Soft soil treatment engineering settlement real-time monitoring method and system

Abstract

The present application relates to the technical field of intelligent control, and discloses a soft soil treatment engineering settlement real-time monitoring method and system, the method comprising: collecting multi-dimensional strain data flow in the soft soil in real time through a continuous monitoring network arranged in the soft soil treatment area; extracting a strain gradient tensor of the spatial position coordinates in the spatial gradient field in the multi-dimensional strain data flow, and analyzing the strain change rate of the spatial position based on the time stamp in the multi-dimensional strain data flow; cross-fusing the strain gradient tensor and the strain change rate to obtain a dynamic coupling feature matrix; marking a multi-level risk area according to the gradient mutation feature; superimposing the corresponding strain change rate to the spatial coordinates of the soft soil treatment area to obtain a settlement evolution cloud chart; generating a monitoring parameter adjustment strategy dynamically through fuzzy control rules based on the settlement evolution cloud chart, and outputting the monitoring parameter adjustment strategy to an execution terminal of the construction machinery; and the present application can improve the accuracy of soft soil treatment engineering settlement real-time monitoring.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology, and in particular to a method and system for real-time monitoring of settlement in soft soil treatment projects. Background Technology

[0002] In the field of settlement monitoring in soft soil treatment projects, existing technologies mostly rely on discrete monitoring point layouts, making it difficult to comprehensively capture the strain distribution characteristics within the soft soil area, resulting in biased judgments on settlement trends. Furthermore, their limited data acquisition frequency and coverage fail to reflect the dynamic changes of soft soil during construction in real time, leading to delays in identifying potential risks and failing to meet the immediate needs of engineering safety management.

[0003] Meanwhile, in existing technologies, the data processing and analysis stages are relatively independent, failing to effectively integrate the spatial and temporal characteristics of strain data, resulting in insufficient accuracy of the generated monitoring results. Engineering adjustment strategies developed based on this often lack specificity and cannot be dynamically optimized according to the actual settlement evolution, thus affecting the construction efficiency and quality control of soft soil treatment projects. Summary of the Invention

[0004] This invention provides a method and system for real-time monitoring of settlement in soft soil treatment projects, in order to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides a method for real-time monitoring of settlement in soft soil treatment engineering, comprising:

[0006] S1. Real-time acquisition of multi-dimensional strain data streams within the soft soil through a continuous monitoring network deployed in the soft soil treatment area;

[0007] S2. Extract the strain gradient tensor of the spatial location coordinates in the spatial gradient field from the multidimensional strain data stream, and analyze the strain change rate of the spatial location based on the timestamp in the multidimensional strain data stream.

[0008] S3. Perform feature cross-fusion of the strain gradient tensor and the strain change rate to obtain the dynamic coupling feature matrix of the soft soil treatment area;

[0009] S4. Mark the multi-level risk areas in the soft soil treatment area according to the gradient mutation characteristics in the dynamic coupling feature matrix;

[0010] S5. Based on the strain change rate corresponding to the multi-level risk area, superimpose it onto the spatial coordinates of the soft soil treatment area to obtain the settlement evolution cloud map of the soft soil treatment area.

[0011] S6. The settlement evolution cloud map is dynamically used to generate a monitoring parameter adjustment strategy for the soft soil treatment area through fuzzy control rules, and the monitoring parameter adjustment strategy is output to the execution terminal of the construction machinery.

[0012] In a preferred embodiment, the real-time acquisition of multidimensional strain data streams within the soft soil through a continuous monitoring network deployed in the soft soil treatment area includes:

[0013] Receives the light signal reflected from the soft soil treatment area;

[0014] The optical signal is converted into an electrical signal to obtain a multidimensional strain data stream inside the soft soil.

[0015] In a preferred embodiment, extracting the strain gradient tensor of the spatial location coordinates in the spatial gradient field from the multidimensional strain data stream includes:

[0016] Obtain the spatial coordinates of the sensing nodes in the multidimensional strain data stream;

[0017] The spatial location coordinates are mapped to the three-dimensional spatial grid coordinate system of the soft soil treatment area to obtain the gridded spatial topology of the soft soil treatment area.

[0018] Based on the aforementioned gridded spatial topology, the strain component differences of the sensing nodes in the orthogonal directions are calculated, wherein the formula for calculating the strain component differences is as follows:

[0019]

[0020] In the formula, For sensor nodes in Difference in strain components in the direction For sensing nodes in Strain component values ​​in the direction, For directly adjacent sensor nodes in Strain component values ​​in the direction;

[0021] The strain component differences are combined according to the tensor arrangement rules to form the strain gradient tensor of the soft soil treatment area.

