Membership value unit: For each sampling point that is not in any of the first environment deformation cluster sets, calculate the membership value of each sampling point and each environment deformation cluster set; Second environmental deformation cluster set unit: Based on the membership values ​​of all sampling points not in any environmental deformation cluster set and all first environmental deformation cluster sets, multiple second environmental deformation cluster sets are determined. The second environmental deformation cluster set includes multiple sampling points and the environmental feature vector and formation deformation vector of each sampling point. Non-convergent unit: For all sampling points that are not in any second environmental deformation cluster set, as well as the environmental feature vector and formation deformation vector of each sampling point, determine a second environmental deformation cluster set and mark it as non-convergent.

5. The real-time monitoring system for formation deformation based on intelligent sensors according to claim 1, characterized in that, The formation characteristics module includes: Physical property vector unit: Physical property tests are performed on the test sub-soil at each sampling point in the sample soil data to determine the physical property vector of the test sub-soil at each sampling point; Mechanical property vector unit: The mechanical properties of the test sub-soil at each sampling point in the soil sample data are tested to determine the mechanical property vector of the test sub-soil at each sampling point; Microscopic property vector unit: Perform microstructure analysis on the test sub-soil at each sampling point in the soil sample data to determine the microscopic property vector of the test sub-soil at each sampling point; Stratigraphic characteristic sub-vector unit: Based on the physical property vector, mechanical property vector and microscopic property vector of the test sub-soil at each sampling point in the soil sample data, the stratigraphic characteristic sub-vector of the test sub-soil at each sampling point in the soil sample data is determined. Converging stratigraphic data unit: Based on all sampling points in each second environmental deformation cluster set without non-convergence labels and the stratigraphic characteristic sub-vector of each sampling point, the convergent stratigraphic data of each second environmental deformation cluster set without non-convergence labels is determined; Non-convergent stratigraphic data unit: Based on all sampling points of the second environmental deformation cluster set with non-convergent labels and the stratigraphic characteristic sub-vector of each sampling point, non-convergent stratigraphic data of the second environmental deformation cluster set with non-convergent labels are determined; Stratigraphic characteristic data unit: Based on the convergent stratigraphic data of all second environmental deformation clusters without non-convergence labels, and the non-convergence stratigraphic data of second environmental deformation clusters with non-convergence labels, the stratigraphic characteristic data is determined.

6. The real-time monitoring system for formation deformation based on intelligent sensors according to claim 1, characterized in that, The real-time monitoring module includes: Formation deformation prediction sub-model unit: The environmental feature vectors of all sampling points in each second environmental deformation cluster set, and the formation characteristic sub-vectors of all sampling points in convergent or non-convergent formation data are used as inputs to the formation deformation prediction sub-model. The formation deformation vectors of all sampling points in each second environmental deformation cluster set are used as outputs to construct the formation deformation prediction sub-model for each second environmental deformation cluster set. Stratigraphic Deformation Prediction Model Unit: Based on the stratigraphic deformation prediction sub-model of all second-environment deformation cluster sets, the stratigraphic deformation prediction model of the stratum to be monitored is determined; Real-time monitoring unit: Based on intelligent sensor array, it monitors the real-time environmental data and real-time formation deformation data of the formation to be monitored. Prediction Unit: Inputs real-time environmental data and real-time stratum deformation data into the stratum deformation prediction model. Based on the output of the stratum deformation prediction model, it determines the predicted deformation data and deformation early warning data of the stratum to be monitored. Based on the predicted deformation data and deformation early warning data, it realizes real-time monitoring of stratum deformation of the stratum to be monitored.

7. A real-time monitoring device for formation deformation based on intelligent sensors, characterized in that, Used to implement the real-time monitoring system for formation deformation based on smart sensors according to any one of claims 1 to 6.

Images