Layered monitoring method for land subsidence based on force sensor

Abstract

The invention relates to the technical field of engineering geological survey, in particular to a land subsidence layered monitoring method based on a force sensor. A force sensor is arranged on the optimal monitoring deployment site of each settlement monitoring layer to obtain a pressure stress sensing reading, and contour surface interpolation of an anti-settlement distance is carried out on a multi-dimensional settlement cell constructed by analyzing the pressure stress sensing reading through settlement, so that a settlement vector applied by pressure stress between the upper and lower settlement monitoring layers is obtained; the settlement vector is deduced through inversion to obtain the relative settlement displacement amplitude of each settlement monitoring layer, the relative settlement displacement amplitude is linearly associated to execute trend reconstruction and quantitative projection embedding of relative displacement between the settlement monitoring layers, and the ground settlement monitoring quantity of the target layered monitoring area is obtained. According to the invention, the force sensor can be utilized to carry out settlement layering, settlement vector monitoring analysis and settlement trend quantitative calculation on the target area, so that monitoring results of different levels of settlement of the ground can be accurately output.

Description

Technical Field

[0001] The present invention relates to the technical field of engineering geological exploration, and particularly relates to a method for layered monitoring of ground settlement based on force sensors. Background Art

[0002] Ground settlement is a phenomenon in which the surface height gradually decreases due to reasons such as excessive exploitation of water resources in the stratum, soil layer compression, or underground engineering activities, and it widely exists in fields such as urban construction, transportation, mineral development, and water conservancy projects. To ensure the safety of engineering structures and the stability of the geological environment, it is of great significance to monitor ground settlement in a timely and accurate manner. Currently, common ground settlement monitoring technologies include leveling measurement, GNSS monitoring, InSAR technology, laser ranging method, etc. These technologies are mostly used for macroscopic monitoring and are suitable for overall settlement analysis of large areas. However, at the microscopic level, especially in the layered settlement monitoring of the soil layer inside, it is impossible to accurately obtain the settlement amounts of soil layers at different depths, resulting in the inability to accurately identify the key layers and their deformation characteristics during the settlement process; and it is difficult for traditional methods to infer the influence of the settlement vector and trend of the upper soil layer on the lower soil layer when the upper soil layer settles through the application of stress; at the same time, the calculation of the settlement degree requires manual operation, which may have a large manual intervention error, resulting in inaccurate monitoring results. Summary of the Invention

[0003] The present invention overcomes the deficiencies of the prior art and provides a method for layered monitoring of ground settlement based on force sensors.

[0004] To achieve the above object, the technical solution adopted by the present invention is as follows: The first aspect of the present invention provides a method for layered monitoring of ground settlement based on force sensors, including the following steps: Construct a feature transition inference network to estimate several historical spatio-temporal rock layer sample maps in the target layered monitoring area, generate a directed edge map of rock layer settlement, and use the forward and backward message passing of the directed edge map to infer the edge probability of rock layer feature transition, so as to determine the settlement monitoring levels of the target layered monitoring area; Extract the ground rock layer structure and historical compression coefficient of each settlement monitoring level in the target layered monitoring area, avoid areas prone to relative settlement based on the ground rock layer structure and historical compression coefficient, make a reasonable layout of sites, obtain the optimal monitoring deployment sites and add them to the force sensor deployment plan; Deploy force sensors at the optimal monitoring deployment sites of each settlement monitoring level to obtain compressive stress sensing readings, and perform isosurface interpolation of the anti-settlement distance by analyzing the multi-dimensional settlement cells constructed by the compressive stress sensing readings, so as to obtain the settlement vector of the compressive stress applied between the upper and lower settlement monitoring levels; The settlement vector is inverted and inferred to obtain the relative settlement displacement amplitude of each settlement monitoring layer. The relative settlement displacement amplitude is linearly correlated to perform trend reconstruction and quantitative projection embedding of the relative displacement between each settlement monitoring layer to obtain the ground settlement monitoring amount in the target layered monitoring area.

[0005] More specifically, the construction of the feature change inference network estimates several historical spatiotemporal rock layer sample maps in the target stratified monitoring area, generates a directed edge map of rock layer settlement, and uses the directed edge map to forward and reversely transfer messages to infer the edge probability of rock layer feature changes to determine the settlement monitoring level of the target stratified monitoring area, specifically including the following steps: Obtain the target hierarchical monitoring area and monitoring tasks for ground subsidence, and define the termination monitoring point of ground subsidence according to the monitoring tasks; By monitoring the central control log, several historical spatiotemporal rock layer sample maps of the target layered monitoring area where the ground subsided to the termination monitoring point within the preset time period were extracted, and the local binary pattern algorithm was introduced to calculate the characteristics of each historical spatiotemporal rock layer sample map to obtain the LBP rock layer characteristic value of each historical spatiotemporal rock layer sample map; Construct a characteristic change inference network, and estimate the conditional probability of the rock formation characteristics of a certain historical space-time rock formation sample map transferring to another historical space-time rock formation sample map as the sedimentation time series is followed based on the LBP rock formation characteristic values ​​in the characteristic change inference network, and obtain the joint probability of rock formation characteristic changes between each historical space-time rock formation sample map; Define the historical spatiotemporal rock layer sample graph as a random variable node, construct the factor node of each random variable node according to the joint probability of rock layer characteristic change, and connect each random variable node based on the structure of factor nodes to build a directed edge graph of rock layer settlement; The stratum lithology characteristics and settlement elements of the target layered monitoring area are obtained through big data, and based on the stratum structural characteristics and settlement elements, messages are forwardly transmitted from the random variable node to the factor node along the stratum characteristic change edge in the directed edge graph. After the forward transmission is completed, messages are reversely transmitted from the factor node to the random variable node along the stratum characteristic change edge; Repeat the above steps to continuously update all messages of each historical spatiotemporal rock layer sample graph by forward and reverse transmission of messages along the edge in the directed edge graph until the maximum iteration frequency is reached, and generate multiple forward edge probability reasoning messages and multiple reverse edge probability reasoning messages; The edge restoration is calculated by combining a plurality of the forward edge probability inference messages with a plurality of reverse edge probability inference messages to obtain the edge probability of the rock formation characteristic change, and the settlement monitoring level of the target stratified monitoring area is determined based on the edge probability of the rock formation characteristic change.