[0022] In a preferred embodiment, the step of resolving the strain change rate based on the timestamp in the multidimensional strain data stream includes:

[0023] Extract historical strain data within a preset time window from the multidimensional strain data stream;

[0024] The strain history dataset is reconstructed sequentially into a spatiotemporal cube data structure of the soft soil treatment area according to the timestamp order.

[0025] Based on the aforementioned spatiotemporal cube data structure, a first-order differential operation in the time dimension is performed on the spatial location coordinates to obtain the strain change between adjacent time stamps. The formula for calculating the strain change is as follows:

[0026]

[0027] In the formula, In the time dimension The strain change between two adjacent time stamps at a point. In the time dimension The Middle Strain value for each timestamp In the time dimension Point the first The strain value of each timestamp;

[0028] Based on the timestamp information in the spatiotemporal cube data structure, the strain change rate of the soft soil treatment area is calculated, wherein the formula for calculating the strain change rate is as follows:

[0029]

[0030] In the formula, In the time dimension The rate of change of strain on In the time dimension The strain change between two adjacent time stamps at a point. This represents the time interval between adjacent timestamps.

[0031] In a preferred embodiment, the step of performing feature cross-fusion of the strain gradient tensor and the strain change rate to obtain the dynamic coupling feature matrix of the soft soil treatment area includes:

[0032] Align the spatial coordinates of the strain gradient tensor with the spatial coordinates of the strain rate of change to obtain a coordinate mapping index table for the soft soil treatment area.

[0033] Based on the spatial coordinates of the strain gradient tensor, the elements of the strain gradient tensor are extracted to obtain the three-dimensional tensor elements of the strain gradient tensor.

[0034] Based on the spatial coordinates of the strain rate of change, a scalar value is extracted from the strain rate of change to obtain the scalar value of the strain rate of change.

[0035] The feature dimensions of the three-dimensional tensor elements and the scalar value of the rate of change are fused to obtain a four-dimensional fused feature vector of the soft soil treatment area.

[0036] The four-dimensional fused feature vectors are arranged in spatial topological order;

[0037] Based on the coordinate mapping index table, the arranged four-dimensional fused feature vectors are filled into the three-dimensional spatial matrix framework of the soft soil treatment area;

[0038] The dynamic coupling feature matrix of the soft soil treatment area is obtained by performing null value interpolation on the filled three-dimensional spatial matrix framework.

[0039] In a preferred embodiment, marking the multi-level risk areas in the soft soil treatment area based on the gradient abrupt change characteristics in the dynamic coupling feature matrix includes:

[0040] The dynamic coupling feature matrix is ​​spatially divided to obtain a grid array of the soft soil treatment area;

[0041] Identify the aggregation state of feature vector directions in the grid array;

[0042] When the aggregation state is non-uniform, the grid array is risk-marked to obtain the potential risk spatial distribution mark of the grid array;

[0043] By performing regional connectivity on the spatially adjacent potential risk spatial distribution markers, the initial risk area of ​​the soft soil treatment area is obtained;

[0044] In the initial risk area, based on the difference in aggregation intensity of the feature vector direction, the aggregation intensity of the feature vector direction is classified to obtain a multi-risk level area of ​​the soft soil treatment area;

[0045] By assigning corresponding risk level identifiers to the multi-risk level areas, a multi-level risk area distribution map of the soft soil treatment area is obtained.

[0046] In a preferred embodiment, the step of superimposing the strain change rates corresponding to the multi-level risk zones onto the spatial coordinates of the soft soil treatment area to obtain a settlement evolution cloud map of the soft soil treatment area includes:

[0047] Spatial location boundary information of the multi-level risk areas is extracted to obtain the spatial outline of the multi-level risk areas;

[0048] In the strain change rate, the spatial position of the spatial contour matching is located to obtain the coordinate mapping point of the multi-level risk area;

[0049] Based on the strain change rate corresponding to the coordinate mapping point, the monitoring feature values ​​of the multi-level risk area are integrated to obtain the exclusive change rate data of the multi-level risk area.

[0050] The specific rate of change data is registered with the three-dimensional spatial coordinate system of the soft soil treatment area to obtain the spatially registered rate of change data of the soft soil treatment area.

