Settlement monitoring method and system based on data provenance

Abstract

The application provides a kind of based on data traceability's subsidence monitoring method and system, first obtain the time series subsidence observation data of each monitoring point in subsidence monitoring area in continuous observation period, including subsidence and corresponding observation time environmental disturbance information, then build subsidence data traceability association network, associate time series subsidence observation data with monitoring point geological basement information, observation equipment traceability identifier and data processing node information, then trace the generation link of each monitoring point subsidence data, extract equipment calibration record etc., generate traceability link feature, then coupling analysis traceability link feature and subsidence, determine traceability reliability index, finally according to index screening data, generate dynamic subsidence trend model and subsidence monitoring analysis report containing traceability link feature in combination with monitoring point spatial distribution characteristics, improve the accuracy and reliability of subsidence monitoring.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer, in particular, to a settlement monitoring method and system based on data provenance. BACKGROUND

[0002] In the field of settlement monitoring, accurately obtaining and analyzing settlement data is of great significance for evaluating the stability of engineering structures, geological disaster warning, etc. Currently, settlement monitoring mainly relies on arranging multiple monitoring points in the monitoring area and using various observation equipment to obtain settlement observation data of each monitoring point. However, the existing settlement monitoring method has many limitations.

[0003] On the one hand, the traditional settlement monitoring data acquisition method often only focuses on the settlement itself, and the environmental interference information during the observation process is not recorded. Environmental factors such as temperature changes, groundwater level fluctuations, and surrounding construction activities can have a significant impact on settlement observation data, but these interference information is often ignored in data collection and analysis, making it difficult to guarantee the accuracy and reliability of settlement data.

[0004] On the other hand, the existing settlement data analysis method lacks the ability to trace the data generation process. Settlement data goes through multiple processing links from collection to final use, including device calibration, data correction, etc. However, the traditional method cannot effectively associate the information of these links, making it difficult to determine whether the data has been disturbed improperly during the generation process and how much the disturbance is. This makes the reliability assessment of settlement monitoring results lack scientific basis, and further affects the accuracy and effectiveness of the decisions made based on settlement monitoring data. SUMMARY

[0005] In view of the above-mentioned problems, in combination with the first aspect of the present application, the embodiments of the present application provide a settlement monitoring method based on data provenance, which comprises:

[0006] Obtaining time-series settlement observation data of each monitoring point in the settlement monitoring area, the time-series settlement observation data containing the settlement amount of each monitoring point in the continuous observation period and the environmental interference information corresponding to the observation time;

[0007] Constructing a settlement data provenance association network, associating the time-series settlement observation data with the geological basement information of the monitoring point, the observation equipment provenance identifier and the data processing node information, forming settlement data provenance association network information containing data sources and associated elements;

[0008] Tracing the generation link of each monitoring point settlement data based on the settlement data provenance association network information, extracting the device calibration record, environmental disturbance source intensity and data correction operation record during data collection in different observation periods, and generating settlement data provenance link features;

[0009] coupling analysis of the sedimentation data provenance link feature and the corresponding sedimentation amount, determining a provenance reliability index of the sedimentation data of each monitoring point, the provenance reliability index being used to represent the degree of interference of the sedimentation data in the generation link;

[0010] According to the provenance reliability index, sedimentation data meeting the threshold requirement of the provenance reliability index is screened out, a dynamic sedimentation trend model of the sedimentation monitoring area is generated in combination with the spatial distribution characteristics of the monitoring points, and a sedimentation monitoring analysis report containing the sedimentation data provenance link feature is generated.

[0011] In still another aspect, the embodiment of the present application also provides a sedimentation monitoring system based on data provenance, characterized in that it comprises:

[0012] a processor; a machine readable storage medium for storing machine executable instructions of the processor; wherein the processor is configured to execute the above-mentioned sedimentation monitoring method based on data provenance by executing the machine executable instructions.

[0013] In still another aspect, the embodiment of the present application also provides a computer program product, the computer program product comprising machine executable instructions stored in a computer readable storage medium, a processor of a computer device reading the machine executable instructions from the computer readable storage medium, and the processor executing the machine executable instructions so that the computer device executes the above-mentioned sedimentation monitoring method based on data provenance.

[0014] Based on the above aspects, by acquiring the time series sedimentation observation data of each monitoring point in the sedimentation monitoring area and recording the environmental interference information of each observation time in detail, a sedimentation data provenance correlation network is constructed, the time series sedimentation observation data is associated with the geological basement information of the monitoring point, the observation equipment provenance identifier and the data processing node information, a comprehensive information network containing data sources and associated elements is formed, the generation link of the sedimentation data of each monitoring point is traced based on the provenance correlation network information, key information such as equipment calibration records, environmental interference source intensity and data correction operation records is extracted, sedimentation data provenance link features are generated, and each link in the data generation process can be accurately located. Coupling analysis of the provenance link feature and the corresponding sedimentation amount determines the provenance reliability index, which scientifically represents the degree of interference of the sedimentation data in the generation link. According to the provenance reliability index, sedimentation data meeting the requirement is screened out, a dynamic sedimentation trend model is generated in combination with the spatial distribution characteristics of the monitoring points, and a sedimentation monitoring analysis report containing the provenance link feature is generated, which provides accurate, reliable and traceable results for sedimentation monitoring. BRIEF DESCRIPTION OF DRAWINGS

[0015] Figure 1is an execution flow diagram of the data source-based settlement monitoring method provided by an embodiment of the present application.

[0016] Figure 2 is a schematic diagram of exemplary hardware and software components of the data source-based settlement monitoring system provided by an embodiment of the present application. DETAILED DESCRIPTION

[0017] The present application will be described in detail below with reference to the accompanying drawings, Figure 1 is a flow diagram of the data source-based settlement monitoring method provided by an embodiment of the present application, which will be described in detail below.

[0018] Step S110: Obtain time-series settlement observation data of each monitoring point in the settlement monitoring area, wherein the time-series settlement observation data includes the settlement amount of each monitoring point in a continuous observation period and environmental interference information corresponding to the observation time.

[0019] Taking a settlement monitoring area along a subway line in a city as an application scenario, monitoring points are arranged along the subway line, and the spacing between the monitoring points is set according to the complexity of the geological conditions. The spacing between monitoring points is smaller in a complex geological condition section and larger in a stable geological condition section. Each monitoring point is installed with an automatic settlement monitoring device, and the observation period is set to once a day, and the observation time is selected in the early morning period when the external environment is relatively stable.

[0020] The time-series settlement observation data is collected by the built-in sensor of the device, the settlement amount is obtained by a high-precision displacement measurement module, the unit is millimeter, and the data is retained to three decimal places. The environmental interference information is collected by an integrated multi-parameter sensor, including the atmospheric temperature, relative humidity, atmospheric pressure, ground vibration acceleration around the monitoring point, underground water level depth and other parameters at the observation time. Among them, the atmospheric temperature unit is Celsius, retained to one decimal place; the relative humidity is expressed in percentage, retained to an integer; the atmospheric pressure unit is hundred pascal, retained to an integer; the ground vibration acceleration is collected by a vibration sensor and converted into a dimensionless vibration level, which is divided into multiple intervals; the underground water level depth unit is meter, retained to two decimal places.

[0021] After data collection is completed, the device automatically adds a time stamp to the original data, the time stamp is accurate to seconds, and is transmitted to the data center through an encrypted wireless network. After receiving the data, the data center first performs data integrity check, checks whether the data field is complete and the format is correct, and sends retransmission instructions to the monitoring device for incomplete or incorrect format data. The data that passes the check is stored in a distributed database, and is stored in partitions according to the monitoring point number and the observation date, which is convenient for subsequent data query and processing.

[0022] Step S120: Construct a settlement data provenance correlation network, correlate the time-series settlement observation data with the geological basement information of the monitoring point, the observation equipment provenance identifier, and the data processing node information, and form the settlement data provenance correlation network information containing data sources and correlation elements.

[0023] In the above subway line settlement monitoring scene, in order to realize the full-link tracing of the settlement data from generation to processing, it is necessary to construct a settlement data provenance correlation network to correlate the time-series settlement observation data with multi-dimensional elements affecting data quality.

[0024] Step S121: Analyze the metadata field of the time-series settlement observation data, extract the location coding information and observation period identifier of each monitoring point, and establish the index system of the time-series settlement observation data.

[0025] The metadata field is extracted from the time-series settlement observation data stored in the data center, and the metadata field includes the monitoring point number, observation date, observation time, data acquisition equipment number, data transmission state, etc. Among them, the monitoring point number is composed of letters and numbers, the first two letters represent the monitoring area code, and the last four digits represent the monitoring point serial number; the observation date is represented by a specific format string; the observation time is accurate to the minute. According to the monitoring point number, the location coding information is extracted, which includes the line number, mileage pile number, offset, etc. Parameters, through which the specific position of the monitoring point in the subway line can be uniquely determined. The observation period identifier is composed of the observation date and the observation period code, and the observation period code is divided according to different observation time intervals in a day. Based on the location coding information and the observation period identifier, the index system of the time-series settlement observation data is established. The index system adopts a multi-level index structure, the first level index is the monitoring point number, the second level index is the observation date, and the third level index is the observation period code. Through the index system, the settlement observation data of any monitoring point in any observation period can be quickly located.

[0026] Step S122: Collect the geological basement information of each monitoring point, which includes the stratum bearing capacity distribution, the rock-soil compression parameter, and the geological structure fracture zone distribution, and assign a unique geological information code to each monitoring point.

[0027] Step S1221: Obtain the geological survey report and drilling data of the settlement monitoring area, and extract the physical and mechanical property parameters of each soil layer from them, including the soil density, internal friction angle, and cohesion.

[0028] The detailed geological survey report and drilling data of the settlement monitoring area along the subway line are obtained by accessing the urban geological survey database or calling the survey unit. The geological survey report contains drill column chart, soil test result table and other contents; the drilling data includes the depth of each drill hole, soil layer name, layer bottom elevation, sampling location and other information. The physical and mechanical property parameters of each soil layer are extracted from the soil test result table, the soil density is determined by the ring cutter method test, which represents the weight of unit volume of soil; the internal friction angle is determined by the direct shear test or triaxial compression test, which reflects the shear strength characteristics of the soil; the cohesion is also obtained by the shear strength test, which characterizes the cohesive force between soil particles. The above parameters are classified and arranged according to the soil layer name and depth, and a soil layer physical and mechanical property parameter table is established, each record corresponds to the parameter value of a soil layer at a specific depth.

[0029] Step S1222: Calculate the stratum bearing capacity characteristic value of each drilling point based on the physical and mechanical property parameters, and use the spatial interpolation method to extend the discrete drilling point bearing capacity characteristic value to the entire monitoring area to generate a stratum bearing capacity distribution raster map.

[0030] According to the calculation formula of stratum bearing capacity characteristic value in the geological survey specification, the physical and mechanical property parameters such as soil density, internal friction angle and cohesion extracted in step S1221 are used to calculate the stratum bearing capacity characteristic value of each soil layer at each drilling point. For different depths of soil layers at the same drilling point, their bearing capacity characteristic values are calculated respectively, and the corresponding soil layer thickness is marked. The Kriging interpolation method in the spatial interpolation method is used to spatially extend the discrete drilling point bearing capacity characteristic value. First, the interpolation area range is determined as the entire settlement monitoring area, and the size of the interpolation grid is set. The grid cell length is determined according to the drilling point density. By constructing a semi-variogram function model, the spatial correlation of the bearing capacity characteristic value is analyzed, and then the bearing capacity characteristic value of each grid cell is predicted according to the bearing capacity values of the surrounding known drilling points, and finally a stratum bearing capacity distribution raster map is generated. The value of each pixel in the raster map represents the stratum bearing capacity characteristic value at that position, and the pixel value is represented by different color gradients, which facilitates intuitive viewing of the spatial distribution difference of the bearing capacity.

[0031] Step S1223: Extract the compression modulus and compression coefficient of the rock-soil mass from the drilling data, divide the rock-soil mass at different depths into a preset number of compression grades according to the compression modulus value, and divide the monitoring area into multiple compression subareas according to the compression grade to form a rock-soil mass compression parameter distribution map.