[0006] More specifically, for each settlement monitoring level in the extraction target hierarchical monitoring area, the ground rock formation and historical compression coefficient are obtained. Based on the ground rock formation and historical compression coefficient, reasonable layout sites are avoided in the areas prone to relative settlement, and the optimal monitoring deployment sites are obtained and added to the force sensor deployment plan, which specifically includes the following steps: Obtain the corresponding model basic information of the force sensors to be used for ground settlement monitoring in the target hierarchical monitoring area, and retrieve the monitoring specification parameters of the force sensors based on the model basic information in the big data network; Based on big data, obtain the settlement cases of the target hierarchical monitoring area, and extract the ground rock formation and historical compression coefficient of each settlement monitoring level in the target hierarchical monitoring area through the settlement cases; Establish a deployment sampling area for each settlement monitoring level according to the ground rock formation, divide each deployment sampling area into several sub-sampling areas according to the monitoring specification parameters, and independently sample the layout sites of the force sensors on each sub-sampling area of each deployment sampling area to obtain N layout site samples of the force sensors randomly provided on each sub-sampling area; Calculate the difference in interlayer compression characteristics between each sub-sampling area based on the historical compression coefficient. If the difference in interlayer compression characteristics is greater than the preset interlayer compression characteristic difference threshold, extract the sub-sampling area corresponding to the compression characteristic difference, mark it as the area prone to relative settlement, and generate the pattern of areas prone to relative settlement between each settlement monitoring level; Preset the settlement azimuth vector according to the pattern of areas prone to relative settlement, and use the N layout site samples to estimate the layout azimuth of the force sensors with the settlement azimuth vector as the avoidance reference benchmark to obtain the expected layout estimation azimuth of each settlement monitoring level; Preset the minimum deployment mean square error, obtain the simulated monitoring variance and simulated monitoring cost when the force sensors are simulated and deployed on the expected layout estimation azimuth of each layer, and calculate the deployment mean square error of each layout site sample based on the simulated monitoring variance and simulated monitoring cost; Only extract the layout site samples corresponding to the unique deployment mean square error less than the minimum deployment mean square error, calibrate them as the optimal monitoring deployment sites, and add the optimal monitoring deployment sites to the force sensor deployment plan.

[0007] More specifically, when force sensors are arranged at the optimal monitoring deployment sites of each settlement monitoring level to obtain compressive stress sensing readings, and the multi-dimensional settlement cells constructed by analyzing the compressive stress sensing readings are used for isosurface interpolation of the anti-settlement distance, the settlement vector of the compressive stress applied between the upper and lower settlement monitoring levels is obtained, which specifically includes the following steps: Sensing and monitoring the ground settlement of the target hierarchical monitoring area by arranging and installing force sensors at the optimal monitoring deployment sites of each settlement monitoring level to obtain the compressive stress sensing readings of each settlement monitoring level in the target hierarchical monitoring area; Construct a multi-dimensional compressive stress density space for the settlement monitoring level, obtain the contour threshold of the multi-dimensional compressive stress density space, assign the contour value of the compressive stress density according to the compressive stress sensing reading to construct scalar corner points, and form multi-dimensional settlement cells of the compressive stress density based on the scalar corner points; If the contour values of one or more scalar corner points on the multi-dimensional settlement cell cross below the contour threshold of the multi-dimensional compressive stress density space, it means that the multi-dimensional settlement cell is crossed by settlement, and it is marked as a multi-dimensional crossed settlement cell, and the scalar corner points passing through the isosurface on the boundary of the multi-dimensional crossed settlement cell are stripped out and defined as settlement crossing scalar corner points; Preset the reverse settlement distance weight function of the settlement change in time series for the compressive stress sensing reading, calculate the distance from each compressive stress sensing reading to the settlement monitoring level when the force sensor does not generate a reading, obtain the reverse settlement distance, and calculate the reverse settlement distance through the reverse settlement distance weight function to obtain the reverse settlement distance weight value; Use the linear intersection interpolation of the reverse settlement distance weight value to determine the settlement crossing intersection point and the compressive stress normal vector, solve and connect the isogrid vertices based on the settlement crossing intersection point and the compressive stress normal vector to determine the settlement vector of the soil layer compressive stress applied by the upper settlement monitoring level to the lower settlement monitoring level.

[0008] More specifically, the method of using the linear intersection interpolation of the reverse settlement distance weight value to determine the settlement crossing intersection point and the compressive stress normal vector, solving and connecting the isogrid vertices based on the settlement crossing intersection point and the compressive stress normal vector to determine the settlement vector and settlement trend of the soil layer compressive stress applied by the upper settlement monitoring level to the lower settlement monitoring level specifically includes the following steps: Perform linear intersection interpolation on one or more settlement crossing scalar corner points through the reverse settlement distance weight value to determine the settlement crossing intersection point, and obtain the compressive stress normal vector of the settlement crossing intersection point in the multi-dimensional compressive stress density space; Define a crossing isosurface vertex for each multi-dimensional crossed settlement cell, construct a quadratic error function based on the settlement crossing intersection point and the compressive stress normal vector, solve the quadratic error function, and generate the isogrid vertices of the multi-dimensional crossed settlement cell; If there are crossing isosurface vertices in the cells adjacent to the current multi-dimensional crossed settlement cell, connect the isogrid vertices of each other to obtain the settlement compressive stress isosurface map of the settlement monitoring level, and analyze the settlement compressive stress isosurface map to determine the settlement vector of the soil layer compressive stress applied by the upper settlement monitoring level to the lower settlement monitoring level.

[0009] More specifically, the inversion infers the settlement vector to obtain the relative settlement displacement amplitude of each settlement monitoring level, and linearly correlates the relative settlement displacement amplitude to perform trend reconstruction and quantitative projection embedding of the relative displacement between each layer of settlement monitoring levels, so as to obtain the ground settlement monitoring quantity of the target stratified monitoring area, specifically including the following steps: Obtain the prior distribution of the belief displacement of soil layer settlement in the target stratified monitoring area through the ground rock formation structure, formation lithology characteristics and settlement elements in the big data network, and construct a displacement inversion model based on the prior node pattern and parameters of the prior distribution of belief displacement; Introduce the maximum likelihood method to calculate the settlement vector between each settlement monitoring level, obtain the settlement likelihood function of each settlement monitoring level, and use the displacement inversion model to infer the posterior displacement of the settlement monitoring level under the condition of the settlement vector corresponding to the settlement likelihood function, so as to obtain the relative settlement displacement amplitude of each settlement monitoring level; Obtain the response time series step length when the force sensor of the adjacent settlement monitoring level monitors the output compressive stress sensing reading, and preset the reconstruction error threshold of the relative displacement between the adjacent settlement monitoring levels based on the response time series step length; Introduce the least squares method to linearly correlate the weights between the relative settlement displacement amplitude of each layer of settlement monitoring level and the relative settlement displacement amplitude of the adjacent settlement monitoring level until the reconstruction error of the neighbor linear correlation is less than the relative displacement reconstruction error threshold, so as to obtain the global correlation weight matrix; Construct a low-dimensional settlement space, and use the global correlation weight matrix to solve and obtain K settlement trend eigenvalues and the settlement trend eigenvectors corresponding to each settlement trend eigenvalue; Based on the K settlement trend eigenvalues, embed the settlement trend eigenvectors into the low-dimensional settlement space to perform quantitative projection expression of the relative displacement between each layer of settlement monitoring levels, and finally obtain the ground settlement monitoring quantity of the target stratified monitoring area.