[0051] Based on the spatial registration change rate data, the spatial trend of the change rate of the spatial location is reconstructed to obtain the settlement trend characterization value of the soft soil treatment area.

[0052] Spatial morphological extrapolation was performed on the settlement trend characterization values ​​to obtain the continuous settlement change surface of the soft soil treatment area;

[0053] The continuous settlement change surface is visualized and rendered to obtain a settlement evolution cloud map of the soft soil treatment area.

[0054] In a preferred embodiment, the step of dynamically generating a monitoring parameter adjustment strategy for the soft soil treatment area using the settlement evolution cloud map through fuzzy control rules, and outputting the monitoring parameter adjustment strategy to the execution terminal of the construction machinery, includes:

[0055] Based on the regional colorimetric characteristics of the settlement evolution cloud map, the risk level of the settlement evolution cloud map is extracted to obtain the colorimetric distribution area of ​​the soft soil treatment area;

[0056] Based on the mapping relationship between the regional chromaticity characteristics and the settlement trend in the settlement evolution cloud map, the settlement trend intensity is matched for the chromaticity distribution area to obtain the settlement trend intensity classification map of the soft soil treatment area.

[0057] By matching and mapping the regional features in the settlement trend intensity grading map with the fuzzy control rule base, a preliminary parameter adjustment scheme for the soft soil treatment area is obtained.

[0058] Based on the constraint library of construction machinery, the preliminary parameter adjustment scheme is adapted to engineering constraints to obtain the monitoring parameter adjustment strategy for the soft soil treatment area.

[0059] In a preferred embodiment, the step of dynamically generating a monitoring parameter adjustment strategy for the soft soil treatment area using the settlement evolution cloud map through fuzzy control rules, and outputting the monitoring parameter adjustment strategy to the execution terminal of the construction machinery, includes:

[0060] Based on the communication protocol library of the construction machinery, the monitoring parameter adjustment strategy of the soft soil treatment area is encoded using engineering protocol to obtain the control command of the soft soil treatment area.

[0061] Based on industrial communication protocols, control commands that add spatial positioning identifiers and timestamp identifiers are encapsulated to obtain data packets of identifier commands;

[0062] The data packet is distributed to the corresponding construction machinery execution terminal.

[0063] To address the above problems, the present invention also provides a real-time settlement monitoring system for soft soil treatment projects, the system comprising:

[0064] The multidimensional strain monitoring module is used to collect multidimensional strain data streams inside the soft soil in real time through a continuous monitoring network deployed in the soft soil treatment area.

[0065] The strain field analysis module is used to extract the strain gradient tensor of the spatial location coordinates in the spatial gradient field in the multidimensional strain data stream, and to analyze the strain change rate of the spatial location based on the timestamp in the multidimensional strain data stream.

[0066] The dynamic coupling matrix generation module is used to perform feature cross-fusion of the strain gradient tensor and the strain change rate to obtain the dynamic coupling feature matrix of the soft soil treatment area.

[0067] A multi-level risk labeling module is used to label multi-level risk areas in the soft soil treatment area based on gradient mutation characteristics in the dynamic coupling feature matrix.

[0068] The settlement cloud map construction module is used to superimpose the strain change rate corresponding to the multi-level risk area onto the spatial coordinates of the soft soil treatment area to obtain the settlement evolution cloud map of the soft soil treatment area.

[0069] The strategy output execution module is used to dynamically generate a monitoring parameter adjustment strategy for the soft soil treatment area from the settlement evolution cloud map through fuzzy control rules, and output the monitoring parameter adjustment strategy to the execution terminal of the construction machinery.

[0070] Compared with the prior art, the present invention has the following beneficial effects:

[0071] 1. This invention deploys a continuous monitoring network to collect multidimensional strain data streams in real time, extracts the strain gradient tensor and analyzes the strain rate of change field, and performs feature cross-fusion to form a dynamic coupled feature matrix. This can accurately capture the strain dynamic characteristics of the soft soil treatment area, providing comprehensive and detailed data support for settlement monitoring. It effectively improves the ability to perceive the complex strain state inside the soft soil, making the monitoring process more real-time and accurate.