[0032] The compression modulus of each soil layer is extracted from the geotechnical test results of the drilling data. The compression modulus reflects the ability of the soil to compress under lateral restraint, and the compression coefficient represents the change rate of the void ratio of the soil under the action of unit pressure increment. The compression modulus is divided into a preset number of compression grades in order from small to large, for example, divided into five grades of high compression, medium-high compression, medium compression, medium-low compression and low compression, each grade corresponding to a compression modulus range. According to the compression modulus value and the distribution depth of each soil layer of each drilling point, the dominant compression grade of the location of the drilling point (i.e. the grade corresponding to the compression grade soil layer with the largest proportion) is determined. Using the polygon partitioning method, the areas with the same or similar compression grade in the monitoring area are connected to form a closed polygon, forming multiple compression zoning areas, each zoning area is labeled with the corresponding compression grade and representative compression modulus and compression coefficient value, thereby generating a geotechnical compression parameter distribution map.

[0033] Step S1224: Analyze the geological structure description in the geological survey report, identify the location, trend and tendency of the fault zone in the monitoring area in combination with the seismic exploration data, determine the influence width and activity characteristics of the fault zone through the activity frequency and influence range of the fault zone, and generate a geological structure fault zone distribution map.

[0034] Carefully read the chapter on regional geological structure in the geological survey report, and extract information such as the name, approximate location and scale of the fault zone described therein. In combination with seismic exploration data such as reflection seismic profile of the region, the spatial position, trend (extension direction of the fault zone in the plane) and tendency (inclination direction of the fault surface) of the fault zone are accurately identified by analyzing the abnormal change area of the seismic wave propagation velocity. For each identified fault zone, collect its historical activity records, count the activity frequency (such as the number of significant activities within a hundred years), and evaluate its influence range according to the scale and activity intensity of the fault zone. The determination of the influence width considers the actual width of the fault zone and the possible secondary influence area, and the activity characteristics include the activity mode (such as creep, sudden dislocation) and average sliding rate and other parameters. In the geographic information system software, the planar position of the fault zone is drawn in proportion, different line types are used to represent fault zones with different activity characteristics, and the name, trend angle, tendency angle, influence width and other attributes of the fault zone are labeled, to generate a geological structure fault zone distribution map.

[0035] Step S1225: Spatially superimpose the stratum bearing capacity distribution grid map, the geotechnical compression parameter distribution map and the geological structure fault zone distribution map to form a comprehensive geological basement information map.

[0036] The stratum bearing capacity distribution grid map generated in step S1222, the rock-soil compressibility parameter distribution map formed in step S1223 and the geological structure fracture zone distribution map obtained in step S1224 are loaded into the same coordinate system by using the spatial overlay function of the geographic information system software. First, coordinate registration is performed to ensure that the spatial positions of the three maps accurately correspond, and then a layer overlay operation is performed to fuse the spatial elements and attribute information of each layer. In the comprehensive geological base information map, the bearing capacity values of the grid cells, the compressibility partitions of the polygons and the linear elements of the fracture zones are displayed simultaneously, and the transparency of each layer is set to make different types of geological information clear and visible. The comprehensive geological base information map can intuitively reflect the spatial relationship between the stratum bearing capacity, the rock-soil compressibility and the geological structure in the monitoring area.

[0037] Step S1226: The comprehensive geological base information map is gridized according to a preset grid precision, and each grid cell corresponds to a unique spatial coordinate range.

[0038] According to the accuracy requirements of settlement monitoring and the convenience of data processing, a preset grid precision is set, for example, the grid cell length is a specific length. A regular square grid is created on the comprehensive geological base information map to cover the entire monitoring area. Each grid cell is defined by its unique spatial coordinate range through the coordinates of the upper left corner and the lower right corner, and the coordinates adopt the Gauss-Kruger projection coordinate system. The grid cell number is identified by row and column, for example, the grid cell number of the mth row and the nth column is a specific format. For each grid cell, the average stratum bearing capacity value, the main compressibility grade and whether it contains a fracture zone and other information within its range are extracted as the attribute data of the grid cell.

[0039] Step S1227: A geological information code is assigned to each grid cell, and the geological information code rule includes the stratum bearing capacity grade code, the rock-soil compressibility grade code and the fracture zone influence degree code of the region.

[0040] A geological information code rule is formulated, and the code length is fixed for several digits, each digit representing different geological meanings. For example, the first two digits are the stratum bearing capacity grade code, which is coded according to the interval to which the average stratum bearing capacity characteristic value within the grid cell belongs, such as one code for high bearing capacity, another code for medium-high bearing capacity, etc.; the middle two digits are the rock-soil compressibility grade code, which is coded according to the main compressibility grade of the grid cell; and the last two digits are the fracture zone influence degree code, which is divided into strong influence, medium influence and slight influence grades and assigned corresponding codes according to the fracture zone influence width and activity characteristics if the grid cell contains a fracture zone, or coded as a specific value if there is no fracture zone. Each grid cell obtained in step S1226 is assigned a unique geological information code according to this rule, the code is stored as a string format, and is stored in association with the spatial coordinate range of the grid cell.

[0041] Step S1228: match the position coordinates of each monitoring point with the spatial coordinate range of the grid unit, so that each monitoring point corresponds to a unique grid unit, thereby obtaining the geological information code of the monitoring point and the corresponding stratum bearing capacity distribution, rock-soil compressibility parameters and geological structure fracture zone distribution.

[0042] The precise plane coordinates (such as X and Y coordinate values) of each monitoring point are extracted from the position encoding information of the monitoring point. In the geographic information system software, the monitoring point coordinates are matched with the grid units generated in step S1226 in terms of spatial position. Through spatial relationship judgment of points and polygons, it is determined which grid unit the spatial coordinate range of each monitoring point falls into, so that the monitoring point corresponds to a unique grid unit. According to the geological information code associated with the grid unit, the geological information code of the monitoring point is obtained. At the same time, the corresponding stratum bearing capacity distribution characteristics (such as average bearing capacity value, bearing capacity level), rock-soil compressibility parameters (such as main compressibility level, representative compression modulus) and geological structure fracture zone distribution (such as whether affected by fracture zone, influence degree code) are extracted from the attribute data of the grid unit, and the above geological basement information is bound with the monitoring point number and stored in the monitoring point geological information table.

[0043] Step S123: obtain the traceability identification of the observation equipment used by each monitoring point, which includes the equipment production batch number, the serial number of the calibration certificate and the equipment operation and maintenance record number, and establish the full life cycle traceability file of the observation equipment.

[0044] The traceability identification information of the observation equipment used by each monitoring point is retrieved from the equipment management department. The equipment production batch number is provided by the manufacturer and includes information such as production year, production line code and batch number. The serial number of the calibration certificate is obtained from the calibration record account book of the equipment, and a unique certificate number is issued by the measurement agency after each calibration. The equipment operation and maintenance record number is generated by the equipment maintenance department in the order of maintenance, recording the installation, repair and replacement of parts of the equipment. The above traceability identification information is associated with the equipment serial number to establish the full life cycle traceability file of the observation equipment. The file is stored in a relational database, the main table records the basic information of the equipment (equipment model, serial number, manufacturer, purchase date), and the sub-tables record the production batch information, calibration record and operation and maintenance record respectively. The association query between the main table and the sub-tables is realized through the equipment serial number.

[0045] Step S124: collect all data processing node information involved in the processing process of the time series settlement observation data, which includes node processing algorithm version, processing personnel qualification number and processing timestamp, to form a data processing link list.

[0046] The processing flow of the time series subsidence observation data from the original data to the final result data is tracked, and all data processing nodes involved in the processing process are identified. The data processing nodes include data preprocessing nodes, data quality checking nodes, data correction nodes, data statistical analysis nodes, etc. Each data processing node is identified by a unique node number. For each data processing node, the node processing algorithm version, the processing personnel qualification number, and the processing timestamp are collected. The node processing algorithm version refers to the software algorithm version number used by the node; the processing personnel qualification number refers to the professional qualification certificate number of the personnel operating the processing node; and the processing timestamp refers to the start time and end time of the processing of the data by the node, accurate to the second. According to the sequence of data processing, the information of each data processing node is sorted to form a data processing link list.

[0047] Step S125: Establishing the association relationship between the time series subsidence observation data and the geological basement information, binding the time series subsidence observation data of each monitoring point to the corresponding stratum bearing capacity distribution, rock-soil compression parameters, and geological structure fracture zone distribution through the mapping of the monitoring point location coding information and the geological information coding.

[0048] Based on the monitoring point location coding information extracted in step S121 and the geological information coding assigned in step S122, the mapping relationship between the two is established. The mapping relationship is realized through an association table, which contains fields such as monitoring point number, location coding information, and geological information coding. By querying the association table, the corresponding geological information coding can be found according to the location coding information of the monitoring point. The corresponding geological basement information such as stratum bearing capacity distribution, rock-soil compression parameters, and geological structure fracture zone distribution is retrieved from the geological database according to the geological information coding. The above-mentioned geological basement information is bound with the time series subsidence observation data of the monitoring point. The binding method is to add a geological information coding field in the attribute field of the time series subsidence observation data, and to establish an association index between the time series subsidence observation data record and the geological basement information record.

[0049] Step S126: Establishing the association relationship between the time series subsidence observation data and the observation equipment traceability identifier, binding the subsidence data of each observation period to the corresponding equipment production batch number and the previous calibration certificate number through the matching of the observation period identifier and the equipment operation and maintenance record number.

[0050] Obtain the observation period identifier from the time series settlement observation data index system established in step S121, and obtain the equipment operation and maintenance record number from the observation equipment full life cycle traceability archive established in step S123. The equipment operation and maintenance record number contains time information such as the installation date and maintenance date of the equipment, which is matched with the observation date in the observation period identifier. The matching rule is: for the settlement data of a certain observation period, find the operation and maintenance record number of the nearest last equipment maintenance before the observation period, and bind the equipment production batch number and the calibration certificate number of each time corresponding to the operation and maintenance record number to the settlement data of the observation period.

[0051] Step S127: Establish the association relationship between the time series settlement observation data and the data processing node information. According to the sequence of data processing, associate the original observation data with the algorithm version, processing personnel qualification number and processing time stamp of the first processing node, and then sequentially associate the information of the previous processing node with the information of the next processing node.

[0052] According to the data processing link list formed in step S124, determine the data processing node sequence experienced by the time series settlement observation data from the original data to the final processing result. The original observation data refers to the data that is directly transmitted from the monitoring equipment to the data center without any processing, and the data identifier is in the input data identifier field of the data processing link list. Associate the original observation data with the first processing node in the data processing link list, and the association content includes the algorithm version, processing personnel qualification number and processing time and end time of the first processing node. The association method is to add the related information field of the first processing node in the metadata of the original observation data. Then, according to the sequence of data processing, match the output data identifier of the previous processing node with the input data identifier of the next processing node, so as to associate the information of the previous processing node with the information of the next processing node.

[0053] Step S128: Take the monitoring point as the core node, take the time series settlement observation data, geological basement information, observation equipment traceability identifier and data processing node information as the association edge, and construct a multi-dimensional settlement data traceability association network. Each association edge in the settlement data traceability association network carries a corresponding association strength parameter, forming settlement data traceability association network information containing data sources and associated elements.

[0054] The monitoring point is taken as the core node of the settlement data traceability association network, and the attributes of the core node include the monitoring point number, position coordinates, installation date, etc. The time series settlement observation data, geological basement information, observation equipment traceability identifier and data processing node information are taken as the association edges connected with the core node. Each association edge represents the association relationship between the core node and a certain type of information, and the attributes of the association edge include the association type and the association strength parameter. The association strength parameter is used to quantify the association closeness between the core node and the associated information, and the value range is between 0 and 1. The greater the value, the closer the association. The multi-dimensional settlement data traceability association network constructed in the above manner can directly display the association relationship between the settlement data of each monitoring point and the related factors, and the association network information is stored in a graph database, which is convenient for network topology analysis and data traceability query.

[0055] Step S130: Based on the settlement data traceability association network information, the generation link of each monitoring point settlement data is traced, the equipment calibration record at the time of data collection in different observation periods, the intensity of environmental interference source and the data correction operation record are extracted, and the settlement data traceability link feature is generated.