[0010] The second aspect of the present invention provides a ground settlement stratified monitoring system based on a force sensor. The ground settlement stratified monitoring system includes a memory and a processor. A ground settlement stratified monitoring method program is stored in the memory. When the ground settlement stratified monitoring method program is executed by the processor, the steps of any one of the ground settlement stratified monitoring methods are realized.

[0011] The present invention solves the technical defects existing in the background technology. The beneficial technical effects of the present invention are as follows: Construct a feature change inference network to estimate several historical spatio-temporal rock layer sample maps in the target stratified monitoring area, generate a directed edge map of rock layer settlement, and use the forward and backward transmission of messages in the directed edge map to infer the edge probability of rock layer feature changes, so as to determine the settlement monitoring level of the target stratified monitoring area; extract the ground rock layer structure and historical compression coefficient of each settlement monitoring level in the target stratified monitoring area, avoid decision-making in areas prone to relative settlement based on the ground rock layer structure and historical compression coefficient, rationally arrange sites, obtain the optimal monitoring deployment sites and add them to the force sensor deployment plan; deploy force sensors at the optimal monitoring deployment sites of each settlement monitoring level to obtain compressive stress sensing readings, and perform isosurface interpolation of the anti-settlement distance by analyzing the multi-dimensional settlement cells constructed by the compressive stress sensing readings to obtain the settlement vector of the compressive stress applied between the upper and lower settlement monitoring levels; invert and infer the settlement vector to obtain the relative settlement displacement amplitude of each settlement monitoring level, and linearly associate the relative settlement displacement amplitude of the neighbors to perform trend reconstruction and quantitative projection embedding of the relative displacement between each layer of settlement monitoring levels, so as to obtain the ground settlement monitoring quantity of the target stratified monitoring area. The present invention can use force sensors to conduct settlement stratification, settlement vector monitoring analysis, and settlement trend quantitative calculation on the target area, so as to achieve a layered settlement monitoring effect, accurately output the settlement degree of different ground levels, and significantly improve the reliability and credibility of ground settlement monitoring results. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0013] Figure 1 FIG. shows a first method flow chart of a ground settlement stratified monitoring method based on a force sensor; Figure 2 FIG. shows a second method flow chart of a ground settlement stratified monitoring method based on a force sensor; Figure 3 FIG. shows a system framework diagram of a ground settlement stratified monitoring system based on a force sensor. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0014] In order to more clearly understand the above objects, features, and advantages of the present invention, the following will further describe the present invention in detail with reference to the drawings and specific embodiments. It should be noted that, without conflict, the embodiments of the present application and the features in the embodiments can be combined with each other.

[0015] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the scope of protection of the present invention is not limited by the specific embodiments disclosed below.

[0016] The first aspect of the present invention provides a method for layered monitoring of land subsidence based on a force sensor, as Figure 1 shown, which includes the following steps: Construct a feature transition inference network to estimate several historical spatio-temporal rock layer sample maps in the target layered monitoring area, generate a directed edge map of rock layer subsidence, and use the forward and backward transmission of messages in the directed edge map to infer the edge probability of rock layer feature transition, so as to determine the subsidence monitoring level of the target layered monitoring area; Extract the ground rock layer structure and historical compression coefficient of each subsidence monitoring level in the target layered monitoring area, avoid the decision-making of easily relatively subsiding areas based on the ground rock layer structure and historical compression coefficient, rationally arrange the sites, obtain the optimal monitoring deployment sites and add them to the force sensor deployment plan; Install force sensors at the optimal monitoring deployment sites of each subsidence monitoring level to obtain compressive stress sensing readings. Analyze the multi-dimensional subsidence cells constructed by the compressive stress sensing readings through subsidence traversal, and perform isosurface interpolation of the anti-subsidence distance to obtain the subsidence vector of the compressive stress applied between the upper and lower subsidence monitoring levels; Invert and infer the subsidence vector to obtain the relative subsidence displacement amplitude of each subsidence monitoring level. Linearly correlate the relative subsidence displacement amplitude of the neighbors to perform trend reconstruction and quantitative projection embedding of the relative displacement between each layer of subsidence monitoring levels, and obtain the ground subsidence monitoring quantity of the target layered monitoring area.

[0017] More specifically, the construction of the feature transition inference network to estimate several historical spatio-temporal rock layer sample maps in the target layered monitoring area, generate a directed edge map of rock layer subsidence, and use the forward and backward transmission of messages in the directed edge map to infer the edge probability of rock layer feature transition, so as to determine the subsidence monitoring level of the target layered monitoring area, specifically includes the following steps: Obtain the target layered monitoring area of land subsidence and the monitoring task, and define the termination monitoring landing point of land subsidence according to the monitoring task; Extract several historical spatio-temporal rock layer sample maps of the target layered monitoring area where land subsidence occurs to the termination monitoring landing point during a preset time period through the monitoring control log, introduce the local binary pattern algorithm to calculate the features of each historical spatio-temporal rock layer sample map, and obtain the LBP rock layer feature value of each historical spatio-temporal rock layer sample map; Construct a characteristic change inference network, and estimate the conditional probability of the rock formation characteristics of a certain historical space-time rock formation sample map transferring to another historical space-time rock formation sample map as the sedimentation time series is followed based on the LBP rock formation characteristic values ​​in the characteristic change inference network, and obtain the joint probability of rock formation characteristic changes between each historical space-time rock formation sample map; Define the historical spatiotemporal rock layer sample graph as a random variable node, construct the factor node of each random variable node according to the joint probability of rock layer characteristic change, and connect each random variable node based on the structure of factor nodes to build a directed edge graph of rock layer settlement; The stratum lithology characteristics and settlement elements of the target layered monitoring area are obtained through big data, and based on the stratum structural characteristics and settlement elements, messages are forwardly transmitted from the random variable node to the factor node along the stratum characteristic change edge in the directed edge graph. After the forward transmission is completed, messages are reversely transmitted from the factor node to the random variable node along the stratum characteristic change edge; Repeat the above steps to continuously update all messages of each historical spatiotemporal rock layer sample graph by forward and reverse transmission of messages along the edge in the directed edge graph until the maximum iteration frequency is reached, and generate multiple forward edge probability reasoning messages and multiple reverse edge probability reasoning messages; The edge restoration is calculated by combining a plurality of the forward edge probability inference messages with a plurality of reverse edge probability inference messages to obtain the edge probability of the rock formation characteristic change, and the settlement monitoring level of the target stratified monitoring area is determined based on the edge probability of the rock formation characteristic change.