[0056] Step S131: The observation equipment traceability identifier of each monitoring point is located from the settlement data traceability association network information, the initial calibration parameters of the equipment at the time of factory delivery are queried according to the equipment production batch number, and then the deviation correction value and the calibration environment condition of each calibration are called through the serial number of the calibration certificate, to form the equipment calibration record sequence.

[0057] Through the graph database query function of the settlement data traceability association network information, the corresponding observation equipment traceability identifier association edge is located according to the monitoring point number, and then the equipment production batch number is obtained. The initial calibration parameters of the equipment at the time of factory delivery are queried from the database of the equipment manufacturer according to the equipment production batch number, including the standard values and allowable deviation ranges of zero drift, sensitivity, linearity and other indicators. The serial numbers of the calibration certificates of the observation equipment traceability identifier are extracted and arranged in chronological order. The detailed records of each calibration are called from the database of the metrological calibration institution according to the serial numbers of the calibration certificates, including the calibration date, calibration project, standard equipment used for calibration, deviation correction value of each calibration point, calibration environment condition, etc. The initial calibration parameters of the factory delivery and the calibration records are integrated in chronological order to form the equipment calibration record sequence.

[0058] Step S132: For each observation period, the calibration deviation correction value closest to the observation time is extracted from the equipment calibration record sequence, which is marked as the effective calibration parameter of the observation period, and the interval time length between the calibration time and the observation time is recorded.

[0059] For each observation period of each monitoring point, find the calibration record with the calibration time before the observation time and the minimum interval from the observation time in the device calibration record sequence formed in step S131. Extract the deviation correction value in the calibration record and mark it as the effective calibration parameter of the observation period. The effective calibration parameter is stored by calibration item, such as zero drift deviation correction value, sensitivity deviation correction value, etc. Calculate the interval length of the calibration time and the observation time, the interval length is in units of days, accurate to one decimal place. Record the interval length as an additional attribute of the effective calibration parameter.

[0060] Step S133: Analyze the environmental disturbance information in the settlement data provenance correlation network information, and identify the type of environmental disturbance source that may affect the settlement data in each observation period in combination with the distribution of geological structure fracture zones in the geological basement information. The environmental disturbance source type includes peripheral vibration source, underground water level change and atmospheric pressure fluctuation.

[0061] Extract the environmental disturbance information associated with the time series settlement observation data from the settlement data provenance correlation network information, including ground vibration level, underground water level, atmospheric pressure and other parameters at the observation time. Analyze the relative position relationship between the monitoring point and the fracture zone in combination with the distribution of geological structure fracture zones in the geological basement information, and judge whether the fracture zone activity may affect the monitoring point. According to the parameter characteristics of the environmental disturbance information and the geological structure, identify the type of environmental disturbance source that may affect the settlement data in each observation period. Peripheral vibration sources include subway operation vibration, peripheral construction machinery vibration, road traffic vibration, etc., which are identified by the size and change rule of ground vibration level; underground water level change is identified by the difference in underground water level depth in different observation periods; atmospheric pressure fluctuation is identified by comparing the atmospheric pressure value at the observation time with the historical average atmospheric pressure value.

[0062] Step S134: Through the processing algorithm version in the data processing node information, call the quantitative evaluation result of the environmental disturbance source intensity, which includes the influence range and action intensity of each disturbance source at the observation time, and form the environmental disturbance source intensity data set.

[0063] Obtain data processing node information from the settlement data provenance correlation network information, and extract the processing algorithm version of the data processing node. Determine the algorithm model for quantitatively evaluating the environmental disturbance source intensity according to the processing algorithm version. The algorithm model has been pre-integrated in the data processing software, and different versions of the algorithm model may use different quantitative evaluation methods.

[0064] Step S1341: Analyze the processing algorithm version number in the data processing node information to determine the algorithm model type for evaluating the intensity of the environmental disturbance source, which includes a statistical regression model, a physical field simulation model, and a machine learning prediction model.

[0065] The processing algorithm version number in the data processing node information is analyzed. The version number usually includes a major version number, a minor version number, and a revision number, and different version numbers correspond to different algorithm model types. By querying the correspondence table of algorithm version and model type, the model type of the environmental disturbance source intensity evaluation model used by the current processing algorithm version is determined. For example, if a specific identification character is included in the version number, it corresponds to a statistical regression model; if another identification character is included, it corresponds to a physical field simulation model; if a machine learning related identification is included, it corresponds to a machine learning prediction model. The statistical regression model establishes a regression equation based on the statistical relationship between historical disturbance source intensity and settlement data; the physical field simulation model simulates by establishing mathematical models of physical processes such as vibration wave propagation and groundwater seepage; the machine learning prediction model uses neural networks, support vector machines, and other algorithms to predict the intensity of the disturbance source.

[0066] Step S1342: According to the algorithm model type, the corresponding model input parameters are called, which include vibration frequency monitoring data at the observation time, groundwater level monitoring data, and atmospheric pressure monitoring data.

[0067] According to the algorithm model type determined in step S1341, the corresponding model input parameters are called from the environmental disturbance information of the time series settlement observation data. For the statistical regression model, the input parameters include the vibration frequency monitoring data at the observation time (such as the main frequency and the frequency component of the vibration signal), the groundwater level monitoring data (the difference between the current observation period and the previous period), and the atmospheric pressure monitoring data (the deviation between the current observation time and the daily average atmospheric pressure); for the physical field simulation model, in addition to the above basic monitoring data, medium parameters (such as the density, elastic modulus, and permeability coefficient of the rock-soil body, which are obtained from the geological basement information) also need to be input; for the machine learning prediction model, the input parameters include all the above monitoring data and derived features (such as vibration duration, groundwater level change rate, and atmospheric pressure change trend). The above input parameters are arranged in the required format of the model, converted into numerical arrays or matrices, and the dimensions are unified (such as vibration frequency unit in hertz, groundwater level in meters, and atmospheric pressure in kilopascals).

[0068] Step S1343: Run the algorithm model to calculate the model input parameters and output the intensity quantization value of each environmental disturbance source, which is represented by a preset number of relative grades, and the grade division is based on the influence degree of the disturbance source on the settlement observation.

[0069] The model input parameters after being sorted in step S1342 are input into the corresponding algorithm model for calculation. For a statistical regression model, the influence coefficients of each interference source are calculated by substituting the regression equation, and the intensity quantization value is determined according to the size of the influence coefficient; for a physical field simulation model, the physical quantity (such as vibration acceleration amplitude, pore water pressure change) of the interference source at the monitoring point is obtained by solving the control equation, and then the physical quantity is converted into the intensity quantization value; for a machine learning prediction model, the input parameters are calculated by forward propagation through the trained model, and the intensity prediction value of each interference source is output. The intensity quantization value is divided into a preset number of relative levels, for example, divided into five levels of weak, slight, moderate, relatively strong and strong, and the division threshold is determined according to the analysis results of the influence degree of historical interference source intensity on settlement observation data. The greater the influence degree, the higher the corresponding level.

[0070] Step S1344: Calculate the influence range of each interference source based on the intensity quantization value. The influence range of the vibration source is calculated by a vibration wave propagation and attenuation model, and the model parameters include medium characteristics, amplitude and frequency; the influence range of the underground water level change is calculated by a groundwater seepage model, and the model parameters include permeability coefficient, hydraulic slope and aquifer thickness; the influence range of atmospheric pressure fluctuation is calculated according to the topographic features of the monitoring area, and the topographic features include flat area, hilly area and mountainous area.

[0071] For the vibration source, the method of combining geometric attenuation and material attenuation in the vibration wave propagation and attenuation model is used to calculate the influence range. The medium characteristics (such as damping ratio and shear wave velocity of rock-soil body) in the model parameters are obtained from the physical and mechanical properties of rock-soil body in the geological base information, and the amplitude and frequency are obtained from the vibration frequency monitoring data. The influence range radius is calculated by calculating the energy attenuation of the vibration wave in the propagation process, when the vibration intensity quantization value decreases to the preset level threshold. For the underground water level change, the influence range is calculated by Darcy's law in the groundwater seepage model, the permeability coefficient is extracted from the geological base information according to the type of rock-soil body, the hydraulic slope is calculated from the groundwater level monitoring data (the ratio of water level difference to horizontal distance), and the aquifer thickness is obtained from the geological survey report. The influence radius of the underground water level change is calculated according to the seepage velocity and time. For atmospheric pressure fluctuation, the influence range calculation coefficient is set according to the topographic features of the monitoring area. The influence range of atmospheric pressure in flat area is larger, the influence range in mountainous area is smaller due to the topographic obstruction, and the influence range in hilly area is between the two, and the influence range radius is obtained by multiplying the atmospheric pressure fluctuation amplitude by the influence coefficient of the corresponding topography.

[0072] Step S1345: Associate the intensity quantization value, influence range and corresponding observation time of each environmental interference source to form a single interference source quantization record.

[0073] Create single disturbance source quantitative records for each environmental disturbance source (ambient vibration source, groundwater level change, atmospheric pressure fluctuation), which contain disturbance source type identification (different types represented by different letters), observation time (time stamp accurate to the minute), intensity quantitative value (corresponding relative grade name or grade code), and influence range (radius value in meters). Combine the above information in the order of the fields into structured data, such as in the form of a dictionary or database table record, to ensure that each record corresponds to a unique observation period and disturbance source type.

[0074] Step S1346: Integrate the single disturbance source quantitative records of all environmental disturbance sources, and classify them by disturbance source type. The records of the same type of disturbance source are arranged in chronological order.

[0075] Collect all single disturbance source quantitative records formed in step S1345, classify the records by disturbance source type, and group the records belonging to the same disturbance source type. For example, all records of ambient vibration sources form a vibration source disturbance group, all records of groundwater level changes form a groundwater level disturbance group, and all records of atmospheric pressure fluctuations form an atmospheric pressure disturbance group. Within each disturbance source type group, the records are sorted in chronological order to form a time series. In the case of multiple same-type disturbance sources at the same observation time (e.g., two different location construction vibration sources at the same time), their records are combined, and the maximum or average value of the intensity quantitative value is taken as the comprehensive record of the disturbance source of that type at that time.

[0076] Step S1347: Calculate the comprehensive influence index of different types of disturbance sources at the same observation time. The comprehensive influence index is the weighted sum of the intensity quantitative values of each single disturbance source, and the weight value is determined according to the sensitivity of the disturbance source to the settlement observation.

[0077] For each observation time, extract the intensity quantitative value of that time from the sorted records of each group of disturbance sources in step S1346 (if there is no record of a certain type of disturbance source at that time, the intensity quantitative value is recorded as 0 or the lowest grade). According to the sensitivity analysis results of each disturbance source type to the settlement observation, assign a weight value to each disturbance source type. The higher the sensitivity, the larger the weight value (the sum of the weight values is 1). For example, groundwater level changes have a higher sensitivity to settlement observation and are given a larger weight, while atmospheric pressure fluctuations have a lower sensitivity and are given a smaller weight. Multiply the intensity quantitative value of each single disturbance source (converted to a corresponding numerical value, such as grade 1 corresponding to 1, grade 2 corresponding to 2, etc.) by its respective weight value and add them up to obtain the comprehensive influence index at that observation time. The comprehensive influence index is a continuous numerical value that reflects the total intensity of the combined action of multiple disturbance sources.

[0078] Step S1348: Combine the single interference source quantification record, the classification and integration result, and the comprehensive influence index to form an environmental interference source intensity dataset containing the influence range and action strength of each interference source at the observation time.

[0079] The single interference source quantification record of step S1345, the classification and integration result (record list arranged by interference source type and time sequence) of step S1346, and the comprehensive influence index calculated in step S1347 are combined to construct an environmental interference source intensity dataset. The dataset is stored in a multi-level structure, with the first level being the observation date, the second level being the observation time, and the third level being the interference source type. Each interference source type lists the intensity quantification value, influence range, and other single interference source quantification record information. At the same time, the comprehensive influence index is recorded at the observation time level. The dataset can be stored as a JSON format file or multiple associated tables in a relational database, facilitating subsequent data query and analysis.

[0080] Step S135: Track the processing timestamp sequence in the data processing node information to determine all correction operations experienced by the time-series subsidence observation data during processing, extract the correction algorithm name, correction parameter adjustment value, and data change amount before and after correction for each correction operation, and form a data correction operation record.

[0081] The processing timestamp sequence is extracted from the data processing node information, and each data processing node is arranged in timestamp order. The processing algorithm function of each data processing node is analyzed to identify nodes involved in data correction operations, such as abnormal value correction nodes, temperature drift correction nodes, system error correction nodes, etc. For each data correction operation node, its processing log is retrieved to extract detailed information about the correction operation, including the correction algorithm name, correction parameter adjustment value, data value before correction, data value after correction, and data change amount before and after correction. The above information is integrated in chronological order of the correction operation to form a data correction operation record.

[0082] Step S136: According to the chronological order of the observation period, integrate the effective calibration parameters, interval duration, environmental interference source intensity dataset, and corresponding data correction operation record of each period to form a traceability element combination for each observation period.

[0083] For each monitoring point, the effective calibration parameters and interval duration extracted in step S132, the environmental interference source intensity dataset formed in step S134, and the data correction operation record formed in step S135 are associated and integrated in chronological order of the observation period. The integration method is to create a traceability element combination object for each observation period, which contains the observation period identifier, effective calibration parameter set, calibration interval duration, environmental interference source intensity data subset, and data correction operation record subset.

[0084] Step S137: Analyze the differences between the traceability element combinations of adjacent observation periods, calculate the change rate of the equipment calibration parameters, the fluctuation amplitude of the environmental interference source intensity, and the frequency change of the data correction operation, and generate the dynamic change characteristics of the traceability elements.

[0085] For two consecutive observation periods, the traceability element combinations of the two are extracted, and the change rate of the equipment calibration parameters is calculated. The change rate of the equipment calibration parameters is calculated separately according to the calibration items, and the change rate is (the effective calibration parameter value of the current period minus the effective calibration parameter value of the previous period) divided by the effective calibration parameter value of the previous period. The fluctuation amplitude of the environmental interference source intensity is calculated, and for each interference source type in the environmental interference source intensity data subset, the action intensity level value is extracted, and the fluctuation amplitude is the absolute value of the difference between the current period action intensity level value and the previous period action intensity level value. The number of correction operations in the data correction operation record subset is counted, and the difference between the current period correction operation number and the previous period correction operation number is calculated as the frequency change of the data correction operation. The change rate of the equipment calibration parameters, the fluctuation amplitude of the environmental interference source intensity, and the frequency change of the data correction operation are arranged in order of observation period to generate the dynamic change characteristics of the traceability elements.

[0086] Step S138: Associate the traceability element combination of each observation period with the dynamic change characteristics of the traceability elements to form the settlement monitoring data traceability link characteristics in time sequence, and the settlement monitoring data traceability link characteristics include a static element set and a dynamic change parameter.

[0087] The traceability element combination of each observation period is taken as the static element set, and the static element set includes the effective calibration parameters, the calibration interval length, the environmental interference source intensity data, and the data correction operation records of the observation period, which do not change with time. The dynamic change characteristics of the traceability elements are taken as the dynamic change parameter, which reflects the changes between adjacent observation periods. The static element set and the dynamic change parameter are associated through the observation period identifier to form the settlement monitoring data traceability link characteristics.

[0088] Step S140: Coupling analysis of the settlement data traceability link characteristics and the corresponding settlement amount is performed to determine the traceability reliability index of the settlement data of each monitoring point, and the traceability reliability index is used to represent the degree of interference of the settlement data in the generated link.

[0089] In the subway along the line settlement monitoring scene, in order to screen high-quality settlement data for trend analysis, it is necessary to perform coupling analysis of the settlement data traceability link characteristics and the actual settlement amount, evaluate the interference degree of the data, and determine the traceability reliability index.

[0090] Step S141: Extract the effective calibration parameters of each observation period from the sediment data provenance link feature, compare the effective calibration parameters with the device nominal accuracy range, calculate the calibration deviation rate, and assign a positive calibration weight when the calibration deviation rate is within the preset deviation range; assign a negative calibration weight when the calibration deviation rate exceeds the preset deviation range.

[0091] From the static element set of the sediment data provenance link feature, the effective calibration parameters of each observation period are extracted, including the deviation correction values of each calibration item. The device nominal accuracy range is obtained from the technical specification of the observation device, and each calibration item corresponds to a nominal allowable deviation range. The calibration deviation rate is calculated, which is the ratio of the absolute value of the deviation correction value of the effective calibration parameter to the upper limit value of the device nominal accuracy range. The preset deviation range is set according to the importance of the device and the monitoring accuracy requirement. When the calibration deviation rate is less than or equal to the upper limit value of the preset deviation range, it is determined that the calibration state is good, and a positive calibration weight is assigned; when the calibration deviation rate is greater than the upper limit value of the preset deviation range, it is determined that the calibration state is abnormal, and a negative calibration weight is assigned.

[0092] Step S142: Extract the environmental interference source intensity dataset of each observation period, analyze the correlation between each interference source intensity and the sedimentation amount, calculate the interference correlation degree, and assign a first interference influence weight when the interference correlation degree is greater than a first preset correlation threshold; assign a second interference influence weight when the interference correlation degree is less than or equal to the first preset correlation threshold and greater than a second preset correlation threshold; assign a third interference influence weight when the interference correlation degree is less than or equal to the second preset correlation threshold, wherein the first interference influence weight value is greater than the second interference influence weight value, and the second interference influence weight value is greater than the third interference influence weight value.

[0093] From the static element set of the sediment data provenance link feature, the environmental interference source intensity dataset of each observation period is extracted, which contains the action intensity level values of each interference source type. At the same time, the corresponding sedimentation amount observation values are extracted. The correlation between each interference source intensity and the sedimentation amount is analyzed, and the interference correlation degree is calculated using statistical methods, with a value range of -1 to 1, and the greater the absolute value, the stronger the correlation. The first preset correlation threshold and the second preset correlation threshold are set according to the experience of the influence of the interference source on the sediment observation, and the first preset correlation threshold is greater than the second preset correlation threshold. When the interference correlation degree is greater than the first preset correlation threshold, a first interference influence weight is assigned; when the interference correlation degree is less than or equal to the first preset correlation threshold and greater than the second preset correlation threshold, a second interference influence weight is assigned; when the interference correlation degree is less than or equal to the second preset correlation threshold, a third interference influence weight is assigned.

[0094] Step S143: Extract the data correction operation record of each observation period, count the number of correction operations and the cumulative amount of correction parameter adjustment value, and assign a positive correction weight when both the number of correction operations and the cumulative amount of correction parameter adjustment value are within the preset standard correction range; assign a negative correction weight when the number of correction operations or the cumulative amount of correction parameter adjustment value exceeds the preset standard correction range.

[0095] The data correction operation record of each observation period is extracted from the static element set of the settlement data traceability link feature, the number of data correction operations in the observation period is counted, and the sum of the correction parameter adjustment values of each correction operation is counted. The preset standard correction range is set according to historical correction data and data processing specifications, including the normal range of the number of correction operations and the normal range of the cumulative amount of correction parameter adjustment value. When the number of correction operations in a certain observation period is within the preset number range and the cumulative amount of correction parameter adjustment value is within the preset cumulative amount range, a positive correction weight is assigned; when the number of correction operations exceeds the preset number range or the cumulative amount of correction parameter adjustment value exceeds the preset cumulative amount range, a negative correction weight is assigned.

[0096] Step S144: The positive calibration weight, the negative calibration weight, the first interference influence weight, the second interference influence weight, the third interference influence weight, the positive correction weight and the negative correction weight are weighted and summed to obtain the initial reliability score of each observation period.

[0097] The calibration weight determined in step S141, the interference influence weight determined in step S142, and the correction weight determined in step S143 are weighted and summed. The weighting coefficients of the weights are set according to the influence degree of each factor on the data reliability, and the factors with greater influence degree are assigned greater weighting coefficients. After multiplying each weight by the corresponding weighting coefficient, the initial reliability score is obtained.

[0098] Step S145: In combination with the dynamic change feature of the traceability element in the settlement data traceability link feature, the change gradient of the initial reliability score of the adjacent period is calculated, the initial reliability score is smoothed when the change gradient is less than the preset gradient threshold, and the reliability mutation point is marked and the mutation reason is recorded when the change gradient is greater than or equal to the preset gradient threshold.

[0099] Step S1451: The device calibration parameter change rate, the environmental interference source intensity fluctuation amplitude and the data correction operation frequency change of the adjacent observation period are obtained from the dynamic change feature of the traceability element.

[0100] The traceability element dynamic change characteristic data structure stored in the deposition data traceability link feature is arranged in order according to observation periods, and the dynamic change characteristic of each period includes an equipment calibration parameter change rate array, an environmental interference source intensity fluctuation amplitude array and a data correction operation frequency change value. For two adjacent observation periods (such as period t and period t-1), the equipment calibration parameter change rate (change rate of each calibration item), the environmental interference source intensity fluctuation amplitude (fluctuation amplitude of each interference source type) and the data correction operation frequency change value (difference value between the correction times of the current period and the correction times of the previous period) of period t are extracted, which reflect the dynamic change of the key elements affecting the data reliability in the adjacent periods.

[0101] Step S1452: Calculate the difference value of the initial reliability score of the adjacent observation periods, and perform ratio operation on the difference value and the time interval of the adjacent periods to obtain the change gradient of the initial reliability score.

[0102] From the initial reliability score sequence obtained in step S144, the initial reliability scores of two adjacent observation periods (period t and period t-1) are taken out, and the difference value (initial reliability score t minus initial reliability score t-1) of the two is calculated. The time interval of the two adjacent observation periods is obtained, which is the number of days between the observation date of period t and the observation date of period t-1 (if the observation period is once a day, the time interval is usually 1 day, and if there is data missing, the actual interval is calculated). The difference value of the initial reliability score is divided by the time interval to obtain the change gradient of the initial reliability score, which is in units of "score / day" and represents the average change amount of the reliability score per day.

[0103] Step S1453: Compare the change gradient with the preset gradient threshold value. If the change gradient is less than the preset gradient threshold value, perform smoothing processing on the initial reliability score; if the change gradient is greater than or equal to the preset gradient threshold value, identify that there is a reliability mutation in the observation period.

[0104] The preset gradient threshold value is determined by analyzing historical reliability score change data and is set as an empirical value, which reflects the maximum change rate allowed by the reliability score under normal circumstances. The change gradient calculated in step S1452 is compared with the preset gradient threshold value in value. If the absolute value of the change gradient is less than the absolute value of the preset gradient threshold value, it means that the reliability score fluctuates within the normal range and does not need special processing, and the smoothing processing is directly performed; if the absolute value of the change gradient is greater than or equal to the absolute value of the preset gradient threshold value, it means that the reliability score has an abnormal rapid change, and it is identified that there is a reliability mutation phenomenon in the current observation period (period t).

[0105] Step S1454: When performing smoothing processing, the initial reliability score is processed using a sliding average method, and the size of the sliding window is adaptively determined according to the total number of observation periods by a preset proportion.

[0106] When it is determined that smoothing processing is needed, the sliding average method is selected as the smoothing algorithm. The size of the sliding window is adaptively determined according to the total number N of observation periods by a preset proportion (such as 5%). The window size is N multiplied by the preset proportion, and the result is rounded to an integer (if the result is less than 1, it is taken as 1). For the current observation period t, the sliding window contains periods t and several periods before and after (for example, when the window size is 5, it contains periods t-2, t-1, t, t+1, and t+2. If it is the first or last period of the sequence, the boundary is filled, and the reliability score of the missing period is filled by copying the first or last value). Calculate the arithmetic mean of the initial reliability scores of all periods in the window, and replace the original initial reliability score with the average value as the smoothed reliability score.

[0107] Step S1455: When a reliability mutation is identified, the timestamp of the mutation observation period and the corresponding initial reliability score are extracted.