[0018] It should be noted that traditional ground settlement monitoring usually uses tools such as inclinometers, settlement markers, or liquid settlement gauges. However, ground settlement is often caused by the different compressibilities of different soil layers, changes in the groundwater level, construction disturbances, etc. These traditional monitoring tools are difficult to achieve layered monitoring of rock and soil, thus unable to accurately judge the depth and cause of settlement and timely detect the development of abnormal settlement, resulting in settlement warning errors and missing the opportunity for intervention. Therefore, accurate layered monitoring of rock and soil is crucial. In this regard, this method first extracts the characteristics of historical spatio-temporal rock layer samples in the target layered monitoring area to obtain the LBP rock layer characteristic values of each sample. The LBP rock layer characteristic values depict the category discrimination labels of different geological rock layers contained in the target layered monitoring area and can provide a basis for defining geological monitoring layers with different rock layer distributions. Since the historical spatio-temporal rock layer sample map is a series of historical data, the calculated LBP rock layer characteristic values show a characteristic numerical value of historical changes between different categories of rock layers. Therefore, the joint probability of rock layer characteristic changes between each historical spatio-temporal rock layer sample map can be further inferred using the LBP rock layer characteristic values in the characteristic change inference network, thereby revealing the law of settlement characteristic changes between different rock and soil layers. Then, factor nodes constructed based on the joint probability of rock layer characteristic changes are used to connect each historical spatio-temporal rock layer sample map, enabling the directed factor distribution at the edges of each rock and soil layer in the target layered monitoring area to be clearly defined, clearly representing the dependence relationship between spatio-temporal rock layer samples under historical settlement conditions, and providing a reliable structural basis for the calculation of settlement message transmission at the edges of rock and soil layers.

[0019] It should be noted that the settlement elements include Quaternary loose sediments, groundwater layer distribution and artificial fill. Since the lithological characteristics and settlement elements of different rock and soil layers directly or indirectly lead to their settlement, they are the key information transmission media for the edge change of rock and soil layers. Therefore, this method is based on the formation structural characteristics and settlement elements. The forward and reverse message transmission from the random variable node to the factor node along the rock layer characteristic change edge in the directed edge graph is used to inform the neighboring factor node of its own edge change cognition probability, while the reverse message transmission is to synthesize the messages of all neighboring factor nodes according to the internal joint probability function structure to inform a random variable node of the edge change cognition probability that it should tend to, so that each random variable node gradually approaches the edge change probability under the historical settlement conditions through the message exchange of the factor node, thereby describing the random evolution law of the settlement edge between each rock and soil layer. Finally, multiple forward edge probability inference messages and multiple reverse edge probability inference messages are combined to restore and calculate the edge settlement of each rock and soil layer, so as to infer the reasonable settlement monitoring level that is easy and excellent for monitoring settlement behavior in the target layered monitoring area. This method can be used to infer the historical characteristic change data of different rock and soil layer samples in the target area to perform reasonable monitoring stratification, thereby significantly improving the accuracy of judging the depth and cause of settlement, effectively optimizing the settlement assessment performance of the ground foundation structure, and making ground settlement monitoring more accurate and reliable.

[0020] More specifically, the method of extracting the ground rock structure and historical compression coefficient of each settlement monitoring layer in the target layered monitoring area, avoiding the area prone to relative settlement based on the ground rock structure and historical compression coefficient, and reasonably arranging the sites, obtaining the optimal monitoring deployment site and adding it to the force sensor deployment plan, specifically includes the following steps: Obtain the corresponding model basic information of the force sensor to be used for ground subsidence monitoring in the target layered monitoring area, and retrieve the monitoring specification parameters of the force sensor in the big data network based on the model basic information; Based on big data, the settlement cases of the target layered monitoring area are obtained, and the ground rock structure and historical compression coefficient of each settlement monitoring layer in the target layered monitoring area are extracted through the settlement cases; Establish a deployment sampling area for each settlement monitoring level according to the ground rock structure, divide each deployment sampling area into several sub-sampling areas according to monitoring specification parameters, and independently sample the deployment sites of force sensors in each sub-sampling area of ​​each deployment sampling area to obtain N deployment site samples of force sensors randomly provided in each sub-sampling area; Calculate the difference in interlayer compression characteristics between each subsampling area based on the historical compression coefficient. If the difference in interlayer compression characteristics is greater than the preset interlayer compression characteristic difference threshold, extract the subsampling area corresponding to the compression characteristic difference, mark it as the area prone to relative settlement, and generate the pattern of areas prone to relative settlement between each settlement monitoring level; Preset the settlement azimuth vector according to the pattern of areas prone to relative settlement, and use N layout site samples to estimate the layout azimuth of the force sensor with the settlement azimuth vector as the avoidance reference benchmark, so as to obtain the expected layout estimation azimuth of each settlement monitoring level; Preset the minimum deployment mean square error, obtain the simulated monitoring variance and simulated monitoring cost when the force sensor is simulated and deployed on the expected layout estimation azimuth of each layer, and calculate the deployment mean square error of each layout site sample based on the simulated monitoring variance and simulated monitoring cost; Only extract the layout site samples corresponding to the unique deployment mean square error less than the minimum deployment mean square error, mark them as the optimal monitoring deployment sites, and add the optimal monitoring deployment sites to the force sensor deployment plan.

[0021] It should be noted that compared with traditional monitoring tools, the force sensor can continuously collect minute stress, strain, and relative displacement changes in real time to obtain more comprehensive structural behavior information; at the same time, it can be connected to a data acquisition system (DAS) and a cloud platform to achieve unattended automated monitoring and reduce the large maintenance costs of traditional equipment. However, if the force sensors at each settlement monitoring level are randomly installed and laid out, it may lead to local or global linear and non-linear deviations in the force sensor monitoring data, resulting in large errors in the ground settlement calculation results based on these force sensor data, seriously affecting geological disaster assessment and urban construction project planning. Therefore, it is particularly crucial whether the force sensors at each settlement monitoring level are accurately laid out reasonably. In response to this, this method independently samples layout sites on the deployment sampling area of each settlement monitoring level to obtain layout site samples. Since there is a certain range of compression coefficients with differences between different settlement monitoring levels under different actual monitoring and governance requirements, some local areas between adjacent settlement monitoring levels may have settlement monitoring deviations restricted by interlayer compression characteristics. For this, this method calculates the difference in interlayer compression characteristics between each subsampling area. If the difference in interlayer compression characteristics is greater than the preset interlayer compression characteristic difference threshold, it indicates that a certain local position on this settlement monitoring level is more likely to be affected by the compression of interlayer relative displacement, and the interlayer compression characteristics of the corresponding subsampling area of this local position will seriously interfere with the monitoring performance of the force sensor. Therefore, these areas prone to relative settlement need to be significantly marked to clarify the pattern of areas prone to relative settlement that are extremely likely to reduce the monitoring performance of the force sensor between all settlement monitoring levels, and further provide a hierarchical local chain basis for excluding unreasonable layout and installation positions.