[0108] For the observation period t identified as having a reliability mutation, the timestamp of the period (accurate to the observation date and time) is extracted from the metadata of the time series settlement observation data, and the initial reliability score (the original score without smoothing processing) corresponding to the period is recorded. The timestamp and initial reliability score are stored in temporary variables for subsequent mutation cause analysis and recording.

[0109] Step S1456: Based on the dynamic change characteristics of the traceability elements before and after the mutation observation period, determine the main factor causing the reliability mutation: if the device calibration parameter change rate exceeds the preset change threshold, the main factor is device calibration change; if the intensity fluctuation amplitude of the environmental disturbance source exceeds the preset fluctuation threshold, the main factor is environmental disturbance change; if the data correction operation frequency changes to zero and the fluctuation amplitudes of the remaining factors are all below the preset fluctuation threshold, the main factor is data processing anomaly.

[0110] The dynamic change characteristics of the traceable elements in several periods (such as the previous 3 periods and the next 3 periods) before and after the mutation observation period t are called, and the change rate of the equipment calibration parameters, the fluctuation amplitude of the intensity of the environmental interference source and the operation frequency change of the data correction in period t are analyzed. The preset change threshold (for the change rate of the equipment calibration parameters) and the preset fluctuation threshold (for the fluctuation amplitude of the intensity of the environmental interference source) are determined according to 3 times the standard deviation of the normal fluctuation range in the historical data. If the absolute value of the change rate of any calibration item in the change rate of the equipment calibration parameters exceeds the preset change threshold, it is determined that the main factor is the change of the equipment calibration (such as re-calibration of the equipment or adjustment of the calibration parameters); if the fluctuation amplitude of any interference source type in the fluctuation amplitude of the intensity of the environmental interference source exceeds the preset fluctuation threshold, it is determined that the main factor is the change of the environmental interference (such as sudden strong vibration or sudden change of underground water level); if the frequency change value of the data correction operation is zero (i.e. the number of corrections is the same as that of the previous period), and the change rate of the equipment calibration parameters and the fluctuation amplitude of the intensity of the environmental interference source do not exceed the respective thresholds, it is determined that the main factor is the abnormality of data processing (such as change of the processing algorithm not recorded, or operation error of the processing personnel).

[0111] Step S1457: The time stamp of the mutation observation period, the initial reliability score and the determined main factor are associated and stored to generate mutation point analysis data.

[0112] A mutation point analysis data record is created, including fields: mutation period timestamp, initial reliability score, main factor, detailed description (briefly describe the specific element change that leads to the mutation, such as "equipment sensitivity calibration parameter change rate exceeds threshold by 2 times"). The timestamp and initial reliability score extracted in step S1455 and the main factor and detailed description determined in step S1456 are filled into the record, and then the record is appended to the mutation point analysis data set, which is stored in table form, supporting subsequent query and statistical analysis.

[0113] Step S1458: The reliability scores after smoothing and the reliability scores after identifying mutations are standardized to have a unified data structure.

[0114] The reliability scores after smoothing (for non-mutation periods) and the reliability scores after identifying mutations (for mutation periods, retaining the original scores and adding mutation markers) are uniformly converted to the same data format. The data structure includes: observation period ID, processed reliability score, whether it is a mutation (Boolean value), mutation reason (if it is a mutation period, fill in the main factor, otherwise, it is empty). The standardized data of all observation periods are arranged in chronological order and stored in a reliability score processing result array for subsequent geological adaptability adjustment steps.

[0115] Step S146: According to the stratum bearing capacity distribution in the geological foundation information, the initial reliability score is adjusted in geological adaptability. When the stratum bearing capacity value is greater than the first bearing capacity threshold value in the region, and the settlement data reliability score is less than the first reliability threshold value, the initial reliability score is increased by a preset proportion; when the stratum bearing capacity value is less than the second bearing capacity threshold value in the region, and the settlement data reliability score is greater than the second reliability threshold value, the initial reliability score is reduced by a preset proportion.

[0116] The stratum bearing capacity value of the position of each monitoring point is obtained from the geological foundation information, and the stratum bearing capacity value is the bearing capacity characteristic value converted from the standard penetration test hammering number or the static sounding cone tip resistance of the region. The first bearing capacity threshold value and the second bearing capacity threshold value are set according to the geological conditions, and the first bearing capacity threshold value is greater than the second bearing capacity threshold value. The first reliability threshold value and the second reliability threshold value are set according to the general reliability level of the settlement data, and the first reliability threshold value is greater than the second reliability threshold value. When the stratum bearing capacity value is greater than the first bearing capacity threshold value, if the initial reliability score of the monitoring point in the region is less than the first reliability threshold value, the initial reliability score is increased by a preset proportion; when the stratum bearing capacity value is less than the second bearing capacity threshold value, if the initial reliability score of the monitoring point in the region is greater than the second reliability threshold value, the initial reliability score is reduced by a preset proportion.

[0117] Step S147: The reliability score after smoothing processing, mutation point marking and geological adaptability adjustment is normalized to obtain the traceability reliability index of the settlement data of each monitoring point with uniform value range, and the traceability reliability index is used to represent the degree of interference of the settlement data in the generation link.

[0118] The reliability scores of all observation periods after smoothing processing, mutation point marking and geological adaptability adjustment are collected to find the maximum value and the minimum value. Linear normalization method is adopted to subtract the minimum value from each reliability score and divide by (the maximum value minus the minimum value) to obtain the traceability reliability index with a value range of 0 to 1. The closer the index value is to 1, the smaller the degree of interference of the settlement data in the generation link, and the more reliable the data; the closer the index value is to 0, the greater the degree of interference, and the less reliable the data.

[0119] Step S150: According to the traceability reliability index, the settlement data meeting the traceability reliability index threshold requirement is screened out, and the dynamic settlement trend model of the settlement monitoring area is generated combined with the spatial distribution characteristics of the monitoring points, and the settlement monitoring analysis report containing the traceability link characteristics of the settlement data is generated.

[0120] In the subway along the line settlement monitoring scene, the high reliability settlement data after screening is used to construct a dynamic settlement trend model combined with the spatial position relationship of the monitoring points, to intuitively display the regional settlement change, and generate an analysis report containing traceability information.

[0121] Step S151: Filter out the settlement data in each monitoring point with the traceability reliability index higher than the threshold value of the traceability reliability index in all observation periods as the settlement data meeting the threshold value of the traceability reliability index.

[0122] The threshold value of the traceability reliability index is set, and the value of the threshold value is determined according to the accuracy requirement and data quality target of the settlement monitoring. The settlement data of all observation periods of each monitoring point are traversed, and the data records with the traceability reliability index higher than the threshold value are filtered out to form a high-reliability settlement data set. The high-reliability settlement data set is subjected to effectiveness check, including time continuity and spatial distribution uniformity of the data.

[0123] Step S152: Extract the location coordinates of each monitoring point and the settlement amount of the corresponding observation period in the settlement data meeting the threshold value of the traceability reliability index, and establish a three-dimensional data set.

[0124] The monitoring point number, observation date and settlement amount of each data record are extracted from the high-reliability settlement data set, and the corresponding plane coordinates and elevation coordinates are obtained from the location coding information according to the monitoring point number. The location coordinates, settlement amount and observation date are combined to establish a three-dimensional data set, and each record of the three-dimensional data set contains X coordinate, Y coordinate, Z coordinate, observation date, settlement amount and other fields. The three-dimensional data set is stored in a spatial database to support spatial query and spatial analysis operations.

[0125] Step S153: Based on the three-dimensional data set, the spatial distribution characteristics of the monitoring points are analyzed, the settlement monitoring area is divided into multiple spatial grid units according to the location coordinates, the average settlement amount of all monitoring points in the same observation period in each grid unit is calculated, and a grid unit settlement matrix is generated.

[0126] The spatial clustering analysis is performed on the location coordinates of the monitoring points in the three-dimensional data set, and the density clustering algorithm is used to identify the dense and sparse areas of the spatial distribution of the monitoring points. According to the range of the monitoring area and the density of the monitoring points, the settlement monitoring area is divided into multiple square spatial grid units of the same size, and the side length of the grid unit is determined according to the distance between the monitoring points. For each spatial grid unit, all the monitoring points falling into the unit are counted. For each observation period, the arithmetic mean of the settlement amounts of all the monitoring points in the grid unit is calculated as the average settlement amount of the grid unit in the observation period. The average settlement amounts of all the grid units in each observation period are arranged in the order of grid unit number and observation period to form a grid unit settlement matrix.

[0127] Step S154: Time series analysis is performed on the grid cell settlement matrix to extract the settlement change trend of each grid cell in the continuous observation period, and the grid cells with settlement growth rate greater than the preset growth rate threshold and the grid cells with settlement growth rate less than or equal to the preset stable threshold are determined.

[0128] The average settlement time series of each grid cell in the grid cell settlement matrix is analyzed for trend, and the rate of change of settlement with time (i.e., settlement growth rate) is calculated using linear regression method. The preset growth rate threshold and the preset stable threshold are set according to the settlement control standard, and the preset growth rate threshold is greater than the preset stable threshold. When the settlement growth rate of a certain grid cell is greater than the preset growth rate threshold, it is marked as a settlement active cell; when the settlement growth rate is less than or equal to the preset stable threshold, it is marked as a settlement stable cell; and between the two, it is a settlement slow cell.

[0129] Step S155: Based on the settlement change trend of each grid cell, a spatial interpolation model is constructed to predict the settlement of the area without monitoring points, and settlement spatial distribution data covering the entire monitoring area is generated.

[0130] Step S1551: Determine the type of spatial interpolation model as a Kriging interpolation model based on the constraint of geological basement information, which is configured to modify the interpolation results in combination with the stratum bearing capacity distribution.

[0131] According to the complexity of the geological conditions of the settlement monitoring area, a Kriging interpolation model based on the constraint of geological basement information is selected as the type of spatial interpolation model. The Kriging interpolation model introduces the stratum bearing capacity distribution in the geological basement information as an auxiliary variable to constrain and modify the interpolation process based on the traditional Kriging interpolation. The core configurations of the model include: variogram model type (such as spherical model), geological constraint factor (correlation coefficient of stratum bearing capacity and settlement), and interpolation search radius (dynamically adjusted according to the density of monitoring points). By inputting the stratum bearing capacity distribution grid map as a covariate into the model, the model not only considers the spatial distance factor when predicting the settlement of unknown points, but also considers the stratum bearing capacity characteristics of the point, thereby improving the interpolation accuracy.

[0132] Step S1552: The settlement of the known monitoring points in the grid cell settlement matrix is input as input data, together with the geological information code corresponding to each known monitoring point as a covariate, and input into the spatial interpolation model.

[0133] Extract the settlement data of all grid cells containing monitoring points (grid cells with known monitoring points) from the grid cell settlement matrix, which are the main input data of the spatial interpolation model (i.e. the attribute values of known sample points). At the same time, obtain the geological information code corresponding to each known monitoring point from the monitoring point geological information table, which contains information such as stratum bearing capacity level and rock-soil compressibility level, and convert the geological information code into a numerical covariate (e.g. map different levels to different integer codes). Combine the planar coordinates of known sample points (grid cell center point coordinates), settlement (attribute value) and geological information code covariate into an input data set in the format of a two-dimensional array, with each row representing a sample point and the columns representing X coordinate, Y coordinate, settlement, stratum bearing capacity level code, compressibility level code, etc.

[0134] Step S1553: Configure the semi-variogram parameters of the spatial interpolation model, where the type of semi-variogram is determined according to the settlement spatial correlation characteristics of the monitoring area.

[0135] Analyze the spatial correlation characteristics of the settlement of the monitoring area through experimental semi-variogram. First, calculate the distance and the square of the settlement difference between all pairs of known sample points, then group them by distance interval, and calculate the semi-variogram value (half of the average of the square of the difference) for each distance interval. Draw the experimental semi-variogram graph, and according to the trend of the semi-variogram value with distance, select the appropriate theoretical semi-variogram type. If the semi-variogram value increases first and then stabilizes with increasing distance, select the spherical model; if it shows an exponential growth trend, select the exponential model; if it is a Gaussian curve shape, select the Gaussian model. After determining the semi-variogram type, estimate the model parameters: range (spatial autocorrelation distance), nugget value (asymptotic value of semi-variogram), sill value (semi-variogram value at the origin) by least squares fitting of experimental data, which describe the spatial variation structure of settlement.

[0136] Step S1554: Run the spatial interpolation model to predict the settlement of grid cells without monitoring points, and obtain the predicted settlement and corresponding prediction error of each unmonitored grid cell.

[0137] The input data set prepared in step S1552 is input into the configured Kriging interpolation model based on the constraint of geological base information. The model first calculates the spatial weight matrix between sample points (based on the semi-variation function parameters), and then modifies the weight by using the geological covariates (sample points in areas with high stratum bearing capacity are given higher weight). For each grid cell center point without a monitoring point, the model searches for known sample points within a certain range according to its coordinate position, and calculates the predicted settlement of the center point by weighted average. At the same time, the model outputs the prediction error (such as Kriging variance) corresponding to each predicted settlement, which reflects the uncertainty of the predicted value. The smaller the error, the more reliable the prediction. The predicted settlement and the prediction error are stored according to the grid cell number to form a prediction result table.

[0138] Step S1555: The predicted settlement is fused with the settlement of the known monitoring points to generate preliminary settlement spatial distribution data covering the entire monitoring area.

[0139] The grid cell settlement of the known monitoring points (from the grid cell settlement matrix) is integrated with the predicted settlement of the unmonitored grid cells in step S1554 to ensure that each grid cell has a corresponding settlement value (the known points use the measured average settlement, and the unknown points use the predicted settlement). These settlement values are arranged in order of the spatial coordinates of the grid cells to form a two-dimensional array, with the row and column of the array corresponding to the row number and column number of the grid cells, and the array element value being the settlement. The two-dimensional array is converted into a raster data format, with the pixel size of the raster being consistent with the size of the grid cells, and the pixel value being the settlement, thereby generating preliminary settlement spatial distribution data covering the entire monitoring area.

[0140] Step S1556: Based on the rock-soil compressibility parameter distribution map in the geological base information, the preliminary settlement spatial distribution data is modified: for the grid cells whose rock-soil compressibility level belongs to the preset high compressibility level interval, the predicted settlement is increased by a first preset adjustment coefficient; for the grid cells whose rock-soil compressibility level belongs to the preset low compressibility level interval, the predicted settlement is decreased by a second preset adjustment coefficient.

[0141] The compressibility parameter distribution map of the rock-soil mass is called from the geological basement information, which divides the monitoring area into different compressibility grade partitions. The preset high compressibility grade interval is the grade with a compression modulus less than a certain value (such as a compression modulus < 4 MPa), and the preset low compressibility grade interval is the grade with a compression modulus greater than a certain value (such as a compression modulus > 15 MPa). According to the positive correlation between the compressibility of the rock-soil mass and the settlement, a first preset adjustment coefficient (greater than 1, such as 1.1) and a second preset adjustment coefficient (less than 1, such as 0.9) are set. Each grid cell in the preliminary settlement spatial distribution data is traversed, and the compressibility grade of the rock-soil mass corresponding to the grid cell is queried through spatial superposition. If it belongs to the high compressibility grade interval, the settlement thereof is multiplied by the first preset adjustment coefficient; if it belongs to the low compressibility grade interval, the settlement thereof is multiplied by the second preset adjustment coefficient; the settlement of the grid cell in the medium compressibility grade interval remains unchanged.

[0142] Step S1557: According to the prediction error, the confidence level of the modified settlement spatial distribution data is evaluated, the area with a prediction error less than a first preset error threshold is divided into a high confidence area, the area with a prediction error between the first preset error threshold and a second preset error threshold is divided into a medium confidence area, and the area with a prediction error greater than the second preset error threshold is divided into a low confidence area.

[0143] The prediction error of each grid cell is extracted from the prediction result table of step S1554, and a first preset error threshold and a second preset error threshold (the first preset error threshold is less than the second preset error threshold) are set. These thresholds are determined according to the statistical distribution of all prediction errors (such as the first threshold being the error mean minus the standard deviation, and the second threshold being the error mean plus the standard deviation). For each grid cell in the modified settlement spatial distribution data, the prediction error thereof is compared with the preset threshold: if the prediction error is less than the first preset error threshold, the area where the grid cell is located is a high confidence area; if the prediction error is greater than or equal to the first preset error threshold and less than or equal to the second preset error threshold, it is a medium confidence area; if the prediction error is greater than the second preset error threshold, it is a low confidence area. In space, the grid cells with the same confidence level are connected into a piece to form a high, medium, and low confidence area polygon.

[0144] Step S1558: The confidence level evaluation result is integrated with the modified settlement data to generate the final settlement spatial distribution data covering the entire monitoring area, which includes the settlement value of each grid cell and the corresponding confidence level.

[0145] A final deposition amount spatial distribution data structure is created, which includes two layers: a deposition amount grid layer and a confidence level grid layer. The deposition amount grid layer stores the deposition amount values of each grid cell after the correction in step S1556; the confidence level grid layer stores the confidence levels of each grid cell divided in step S1557 (represented by numbers 1, 2, and 3 for high, medium, and low confidence levels, respectively). The two layers are spatially registered in the geographic information system to ensure one-to-one correspondence of the grid cells. The final generated deposition amount spatial distribution data is stored in the GeoTIFF format, which supports opening and analysis in professional GIS software, and the attributes of each grid cell include both the deposition amount value and the confidence level.

[0146] Step S156: Time series superposition of deposition amount spatial distribution data of different observation periods is performed to analyze the expansion direction of the subsidence area and the position change of the shrinkage area, and to extract the moving track of the subsidence boundary.

[0147] The deposition amount spatial distribution data of each observation period is displayed in time sequence, and different colors are used to represent different deposition amounts to form a subsidence dynamic change map. By comparing the deposition amount spatial distribution of adjacent observation periods, the area with increased deposition amount and the area with decreased deposition amount are identified. For the expansion area, the moving direction and distance of the geometric center are calculated; for the shrinkage area, the moving direction and distance of the geometric center are also calculated. The extraction of the subsidence boundary uses an edge detection algorithm to perform edge detection on the deposition amount spatial distribution data of each observation period to obtain the boundary polygon of the subsidence area. The boundary polygons of different periods are superimposed, the moving track of the boundary polygon vertices is analyzed, and the continuous moving track line of the subsidence boundary is obtained through curve fitting.

[0148] Step S157: A dynamic subsidence trend model is constructed by combining the moving track of the subsidence boundary and the deposition rate of the grid cells, and the dynamic subsidence trend model includes the deposition amount spatial distribution of each time node, the moving speed of the subsidence boundary, and the migration path of the subsidence core area.

[0149] The deposition amount spatial distribution data, the moving track of the subsidence boundary, and the deposition rate of the grid cells are integrated into the dynamic subsidence trend model. The model displays the deposition amount spatial distribution of each time node in the form of time series animation, and the size change of the deposition amount is intuitively reflected through color gradient. The moving speed of the subsidence boundary is calculated, and the moving speed is the ratio of the boundary moving distance to the time interval. The subsidence core area is defined as a continuous set of grid cells with the maximum deposition rate and an area exceeding a preset threshold, and the migration path of the subsidence core area is generated by tracking the position change of the geometric center of the core area. The dynamic subsidence trend model supports interactive operation, and users can select any observation period through the time slider to view the deposition distribution of the period.

[0150] Step S158: Verify the dynamic subsidence trend model, compare the subsidence amount of the subsequent period predicted by the dynamic subsidence trend model with the actual observed subsidence data that meets the traceability reliability index threshold requirement, adjust the spatial interpolation parameters of the dynamic subsidence trend model, and make the deviation between the output of the dynamic subsidence trend model and the actual data within the preset deviation range.

[0151] The actual subsidence data of the recent several observation periods is selected as the verification sample, the subsidence amount of these periods predicted by the dynamic subsidence trend model is compared with the actual observed value, and the prediction deviation of each grid cell is calculated. The average value and the root mean square error of the absolute value of the prediction deviation of all grid cells are calculated to evaluate the prediction accuracy of the model. If the average absolute error or the root mean square error exceeds the preset deviation range, the variogram parameters or the interpolation method of the spatial interpolation model are adjusted, and the interpolation prediction is performed again. Repeat the adjustment and verification process until the model prediction deviation is within the preset deviation range.

[0152] Step S159: Generate a subsidence monitoring analysis report containing subsidence data traceability link features.

[0153] For example, step S1591: Extract key feature parameters from the dynamic subsidence trend model, including the location coordinates of the subsidence core area, the moving speed of the subsidence boundary, and the subsidence growth rate of each grid cell.

[0154] The key feature parameters calculated by the dynamic subsidence trend model are queried and extracted through the data interface of the model. The location coordinates of the subsidence core area are obtained by calculating the geometric center of the set of continuous grid cells with the largest subsidence growth rate, and the geometric center coordinates are the weighted average of the center point coordinates of all grid cells in the region (the weight is the subsidence growth rate of each grid cell); the moving speed of the subsidence boundary is calculated by the distance difference between the corresponding vertices of the subsidence boundary polygon in adjacent periods divided by the time interval, and the average of the moving speeds of all vertices is taken as the overall boundary moving speed; the subsidence growth rate of each grid cell is directly extracted from the grid cell attribute table, which is generated in step S154 and records the linear regression subsidence growth rate value of each grid cell. The above key feature parameters are arranged into time series data according to the observation period, so as to analyze the variation law thereof with time.

[0155] Step S1592: Determine significant traceability elements from the subsidence data traceability link features based on a preset influence significance threshold, the significant traceability elements including observation periods with device calibration deviation rate exceeding the preset deviation range, time periods with environmental interference source intensity reaching the preset influence level, and processing nodes with data correction operation exceeding the preset standard correction range.

[0156] A significant influence threshold is set, including a preset upper limit of a device calibration deviation rate, a preset influence level of an environmental interference source intensity (such as a "strong" level and above), and a preset standard correction range upper limit of a data correction operation (an upper limit of a correction number and a correction parameter cumulative amount). The characteristics of the deposition data traceability link of each monitoring point are traversed. For each observation period, it is checked whether the device calibration deviation rate exceeds the preset deviation range. If it exceeds, the observation period is marked as a device calibration significant element. It is checked whether there is a disturbance source type and time period that reaches the preset influence level in the environmental interference source intensity data set. If there is, the time period is marked as an environmental interference significant element. It is checked whether there is a processing node with a correction number or correction parameter adjustment value cumulative amount exceeding the preset standard correction range in the data correction operation record. If there is, the processing node is marked as a data correction significant element. All marked significant traceability elements are classified by type, and the time stamp, associated monitoring point, and specific exceeding value are recorded.

[0157] Step S1593: The key feature parameters are associated with the significant traceability elements, and the traceability influence factors corresponding to each deposition feature are identified. When the deposition growth rate of the deposition core area is greater than a preset growth rate threshold, the device calibration record and the environmental interference source intensity change in the same period are associated and analyzed.

[0158] An association analysis mechanism between the key feature parameters and the significant traceability elements is established. A time window matching method is used. For each abnormal change point of a key feature parameter (such as a sudden increase in deposition growth rate), it is searched within a certain range (such as ±3 observation periods) before and after the time stamp whether there is a significant traceability element. If the deposition growth rate of the deposition core area is greater than a preset growth rate threshold (which is the same as the preset growth rate threshold in step S154), the device calibration record (obtained from the device calibration record sequence) and the environmental interference source intensity change data (obtained from the dynamic change characteristics of the environmental interference source intensity data set) in the same period (the same observation period) are extracted, and the time correlation between the change trend of the device calibration deviation rate, the fluctuation of the environmental interference source intensity, and the deposition growth rate anomaly is analyzed. For example, if there is a record of a device calibration deviation rate exceeding the standard in the period before the sudden increase in deposition growth rate, it is judged that the device calibration may be one of the influence factors causing the deposition growth rate anomaly. If the environmental interference source intensity reaches the preset influence level in the same period, it is judged that the environmental interference may also be an influence factor.

[0159] Step S1594: The traceability reliability indicators of each monitoring point are counted, the proportion of data meeting the traceability reliability indicator threshold requirement in the total data amount is calculated, and the data distribution characteristics of different spatial areas meeting the threshold requirement are analyzed to generate a data reliability evaluation result.