[0022] It should be noted that when generating the pattern of the easy relative settlement field, the unreasonable points where the force sensor needs to avoid layout and installation in each layer can be determined. Therefore, according to the pattern of the easy relative settlement field, the settlement azimuth vector is preset. This settlement azimuth vector is a guide for the avoidance direction, and can significantly improve the reliability and stability of the force sensor to accurately monitor the settlement data between adjacent layers compared with the traditional installation planning method of randomly selecting the layout azimuth. Finally, by simulating the mean square error of the deployment of the force sensor at the expected layout estimation azimuth of each layer, this deployment mean square error represents the performance index of the force sensor to simulate the monitoring of ground settlement at the expected layout estimation azimuth. If this deployment mean square error is less than the minimum deployment mean square error, it means that the simulation monitoring performance of the force sensor at this expected layout estimation azimuth is the best, so it is selected as the optimal monitoring deployment site of the force sensor. Through this method, the ground rock formation and historical compression coefficient of each settlement monitoring layer in the target stratified monitoring area can be used to sample and screen the layout site samples, so as to make the layout and installation site of the force sensor more reasonable, accurate and stable, optimize the monitoring performance of the force sensor for different settlement monitoring layers to the greatest extent, and significantly improve the accuracy of the ground settlement monitoring results.

[0023] More specifically, a force sensor is arranged at the optimal monitoring deployment site of each settlement monitoring layer to obtain the compressive stress sensing reading. The multi-dimensional settlement cell constructed by the compressive stress sensing reading is analyzed for settlement crossing, and the isosurface interpolation of the back settlement distance is performed to obtain the settlement vector of the compressive stress applied between the upper and lower settlement monitoring layers. As Figure 2 shown, it specifically includes the following steps: S202: A force sensor is arranged and installed at the optimal monitoring deployment site of each settlement monitoring layer to perform sensing monitoring on the ground settlement of the target stratified monitoring area, so as to obtain the compressive stress sensing reading of each settlement monitoring layer in the target stratified monitoring area; S204: Construct a multi-dimensional compressive stress density space of the settlement monitoring layer, obtain the contour threshold of the multi-dimensional compressive stress density space, assign the contour value of the compressive stress density according to the compressive stress sensing reading to construct a scalar corner point, and form a multi-dimensional settlement cell of the compressive stress density based on the scalar corner point; S206: If there is one or more contour values of the scalar corner points on the multi-dimensional settlement cell crossing below the contour threshold of the multi-dimensional compressive stress density space, it means that the multi-dimensional settlement cell is crossed by settlement, and it is marked as a multi-dimensional crossed settlement cell, and the scalar corner points passing through the isosurface on the boundary of the multi-dimensional crossed settlement cell are stripped and defined as settlement crossing scalar corner points; S208: Preset the compressive stress sensing reading in the reverse settlement distance weight function of the settlement change in time series, calculate the distance from each compressive stress sensing reading to the settlement monitoring layer when the force sensor does not produce a reading, obtain the reverse settlement distance, and calculate the reverse settlement distance through the reverse settlement distance weight function to obtain the reverse settlement distance weight value; S210: Use the linear intersection interpolation of the reverse settlement distance weight value to determine the settlement crossing point and the compressive stress normal vector, solve and connect the vertices of the equivalent grid based on the settlement crossing point and the compressive stress normal vector to determine the settlement vector of the soil layer compressive stress applied from the upper settlement monitoring layer to the lower settlement monitoring layer.

[0024] It should be noted that the settlement vector includes the direction, range, and trend of settlement. When the rock and soil layer settles, its surface will collapse or form staggered peaks, etc., which may cause the relative displacement between them to show phenomena such as continuous peaks and valleys or faults. Each layer of force sensors monitors the vertical stress change between the upper and lower soil layers. When the upper soil layer sinks, the settlement range, direction, etc. of the peaks and valleys or faults formed will generate additional compressive stress on the lower nodes, resulting in changes in the force sensor readings. Therefore, when obtaining the force sensor readings of each settlement monitoring layer, this method first spatializes the multi-dimensional compressive stress density of each settlement monitoring layer. Since there may be no settlement peaks, valleys, or faults in the upper soil layer before the force sensor monitors a reading, there is an isoline benchmark at this time, which can be used as the isoline threshold of the multi-dimensional compressive stress density space. When a compressive stress reading occurs, the soil layer sinks, causing certain settlement vector compressive stress application peaks and valleys and fault phenomena on the surface, which results in a fluctuating expression of equivalent undulations compared to before settlement. Therefore, further assign the isoline value of the corresponding compressive stress density according to the compressive stress sensing reading to construct scalar corner points, so as to establish a multi-dimensional settlement cell that reflects the undulations during settlement to visualize the settlement situation. Among them, if the isoline value of one or more scalar corner points on the multi-dimensional settlement cell crosses below the isoline threshold of the multi-dimensional compressive stress density space, it indicates that the soil at this place has sunk below the surface when no compressive stress reading was monitored, showing a settlement crossing sign compared to the initial surface state, representing that the soil surface at this place forms an undulation with a certain settlement vector. Therefore, it is correspondingly marked as a multi-dimensional crossing settlement cell, and at the same time, obtain the settlement crossing scalar corner points on its boundary that cross below the isosurface.

[0025] It should be noted that there may be one or more settlement crossing scalar corners below the equipotential surface. These settlement crossing scalar corners are important source data for equipotential expression. The crossing of the settlement crossing scalar corners causes the gradually generation and change of the applied compressive stress, and there is a close connection between the two. Therefore, in this method, the reverse settlement distance of the settlement monitoring layer is calculated when each compressive stress sensing reading reaches the force sensor without generating a reading. Briefly speaking, the reverse settlement distance is the settlement distance between the initial soil layer surface and the undulation change of the soil layer surface when the compressive stress reading gradually generates towards the force sensor. The reverse settlement distance weight value can reflect the applied weight of each compressive stress during settlement fluctuations. Then, based on the reverse settlement distance weight value, linear intersection interpolation is performed on each settlement crossing scalar corner to determine the settlement crossing intersection point and the compressive stress normal vector, so as to further draw the equipotential surface diagram when the soil layer surface undergoes settlement fluctuations. Through the equipotential surface diagram, the settlement vector of the compressive stress applied to the next soil layer during settlement of each settlement monitoring layer can be clearly and completely determined. Through this method, the vector and trend of the compressive stress applied to the next soil layer during settlement of different soil layers can be accurately and efficiently depicted in the form of an equipotential surface diagram, so as to visualize the current situation of settlement development in ground stratified monitoring, make the subsequent specific analysis and calculation of the settlement degree more accurate and reliable, and greatly improve the credibility of the monitoring results.

[0026] More specifically, the linear intersection interpolation using the reverse settlement distance weight value is used to determine the settlement crossing intersection point and the compressive stress normal vector, and the equipotential grid vertices are solved and connected based on the settlement crossing intersection point and the compressive stress normal vector to determine the settlement vector of the soil layer compressive stress applied by the previous settlement monitoring layer to the next settlement monitoring layer. The specific steps are as follows: Perform linear intersection interpolation on one or more settlement crossing scalar corners through the reverse settlement distance weight value to determine the settlement crossing intersection point, and obtain the compressive stress normal vector of the settlement crossing intersection point in the multi-dimensional compressive stress density space; Define a crossing equipotential vertex for each multi-dimensional crossing settlement cell, construct a quadratic error function based on the settlement crossing intersection point and the compressive stress normal vector, solve the quadratic error function, and generate the equipotential grid vertices of the multi-dimensional crossing settlement cell; If there are crossing equipotential vertices in the cells adjacent to the current multi-dimensional crossing settlement cell, connect the equipotential grid vertices of each other to obtain the settlement compressive stress equipotential surface diagram of the settlement monitoring layer, and analyze the settlement compressive stress equipotential surface diagram to determine the settlement vector of the soil layer compressive stress applied by the previous settlement monitoring layer to the next settlement monitoring layer.

[0027] It should be noted that by linearly intersecting and interpolating one or more settlement crossing scalar corner points through the anti-settlement distance weight value, the applied compressive stress weight can be assigned to the multi-dimensional settlement cells with settlement fluctuations, so that the multi-dimensional settlement cells corresponding to the settlement fluctuations representing the formation of peaks, valleys or faults can define the equivalent data of corresponding hierarchical fluctuations with the change of the compressive stress readings in time series. As a result, the equivalent surface of the settlement fluctuations has a more accurate compressive stress expression value, improving the visualization accuracy of the equivalent surface diagram. Since the grid vertices of the equivalent surface are no longer restricted to the edges of the multi-dimensional crossing settlement cells but can freely drift to the optimal positions, the quadratic error function constructed by the compressive stress normal vectors of the settlement crossing intersection points after interpolation and the corresponding solution can better restore complex or sharp shapes (such as corners), making the description of the equivalent surfaces of different peaks, valleys and faults formed by settlement more plastic and followable, and maximizing the restoration of the detailed expression of the settlement vector and trend. If there are crossing equivalent vertices in the cells adjacent to the current multi-dimensional crossing settlement cells, it indicates that these adjacent cells also have a crossing state due to certain settlement vectors and trends compared with the initial soil layer. Therefore, they are included in the equivalent category of the settlement fluctuations for joint connection expression, and finally the settlement compressive stress equivalent surface diagram of the settlement monitoring level is formed. Through this method, the visualization of the fluctuations of the compressive stress applied to the next soil layer during the settlement of each settlement monitoring level can be realized, so as to more accurately reveal the settlement vectors of each rock and soil layer and provide a more reliable analysis basis for the subsequent calculation of the overall settlement degree.

[0028] More specifically, the inversion and inference of the settlement vector is used to obtain the relative settlement displacement amplitude of each settlement monitoring level, and the relative settlement displacement amplitude is linearly correlated to perform the trend reconstruction and quantitative projection embedding of the relative displacement between each layer of settlement monitoring levels, so as to obtain the ground settlement monitoring quantity of the target stratified monitoring area, which specifically includes the following steps: Obtain the prior distribution of the belief displacement of the soil layer settlement in the target stratified monitoring area through the ground rock formation structure, the formation lithology characteristics and the settlement elements in the big data network, and construct a displacement inversion model based on the prior node pattern and parameters of the prior distribution of the belief displacement; Introduce the maximum likelihood method to calculate the settlement vector between each settlement monitoring level, obtain the settlement likelihood function of each settlement monitoring level, and use the displacement inversion model to infer the posterior displacement of the settlement monitoring level under the condition of the settlement vector corresponding to the settlement likelihood function, so as to obtain the relative settlement displacement amplitude of each settlement monitoring level; Obtain the response time series step length of the force sensor monitoring output compressive stress sensing readings of adjacent settlement monitoring levels, and preset the reconstruction error threshold of the relative displacement between adjacent settlement monitoring levels based on the response time series step length; Introduce the weights between the relative settlement displacement amplitudes of each layer of settlement monitoring levels and those of adjacent settlement monitoring levels through least - squares neighbor linear correlation until the reconstruction error of neighbor linear correlation is less than the relative displacement reconstruction error threshold, and obtain the global correlation weight matrix; Construct a low - dimensional settlement space, and use the global correlation weight matrix to solve for K settlement trend eigenvalues and the corresponding settlement trend eigenvectors of each settlement trend eigenvalue; Based on the K settlement trend eigenvalues, embed the settlement trend eigenvectors into the low - dimensional settlement space to perform a quantitative projection expression of the relative displacement between each layer of settlement monitoring levels, and finally obtain the ground settlement monitoring quantity of the target stratified monitoring area.

[0029] It should be noted that when settlement occurs in each layer of settlement monitoring level, the settlement vector that applies compressive stress to the force sensor is the base structure for settlement amount monitoring, which depicts the magnitude and amplitude of the settlement development of each layer of rock and soil. However, since the settlement vector contains different multi-dimensional settlement indicators, it is difficult to quantify these multi-dimensional settlement indicators into accurate settlement monitoring values using traditional methods. In response to this, this method uses a displacement inversion model constructed based on the prior distribution of belief displacement of soil layer settlement in the target stratified monitoring area to globally invert and infer the settlement vector of each monitoring layer, thereby clarifying the specific amplitude of the relative settlement displacement due to settlement between adjacent settlement monitoring levels. By inferring the actual soil settlement through the prior distribution of historical data, the calculation of relative displacement can be made more accurate and saturated compared to traditional methods, ensuring that the subsequent calculation of the settlement degree is more stable and reliable. Since there will be a certain time sequence step in the monitoring response of the force sensor when relative settlement displacement occurs between each layer of soil under the condition of applying the settlement vector, which is the trend connection error of each multi-dimensional index in the settlement vector, a reconstruction error threshold for the relative displacement between adjacent settlement monitoring levels is preset based on the response time sequence step, and the least squares method is introduced to linearly correlate and combine neighbor points to reconstruct the weight of the relative settlement displacement amplitude between each layer, thereby extracting and encoding the local geometric structure of the data points located within the settlement vector. For example, if a relative settlement displacement amplitude is a bit to the left among its neighbors in the settlement vector, this information is saved in the corresponding weight, thus obtaining a global correlation weight matrix that describes the relative settlement displacement trend of each multi-dimensional index in the settlement vector. Finally, the settlement trend eigenvalue and the corresponding settlement trend eigenvector are solved to further embed them into a low-dimensional settlement space for projection expression of the settlement trend, and then the ground settlement monitoring quantity of the target stratified monitoring area can be accurately quantified. On the one hand, this method can replace the cumbersome steps of traditional manual monitoring for calculating ground settlement, saving time and effort, reducing manual intervention errors and unnecessary calculation processes, and improving the efficiency of stratified monitoring of ground settlement; on the other hand, it uses the various multi-dimensional indicators covered in the inversion settlement vector of the prior distribution to infer the relative settlement amplitude of each layer of monitored soil, discovers and reconstructs the settlement trend in the multi-dimensional indicators, and then quantifies the accurate ground settlement result, improving the monitoring accuracy of the specific settlement of rock and soil between different layers, optimizing the quality of monitoring output, and enhancing the credibility of monitoring results.