[0160] For each monitoring point, the total number of traceability reliability indicators and the number of data records higher than the traceability reliability indicator threshold value in all observation periods are counted, and the ratio of the two (proportion) is calculated to obtain the high reliability data proportion of the monitoring point. The high reliability data proportions of all monitoring points are marked on the monitoring area map according to the spatial position, and different proportion ranges (such as 0-20%, 20-40%, …, 80-100%) are represented using a hierarchical color setting method. The spatial distribution characteristics of the proportion (such as which areas have high proportions and which areas have low proportions) are analyzed. At the same time, the average high reliability data proportion of the entire monitoring area is calculated, and the number and proportion of monitoring points with a proportion higher than the average value are counted. The above statistical results, spatial distribution characteristics description, and proportion analysis of typical areas are integrated to generate a data reliability evaluation result report, which includes statistical tables and spatial distribution maps.

[0161] Step S1595: In combination with the geological basement information, the correlation between stratum bearing capacity distribution and settlement trend is evaluated, the analysis results of the influence of geological conditions on settlement development are generated, and the basis for distinguishing geological factors and external interference factors is generated based on the traceability information.

[0162] A stratum bearing capacity distribution grid is extracted from the geological basement information, which is spatially superimposed and analyzed with the settlement amount spatial distribution data in the dynamic settlement trend model. Using a partition statistical method, the average settlement amount and average settlement increase rate in different stratum bearing capacity level regions are calculated. By comparing and analyzing the correlation between the two (such as drawing a scatter plot of bearing capacity level and average settlement increase rate, and calculating the correlation coefficient), the influence degree of stratum bearing capacity distribution on settlement trend (such as whether the settlement increase rate is higher in low bearing capacity regions) is evaluated. Based on the environmental disturbance source intensity and equipment calibration records in the characteristics of the settlement data traceability link, the geological factors and external interference factors are distinguished: when the stratum bearing capacity level is low and there is no significant external disturbance source (the environmental disturbance source intensity does not reach the preset level, and the equipment calibration is normal), the settlement is mainly caused by geological factors (soil compressibility); when there is a significant external disturbance source and the settlement abnormal change is highly consistent with the occurrence time of the disturbance source, the settlement is mainly caused by external interference factors. The correlation evaluation results, influence degree analysis, and distinguishing basis are integrated to form a geological condition influence analysis report.

[0163] Step S1596: According to the preset report template, the visualization data of the dynamic settlement trend model, the analysis results of significant traceability elements, the data reliability evaluation results, and the geological correlation analysis results are integrated.

[0164] The preset subsidence monitoring analysis report template is called, and the subsidence monitoring analysis report template includes chapters such as cover, table of contents, preface, monitoring area profile, data source and processing, dynamic subsidence trend analysis, data reliability evaluation, geological condition influence analysis, conclusion and suggestion, and the like. The visualization data of the dynamic subsidence trend model, such as the subsidence amount spatial distribution time sequence diagram, the subsidence boundary movement track animation, and the core area migration path diagram, are inserted into the chapter of dynamic subsidence trend analysis; the significant traceability elements determined in step S1592 and the influence analysis of the significant traceability elements on subsidence are inserted into the traceability analysis part of the chapter of data source and processing; the data reliability evaluation result generated in step S1594 is inserted into the chapter of data reliability evaluation; and the geological correlation analysis result in step S1595 is inserted into the chapter of geological condition influence analysis. It is ensured that the contents of each chapter are logically coherent, the figure numbers are standardized, and the data references are accurate.

[0165] Step S1597: In the integration process, each analysis result is associated with the subsidence data traceability link feature on which it is based, and finally a subsidence monitoring analysis report containing the subsidence data traceability link feature is generated.

[0166] In the report integration process, for each analysis conclusion or data chart, the source of the subsidence data traceability link feature on which it is based is explicitly marked in the footnotes or explanatory text thereof. For example, when describing the abnormal subsidence increase rate of a certain area, it is marked that “this conclusion is based on the environmental disturbance source intensity data set (atmospheric pressure fluctuation level reaches ‘intense’) of the observation period t and the equipment calibration record (calibration deviation rate exceeds the preset range by 1.5 times)”; when displaying the high-reliability data proportion chart, it is marked that “the data is derived from the traceability reliability index calculation results of each monitoring point, and the calculation results of the processing algorithm version V2.3 in the data processing node information table are referred to for details”. Through the above association identification, the user of the report can trace back to the original data and traceability information of each analysis result, ensuring the scientificity and verifiability of the report. After completing all content integration and association identification, the report is formatted, checked for typographical errors, and logically audited, and finally a complete subsidence monitoring analysis report containing the subsidence data traceability link feature is generated.

[0167] Based on the same inventive concept, please refer to Figure 2 , which shows the structure schematic block diagram of the data traceability based subsidence monitoring system 100 provided by the embodiments of the present application for executing the above-mentioned data traceability based subsidence monitoring method. The data traceability based subsidence monitoring system 100 can include a communication unit 110, a machine readable storage medium 120, and a processor 130.

[0168] The machine readable storage medium 120 is configured to store machine executable instructions for implementing the scheme of the present application, and the processor 130 is configured to execute the machine executable instructions stored in the machine readable storage medium 120 to implement the data provenance based settlement monitoring method provided by the foregoing method embodiment.

[0169] It should be noted that, in order to simplify the expression of the present disclosure and help to understand one or more embodiments of the present application, in the foregoing description of embodiments of the present application, various features are sometimes combined into one embodiment, figure or description thereof.

Claims

1. A data provenance based settlement monitoring method, characterized in that, The method comprises: Obtaining time-series subsidence observation data of each monitoring point in a subsidence monitoring area, wherein the time-series subsidence observation data comprises the subsidence amount of each monitoring point in a continuous observation period and environmental disturbance information corresponding to the observation time; Building a subsidence data provenance correlation network, correlating the time-series subsidence observation data with geological basement information of the monitoring points, observation equipment provenance identification, and data processing node information, and forming subsidence data provenance correlation network information containing data sources and correlation elements; Based on the subsidence data provenance correlation network information, tracing the generation link of the subsidence data of each monitoring point, extracting the equipment calibration records, environmental disturbance source intensity, and data correction operation records at the time of data collection in different observation periods, and generating subsidence data provenance link features; Coupling analysis of the subsidence data provenance link features and the corresponding subsidence amount is performed to determine the provenance reliability index of the subsidence data of each monitoring point, which is used to represent the degree of disturbance of the subsidence data in the generation link; According to the provenance reliability index, subsidence data meeting the threshold requirements of the provenance reliability index is screened out, and a dynamic subsidence trend model of the subsidence monitoring area is generated in combination with the spatial distribution characteristics of the monitoring points, and a subsidence monitoring analysis report containing subsidence data provenance link features is generated.

2. The data provenance based settlement monitoring method of claim 1, wherein, The building of the subsidence data provenance correlation network, the correlation of the time-series subsidence observation data with the geological basement information of the monitoring points, the observation equipment provenance identification, and the data processing node information, and the formation of the subsidence data provenance correlation network information containing data sources and correlation elements, comprises: Analyzing the metadata field of the time-series subsidence observation data, extracting the location coding information and observation period identification of each monitoring point, and establishing an index system of the time-series subsidence observation data; Collecting the geological basement information of each monitoring point, wherein the geological basement information comprises stratum bearing capacity distribution, rock-soil compression parameters, and geological structure fracture zone distribution, and assigning a unique geological information code to each monitoring point; Obtaining the provenance identification of the observation equipment used by each monitoring point, wherein the provenance identification comprises equipment production batch number, historical calibration certificate number, and equipment operation and maintenance record number, and establishing a full life cycle provenance archive of the observation equipment; Collecting all data processing node information involved in the processing of the time-series subsidence observation data, wherein the data processing node information comprises node processing algorithm version, processing personnel qualification number, and processing timestamp, and forming a data processing link list; Establishing the correlation between the time-series subsidence observation data and the geological basement information, and binding the time-series subsidence observation data of each monitoring point to the corresponding stratum bearing capacity distribution, rock-soil compression parameters, and geological structure fracture zone distribution through the mapping of the monitoring point location coding information and the geological information code; Establishing the correlation between the time-series subsidence observation data and the observation equipment provenance identification, and binding the subsidence data of each observation period to the corresponding equipment production batch number and historical calibration certificate number through the matching of the observation period identification and the equipment operation and maintenance record number; Establish the association between the time series settlement observation data and the data processing node information, associate the original observation data with the first processing node algorithm version, personnel qualification number and timestamp in the order of data processing, and then associate the previous processing node information with the next processing node information; Take the monitoring point as the core node, take the time series settlement observation data, geological basement information, observation equipment traceability identification and data processing node information as the association edge, construct a multi-dimensional settlement data traceability association network, and each association edge in the settlement data traceability association network carries a corresponding association strength parameter, forming settlement data traceability association network information containing data sources and associated elements.

3. The data provenance based settlement monitoring method of claim 2, wherein, The geological basement information of each monitoring point is collected, and the geological basement information includes stratum bearing capacity distribution, rock-soil compression parameter and geological structure fracture zone distribution, and each monitoring point is assigned a unique geological information code, including: Obtain the geological survey report and drilling data of the settlement monitoring area, and extract the physical and mechanical property parameters of each soil layer from the data, including the unit weight, internal friction angle and cohesion of the soil; Based on the physical and mechanical property parameters, the characteristic value of the stratum bearing capacity of each drilling point is calculated, and the spatial interpolation method is used to extend the discrete drilling point bearing capacity characteristic value to the entire monitoring area to generate a stratum bearing capacity distribution grid map; The compression modulus and compression coefficient of the rock-soil body are extracted from the drilling data, the rock-soil body at different depths is divided into a preset number of compression grades according to the compression modulus value, and the monitoring area is divided into multiple compression subareas according to the compression grade to form a rock-soil compression parameter distribution map; Analyze the geological structure description in the geological survey report, identify the location, trend and tendency of the fracture zone in the monitoring area in combination with the seismic exploration data, determine the influence width and activity characteristics of the fracture zone through the fracture zone activity frequency and influence range, and generate a geological structure fracture zone distribution map; The stratum bearing capacity distribution grid map, rock-soil compression parameter distribution map and geological structure fracture zone distribution map are spatially superimposed to form a comprehensive geological basement information map; The comprehensive geological basement information map is grid processed according to the preset grid accuracy, and each grid cell corresponds to a unique spatial coordinate range; Each grid cell is assigned a geological information code, and the geological information code rule includes the stratum bearing capacity level code, rock-soil compression level code and fracture zone influence degree code of the area; The location coordinates of each monitoring point are matched with the spatial coordinate range of the grid cell, so that each monitoring point corresponds to a unique grid cell, and thus the geological information code and corresponding stratum bearing capacity distribution, rock-soil compression parameter and geological structure fracture zone distribution of the monitoring point are obtained.

4. The data provenance based settlement monitoring method of claim 1, wherein, The generation link of each monitoring point settlement data is traced based on the settlement data traceability association network information, the equipment calibration record, environmental interference source intensity and data correction operation record during data collection in different observation periods are extracted, the settlement data traceability link characteristics are generated, including: Locate the observation equipment provenance identifier of each monitoring point from the provenance correlation network information of the subsidence data, query the initial calibration parameters of the equipment at the time of factory delivery according to the equipment production batch number, and then call the deviation correction value and calibration environmental conditions of each calibration through the serial number of the calibration certificate to form the equipment calibration record sequence; For each observation period, extract the calibration deviation correction value closest to the observation time from the equipment calibration record sequence, mark it as the effective calibration parameter of the observation period, and record the interval length between the calibration time and the observation time; Analyze the environmental interference information in the provenance correlation network information of the subsidence data, combine the geological structure fracture zone distribution in the geological basement information, identify the type of environmental interference source that may affect the subsidence data in each observation period, and the environmental interference source type includes surrounding vibration source, underground water level change and atmospheric pressure fluctuation; Call the quantitative evaluation results of the intensity of the environmental interference source through the processing algorithm version in the data processing node information, and the quantitative evaluation results include the influence range and action intensity of each interference source at the observation time, and form the environmental interference source intensity dataset; Track the processing timestamp sequence in the data processing node information, determine all the correction operations experienced by the time series subsidence observation data in the processing process, extract the correction algorithm name, correction parameter adjustment value and data change before and after correction of each correction operation, and form the data correction operation record; According to the order of the observation period, integrate the effective calibration parameter, interval length, environmental interference source intensity dataset and corresponding data correction operation record of each period to form the provenance element combination of each observation period; Analyze the differences between the provenance element combinations of adjacent observation periods, calculate the change rate of the equipment calibration parameter, the fluctuation amplitude of the environmental interference source intensity and the frequency change of the data correction operation, and generate the dynamic change characteristics of the provenance elements; Correlate the provenance element combination of each observation period with the dynamic change characteristics of the provenance elements to form the subsidence data provenance link characteristics in time sequence, and the subsidence data provenance link characteristics include a static element set and dynamic change parameters.

5. The data provenance based settlement monitoring method of claim 4, wherein, The quantitative evaluation results of the intensity of the environmental interference source are called through the processing algorithm version in the data processing node information, and the quantitative evaluation results include the influence range and action intensity of each interference source at the observation time, and form the environmental interference source intensity dataset, including: Analyze the processing algorithm version number in the data processing node information, determine the algorithm model type for evaluating the intensity of the environmental interference source, and the algorithm model type includes statistical regression model, physical field simulation model and machine learning prediction model; According to the algorithm model type, call the corresponding model input parameters, and the model input parameters include vibration frequency monitoring data, underground water level monitoring data and atmospheric pressure monitoring data at the observation time; Run the algorithm model to calculate the model input parameters, output the intensity quantitative value of each environmental interference source, and the intensity quantitative value is represented by a preset number of relative grades, and the grade division is based on the influence degree of the interference source on the subsidence observation; The influence range of each interference source is calculated based on the intensity quantization value, the influence range of the vibration source is calculated by a vibration wave propagation attenuation model, and the model parameters include medium characteristics, amplitude and frequency; the influence range of the groundwater level change is calculated by a groundwater seepage model, and the model parameters include the permeability coefficient, the hydraulic slope and the aquifer thickness; the influence range of the atmospheric pressure fluctuation is calculated according to the topographic features of the monitoring area, and the topographic features include flat areas, hilly areas and mountainous areas; The intensity quantization value, the influence range and the corresponding observation time of each environmental interference source are associated to form a single interference source quantization record; The single interference source quantization records of all environmental interference sources are integrated, classified according to the type of interference source, and the records of the same type of interference source are arranged in chronological order; The comprehensive influence index of different types of interference sources at the same observation time is calculated, and the comprehensive influence index is the weighted sum of the intensity quantization values of each single interference source, and the weight value is determined according to the sensitivity of the interference source to the settlement observation; The single interference source quantization record, the classified and integrated result and the comprehensive influence index are combined to form an environmental interference source intensity data set containing the influence range and the action intensity of each interference source at the observation time.

6. The data provenance based settlement monitoring method of claim 4, wherein, The coupling analysis of the settlement data traceability link features and the corresponding settlement amount is performed to determine the traceability reliability index of the settlement data of each monitoring point, including: The effective calibration parameters of each observation period are extracted from the settlement data traceability link features, the effective calibration parameters are compared with the device nominal accuracy range, the calibration deviation rate is calculated, when the calibration deviation rate is within the preset deviation range, a positive calibration weight is given; when the calibration deviation rate exceeds the preset deviation range, a negative calibration weight is given; The environmental interference source intensity data set of each observation period is extracted, the correlation between the intensity of each interference source and the settlement amount is analyzed, and the interference correlation degree is calculated, when the interference correlation degree is greater than a first preset correlation threshold, a first interference influence weight is given; when the interference correlation degree is less than or equal to the first preset correlation threshold and greater than a second preset correlation threshold, a second interference influence weight is given; when the interference correlation degree is less than or equal to the second preset correlation threshold, a third interference influence weight is given, wherein the first interference influence weight value is greater than the second interference influence weight value, and the second interference influence weight value is greater than the third interference influence weight value; The data correction operation record of each observation period is extracted, the number of correction operations and the cumulative amount of correction parameter adjustment value are counted, when the number of correction operations and the cumulative amount of correction parameter adjustment value are within the preset standard correction range, a positive correction weight is given; when the number of correction operations or the cumulative amount of correction parameter adjustment value exceeds the preset standard correction range, a negative correction weight is given; The positive calibration weight, the negative calibration weight, the first interference influence weight, the second interference influence weight, the third interference influence weight, the positive correction weight and the negative correction weight are weighted and summed to obtain the initial reliability score of each observation period; The change gradient of the initial reliability score of the adjacent period is calculated according to the dynamic change feature of the traceability element in the traceability link feature of the settlement data, when the change gradient is less than a preset gradient threshold, the initial reliability score is smoothed, and when the change gradient is greater than or equal to the preset gradient threshold, a reliability mutation point is marked and a mutation reason is recorded; According to the stratum bearing capacity distribution in the geological basement information, the initial reliability score is adjusted in geological adaptability, when the stratum bearing capacity value is greater than a first bearing capacity threshold, and the settlement data reliability score is less than a first reliability threshold, the initial reliability score is increased by a preset proportion, and when the stratum bearing capacity value is less than a second bearing capacity threshold, and the settlement data reliability score is greater than a second reliability threshold, the initial reliability score is decreased by a preset proportion; The reliability score after the smoothing, the mutation point marking and the geological adaptability adjustment is normalized to obtain the traceability reliability index of the settlement data of each monitoring point with a unified value range, and the traceability reliability index is used to represent the degree of interference of the settlement data in the generation link.

7. The data provenance based settlement monitoring method of claim 6, wherein, The change gradient of the initial reliability score of the adjacent period is calculated according to the dynamic change feature of the traceability element in the traceability link feature of the settlement data, when the change gradient is less than a preset gradient threshold, the initial reliability score is smoothed, and when the change gradient is greater than or equal to the preset gradient threshold, a reliability mutation point is marked and a mutation reason is recorded; The dynamic change feature of the traceability element includes the device calibration parameter change rate, the environmental interference source intensity fluctuation amplitude and the data correction operation frequency change of the adjacent observation period; The difference value of the initial reliability score of the adjacent observation period is calculated, and the difference value is subjected to ratio operation with the time interval of the adjacent period to obtain the change gradient of the initial reliability score; The change gradient is compared with the preset gradient threshold, if the change gradient is less than the preset gradient threshold, the initial reliability score is smoothed, if the change gradient is greater than or equal to the preset gradient threshold, the observation period is identified as having a reliability mutation; When the smoothing is performed, the initial reliability score is processed by using the sliding average method, and the size of the sliding window is adaptively determined according to the total number of observation periods by a preset proportion; When the reliability mutation is identified, the timestamp and the corresponding initial reliability score of the mutation observation period are extracted; Based on the dynamic change feature of the traceability element before and after the mutation observation period, the main factor causing the reliability mutation is determined, if the device calibration parameter change rate exceeds a preset change threshold, the main factor is device calibration change, if the environmental interference source intensity fluctuation amplitude exceeds a preset fluctuation threshold, the main factor is environmental interference change, and if the data correction operation frequency change is zero and the fluctuation amplitudes of the remaining factors are all less than the preset fluctuation threshold, the main factor is data processing abnormality; The timestamp, the initial reliability score and the determined main factor of the mutation observation period are associated and stored to generate mutation point analysis data; The reliability score after the smoothing and the reliability score after the identification of the mutation are subjected to data format standardization processing, so that the reliability scores of all observation periods have a unified data structure. ​ 8. The data provenance based settlement monitoring method of claim 1, wherein, The settlement data meeting the traceability reliability index threshold requirement is screened out according to the traceability reliability index, and a dynamic settlement trend model of the settlement monitoring area is generated in combination with the spatial distribution characteristics of the monitoring points, including: Settlement data with a traceability reliability index higher than the traceability reliability index threshold in all observation periods of each monitoring point is screened out as settlement data meeting the traceability reliability index threshold requirement; The position coordinates of each monitoring point and the settlement amount of the corresponding observation period in the settlement data meeting the traceability reliability index threshold requirement are extracted, and a three-dimensional data set is established; Based on the three-dimensional data set, the spatial distribution characteristics of the monitoring points are analyzed, the settlement monitoring area is divided into multiple spatial grid units according to the position coordinates, the average settlement amount of all monitoring points in the same observation period in each grid unit is calculated, and a grid unit settlement matrix is generated; The grid unit settlement matrix is subjected to time series analysis, the settlement amount variation trend of each grid unit in consecutive observation periods is extracted, and the grid units with a settlement increase rate greater than a preset increase rate threshold and the grid units with a settlement increase rate less than or equal to a preset stability threshold are determined; Based on the settlement amount variation trend of each grid unit, a spatial interpolation model is constructed, the settlement amount of the area without monitoring points is predicted, and settlement amount spatial distribution data covering the entire monitoring area are generated; The settlement amount spatial distribution data of different observation periods are subjected to time series superposition, the expansion direction of the settlement area and the position change of the shrinkage area are analyzed, and the movement trajectory of the settlement boundary is extracted; In combination with the movement trajectory of the settlement boundary and the settlement increase rate of the grid unit, a dynamic settlement trend model is constructed, and the dynamic settlement trend model includes the settlement amount spatial distribution, the settlement boundary movement speed and the migration path of the settlement core area of each time node; The dynamic settlement trend model is verified, the subsequent period settlement amount predicted by the dynamic settlement trend model is compared with the settlement data meeting the traceability reliability index threshold requirement actually observed, the spatial interpolation parameters of the dynamic settlement trend model are adjusted, and the deviation between the output of the dynamic settlement trend model and the actual data is within a preset deviation range.

9. The data provenance based settlement monitoring method of claim 8, wherein, Based on the settlement amount variation trend of each grid unit, a spatial interpolation model is constructed, the settlement amount of the area without monitoring points is predicted, and settlement amount spatial distribution data covering the entire monitoring area are generated, including: The type of the spatial interpolation model is determined as a Kriging interpolation model constrained based on geological basement information, and the Kriging interpolation model is configured to correct the interpolation result in combination with the stratum bearing capacity distribution; The settlement amount of the known monitoring points in the grid unit settlement matrix is input as input data, together with the geological information code corresponding to each known monitoring point as a covariant, and is input into the spatial interpolation model; The semi-variogram function parameters of the spatial interpolation model are configured, and the type of the semi-variogram function is determined according to the settlement spatial correlation characteristics of the monitoring area; The spatial interpolation model is run, the settlement amount of the grid unit without monitoring points is predicted, and the predicted settlement amount and the corresponding prediction error of each unmonitored grid unit are obtained; fuse the predicted settlement amount with the settlement amount of the known monitoring points to generate preliminary settlement amount spatial distribution data covering all grid cells in the entire monitoring area; based on the rock-soil compressibility parameter distribution map in the geological basement information, correct the preliminary settlement amount spatial distribution data: for the grid cells whose rock-soil compressibility level belongs to the preset high compressibility level interval, increase the predicted settlement amount by a first preset adjustment coefficient; for the grid cells whose rock-soil compressibility level belongs to the preset low compressibility level interval, decrease the predicted settlement amount by a second preset adjustment coefficient; according to the prediction error, perform confidence assessment on the corrected settlement amount spatial distribution data, divide the area where the prediction error is less than a first preset error threshold into a high-confidence area, the area where the prediction error is between the first preset error threshold and a second preset error threshold into a medium-confidence area, and the area where the prediction error is greater than the second preset error threshold into a low-confidence area; integrate the confidence assessment result with the corrected settlement amount data to generate final settlement amount spatial distribution data covering the entire monitoring area, which contains the settlement amount value of each grid cell and its corresponding confidence level.

10. A settlement monitoring system based on data provenance, characterized in that, comprise: a processor; a machine-readable storage medium for storing machine-executable instructions of the processor; wherein the processor is configured to perform the data provenance-based settlement monitoring method of any one of claims 1 to 8 via execution of the machine-executable instructions.

Images