[0030] In the second aspect of the present invention, a ground settlement stratified monitoring system based on a force sensor is provided, as Figure 3 shown. The ground settlement stratified monitoring system includes a memory 31 and a processor 32. A ground settlement stratified monitoring method program is stored in the memory 31. When the ground settlement stratified monitoring method program is executed by the processor 32, the steps of any of the ground settlement stratified monitoring methods are implemented.

[0031] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A method for monitoring ground subsidence stratification based on force sensors, characterized in that: The following steps are involved: Construct a characteristic change inference network to estimate several historical spatiotemporal rock layer sample maps in the target stratified monitoring area, generate a directed edge map of rock layer settlement, and use the directed edge map to infer the edge probability of rock layer characteristic change by forward and reverse message transmission to determine the settlement monitoring level of the target stratified monitoring area; Extract the ground rock structure and historical compression coefficient of each settlement monitoring layer in the target layered monitoring area, avoid the area prone to relative settlement based on the ground rock structure and historical compression coefficient, and reasonably arrange the sites, obtain the optimal monitoring deployment site and add it to the force sensor deployment plan; Force sensors are deployed at the optimal monitoring deployment sites of each settlement monitoring level to obtain compressive stress sensor readings. The settlement crosses the multidimensional settlement cells constructed by analyzing the compressive stress sensor readings to perform isosurface interpolation of the anti-settlement distance, and obtain the settlement vector imposed by the compressive stress between the upper and lower settlement monitoring levels. The settlement vector is inverted and inferred to obtain the relative settlement displacement amplitude of each settlement monitoring layer. The relative settlement displacement amplitude is linearly correlated to perform trend reconstruction and quantitative projection embedding of the relative displacement between each settlement monitoring layer to obtain the ground settlement monitoring amount in the target layered monitoring area.

2. A method for monitoring ground subsidence stratification based on a force sensor according to claim 1, characterized in that: The method of constructing a characteristic change inference network to estimate several historical spatiotemporal rock layer sample maps in the target stratified monitoring area, generating a directed edge map of rock layer settlement, and using the directed edge map to forwardly and reversely transfer messages to infer the edge probability of rock layer characteristic change, so as to determine the settlement monitoring level of the target stratified monitoring area, specifically includes the following steps: Obtain the target hierarchical monitoring area and monitoring tasks for ground subsidence, and define the termination monitoring point of ground subsidence according to the monitoring tasks; By monitoring the central control log, several historical spatiotemporal rock layer sample maps of the target layered monitoring area where the ground subsided to the termination monitoring point within the preset time period were extracted, and the local binary pattern algorithm was introduced to calculate the characteristics of each historical spatiotemporal rock layer sample map to obtain the LBP rock layer characteristic value of each historical spatiotemporal rock layer sample map; Construct a characteristic change inference network, and estimate the conditional probability of the rock formation characteristics of a certain historical space-time rock formation sample map transferring to another historical space-time rock formation sample map as the sedimentation time series is followed based on the LBP rock formation characteristic values ​​in the characteristic change inference network, and obtain the joint probability of rock formation characteristic changes between each historical space-time rock formation sample map; Define the historical spatiotemporal rock layer sample graph as a random variable node, construct the factor node of each random variable node according to the joint probability of rock layer characteristic change, and connect each random variable node based on the structure of factor nodes to build a directed edge graph of rock layer settlement; The stratum lithology characteristics and settlement elements of the target layered monitoring area are obtained through big data, and based on the stratum structural characteristics and settlement elements, messages are forwardly transmitted from the random variable node to the factor node along the stratum characteristic change edge in the directed edge graph. After the forward transmission is completed, messages are reversely transmitted from the factor node to the random variable node along the stratum characteristic change edge; Repeat the above steps to continuously update all messages of each historical spatiotemporal rock layer sample graph by forward and reverse transmission of messages along the edge in the directed edge graph until the maximum iteration frequency is reached, and generate multiple forward edge probability reasoning messages and multiple reverse edge probability reasoning messages; The edge restoration is calculated by combining a plurality of the forward edge probability inference messages with a plurality of reverse edge probability inference messages to obtain the edge probability of the rock formation characteristic change, and the settlement monitoring level of the target stratified monitoring area is determined based on the edge probability of the rock formation characteristic change.

3. The method for monitoring ground subsidence stratification based on a force sensor according to claim 1, characterized in that: The method of extracting the ground rock structure and historical compression coefficient of each settlement monitoring layer in the target layered monitoring area, avoiding the area prone to relative settlement based on the ground rock structure and historical compression coefficient, and reasonably arranging the sites, obtaining the optimal monitoring deployment site and adding it to the force sensor deployment plan, specifically includes the following steps: Obtain the corresponding model basic information of the force sensor to be used for ground subsidence monitoring in the target layered monitoring area, and retrieve the monitoring specification parameters of the force sensor in the big data network based on the model basic information; Based on big data, the settlement cases of the target layered monitoring area are obtained, and the ground rock structure and historical compression coefficient of each settlement monitoring layer in the target layered monitoring area are extracted through the settlement cases; Establish a deployment sampling area for each settlement monitoring level according to the ground rock structure, divide each deployment sampling area into several sub-sampling areas according to monitoring specification parameters, and independently sample the deployment sites of force sensors in each sub-sampling area of ​​each deployment sampling area to obtain N deployment site samples of force sensors randomly provided in each sub-sampling area; The inter-layer compression characteristic difference between each sub-sampling area is calculated based on the historical compression coefficient. If the inter-layer compression characteristic difference is greater than the preset inter-layer compression characteristic difference threshold, the sub-sampling area corresponding to the compression characteristic difference is extracted and marked as an area prone to relative settlement, thereby generating a pattern of areas prone to relative settlement between each settlement monitoring layer. According to the pattern of the area prone to relative settlement, the settlement azimuth vector is preset, and the settlement azimuth vector is used as the avoidance reference benchmark to estimate the layout azimuth of the force sensor using N layout site samples, so as to obtain the expected layout estimated azimuth of each settlement monitoring level; The minimum deployment mean square error is preset, the simulated monitoring variance and the simulated monitoring cost of the force sensor simulated deployment at the expected deployment estimated position of each layer are obtained, and the deployment mean square error of each deployment site sample is calculated based on the simulated monitoring variance and the simulated monitoring cost; Only the deployment site samples corresponding to the unique deployment mean square error less than the minimum deployment mean square error are extracted and calibrated as the optimal monitoring deployment site, and the optimal monitoring deployment site is added to the force sensor deployment plan.

4. The method for monitoring ground subsidence stratification based on a force sensor according to claim 1, characterized in that: The method comprises the following steps: placing force sensors at the optimal monitoring deployment sites of each settlement monitoring level to obtain compressive stress sensor readings; performing isosurface interpolation of anti-settlement distances on a multidimensional settlement cell constructed by analyzing the compressive stress sensor readings; and obtaining a settlement vector applied by compressive stress between upper and lower settlement monitoring levels. The ground settlement in the target layered monitoring area is monitored by installing force sensors at the optimal monitoring deployment sites of each settlement monitoring layer, so as to obtain the compressive stress sensor readings of each settlement monitoring layer in the target layered monitoring area; Construct a multi-dimensional compressive stress density space of the settlement monitoring level, obtain the contour threshold of the multi-dimensional compressive stress density space, assign the contour value of the compressive stress density according to the compressive stress sensor reading to construct a scalar corner point, and form a multi-dimensional settlement unit cell of the compressive stress density based on the scalar corner point; If there is one or more scalar corner points on the multidimensional settlement cell whose contour values ​​cross below the contour threshold of the multidimensional compressive stress density space, it means that the multidimensional settlement cell is crossed by settlement and is marked as a multidimensional crossing settlement cell. The scalar corner points on the boundary of the multidimensional crossing settlement cell that cross the contour surface are stripped out and defined as settlement crossing scalar corner points. Preset the anti-settlement distance weight function of the compressive stress sensor reading in the time series settlement change, calculate the distance of each compressive stress sensor reading to the settlement monitoring level when the force sensor does not produce a reading, and obtain the reverse settlement distance. Calculate the reverse settlement distance through the anti-settlement distance weight function to obtain the anti-settlement distance weight value; The settlement crossing intersection point and the compressive stress normal vector are determined by linear intersection interpolation of the anti-settlement distance weight value. Based on the settlement crossing intersection point and the compressive stress normal vector, the equivalent grid vertices are solved and connected to determine the settlement vector of the next settlement monitoring layer that applies the soil compressive stress of the previous settlement monitoring layer.

5. A method for monitoring ground subsidence stratification based on a force sensor according to claim 4, characterized in that: The method uses linear intersection interpolation of anti-settlement distance weight values ​​to determine the settlement crossing intersection point and the compressive stress normal vector, solves and connects the equivalent grid vertices based on the settlement crossing intersection point and the compressive stress normal vector, so as to determine the settlement vector and settlement trend of the next settlement monitoring layer by applying soil compressive stress at the previous settlement monitoring layer, specifically including the following steps: Performing linear intersection interpolation on one or more settlement crossing scalar corner points through anti-settlement distance weight values ​​to determine the settlement crossing intersection point, and obtaining the compressive stress normal vector of the settlement crossing intersection point in the multi-dimensional compressive stress density space; For each multi-dimensional crossing settlement cell, define a crossing equivalent vertex, construct a quadratic error function based on the settlement crossing intersection point and the compressive stress normal vector, solve the quadratic error function, and generate the equivalent mesh vertices of the multi-dimensional crossing settlement cell; If there are crossing equivalued vertices in the cells adjacent to the current multi-dimensional crossing settlement cell, the equivalued mesh vertices are connected to obtain the settlement compressive stress isosurface map of the settlement monitoring level. The settlement compressive stress isosurface map is analyzed to determine the settlement vector of the next settlement monitoring level that applies soil compressive stress at the previous settlement monitoring level.

6. The method for monitoring ground subsidence stratification based on force sensors according to claim 1, characterized in that: The inversion and inference of the settlement vector is used to obtain the relative settlement displacement amplitude of each settlement monitoring layer, and the relative settlement displacement amplitude is linearly associated to perform trend reconstruction and quantitative projection embedding of the relative displacement between each settlement monitoring layer to obtain the ground settlement monitoring amount of the target layered monitoring area, which specifically includes the following steps: The belief displacement prior distribution of soil settlement in the target layer monitoring area is obtained through the ground rock structure, stratum lithology characteristics and settlement elements in the big data network, and the displacement inversion model is constructed based on the prior node pattern and parameters of the belief displacement prior distribution; The maximum likelihood method is introduced to calculate the settlement vectors between the settlement monitoring levels to obtain the settlement likelihood function of each settlement monitoring level. The displacement inversion model is used to infer the displacement posteriori of the settlement monitoring level under the condition of the settlement vector corresponding to the settlement likelihood function to obtain the relative settlement displacement amplitude of each settlement monitoring level. Obtaining the response timing step length when the force sensors of adjacent settlement monitoring levels monitor and output the compressive stress sensing readings, and presetting the reconstruction error threshold of the relative displacement between adjacent settlement monitoring levels based on the response timing step length; The weight between the relative settlement displacement amplitude of each settlement monitoring layer and the relative settlement displacement amplitude of the adjacent settlement monitoring layer is introduced into the least squares neighbor linear association until the reconstruction error of the neighbor linear association is less than the relative displacement reconstruction error threshold, and the global association weight matrix is ​​obtained; Construct a low-dimensional settlement space, and use the global correlation weight matrix to obtain K settlement trend eigenvalues ​​and the settlement trend eigenvectors corresponding to each settlement trend eigenvalue; Based on K settlement trend characteristic values, the settlement trend characteristic vector is embedded into the low-dimensional settlement space to perform quantitative projection expression of the relative displacement between each settlement monitoring layer, and finally the ground settlement monitoring amount of the target layered monitoring area is obtained.

7. A ground subsidence stratification monitoring system based on force sensors, characterized in that: The ground subsidence stratification monitoring system includes a memory and a processor. The memory stores a ground subsidence stratification monitoring method program based on a force sensor. When the ground subsidence stratification monitoring method program is executed by the processor, the ground subsidence stratification monitoring method steps as described in any one of claims 1-6 are implemented.

Images