[0072] 2. This invention marks multi-level risk areas based on a dynamic coupling feature matrix, generates a settlement evolution cloud map by superimposing a strain rate of change field, and dynamically generates monitoring parameter adjustment strategies through fuzzy control rules. This achieves an integrated process from data acquisition and analysis to strategy output, which not only improves the efficiency of real-time settlement monitoring in soft soil treatment projects, but also provides targeted guidance for construction machinery to adjust parameters, thus helping to ensure the safety and stability of engineering construction. Attached Figure Description

[0073] Figure 1 This is a flowchart illustrating a method for real-time monitoring of settlement in soft soil treatment engineering according to an embodiment of the present invention.

[0074] Figure 2 This is a functional module diagram of a real-time settlement monitoring system for soft soil treatment engineering provided in an embodiment of the present invention;

[0075] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0076] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0077] This application provides a method for real-time monitoring of settlement in soft soil treatment projects. The executing entity of this method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the method for real-time monitoring of settlement in soft soil treatment projects can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0078] Reference Figure 1 The diagram shown is a flowchart illustrating a real-time settlement monitoring method for soft soil treatment projects according to an embodiment of the present invention. In this embodiment, the real-time settlement monitoring method for soft soil treatment projects includes:

[0079] S1. Real-time acquisition of multi-dimensional strain data streams within the soft soil through a continuous monitoring network deployed in the soft soil treatment area;

[0080] In this embodiment of the invention, the real-time acquisition of multi-dimensional strain data streams within the soft soil through a continuous monitoring network deployed in the soft soil treatment area includes:

[0081] Receives the light signal reflected from the soft soil treatment area;

[0082] The optical signal is converted into an electrical signal to obtain a multidimensional strain data stream inside the soft soil.

[0083] Specifically, distributed optical fiber sensors are pre-embedded inside the soft soil treatment area. These sensors are evenly distributed in the three-dimensional space of the soft soil treatment area and form a network structure. One end of the sensor is connected to the light source emitting device, and the other end is connected to the signal receiving device. The light source emitting device continuously emits laser signals of a specific wavelength to the distributed optical fiber sensors. During the transmission of the laser signal inside the optical fiber, it will interact with the soft soil medium.

[0084] Furthermore, when the soft soil deforms, the optical fiber is subjected to stress, causing changes in the phase, intensity, and other characteristics of the laser signal transmitted inside it. These laser signals, which carry the effects of the soft soil deformation and have changed characteristics, are reflected back to the signal receiving device along the optical fiber. The signal receiving device receives these reflected optical signals and temporarily stores them in the data cache module.

[0085] Furthermore, the signal receiving device transmits the buffered optical signal to the photoelectric conversion module, which includes a photodiode array. When the optical signal shines on the photodiode, it is converted into a corresponding electrical signal. During the conversion process, the intensity change of the optical signal corresponds to the amplitude of the electrical signal, and the phase change of the optical signal corresponds to the frequency change of the electrical signal.

[0086] Furthermore, through this conversion, the signal, which originally existed in the form of light, is transformed into an electrical signal that can be recognized by the subsequent processing unit. These electrical signals encompass the strain information of the soft soil at different locations and in different directions, and together constitute a multidimensional strain data stream inside the soft soil. This data stream is transmitted to the data storage unit for storage, so as to facilitate subsequent analysis and processing.

[0087] In summary, by receiving the light signals reflected from the soft soil treatment area and converting them into electrical signals to obtain multidimensional strain data streams, the high sensitivity of light signals to soft soil deformation can be used to accurately capture subtle strain changes inside the soft soil, ensuring that the collected data can truly reflect the state of the soft soil.

[0088] In summary, optical signals are less susceptible to external interference during transmission, ensuring data stability and reliability. This provides high-quality raw data support for subsequent steps such as strain gradient tensor extraction and strain rate analysis, helping to improve the accuracy and effectiveness of real-time settlement monitoring in the entire soft soil treatment project. It also enables the monitoring system to grasp the dynamic changes of soft soil more timely and accurately, providing strong data support for project safety management.

[0089] S2. Extract the strain gradient tensor of the spatial location coordinates in the spatial gradient field from the multidimensional strain data stream, and analyze the strain change rate of the spatial location based on the timestamp in the multidimensional strain data stream.

[0090] In this embodiment of the invention, extracting the strain gradient tensor of the spatial location coordinates in the spatial gradient field from the multidimensional strain data stream includes:

[0091] Obtain the spatial coordinates of the sensing nodes in the multidimensional strain data stream;

[0092] The spatial location coordinates are mapped to the three-dimensional spatial grid coordinate system of the soft soil treatment area to obtain the gridded spatial topology of the soft soil treatment area.

[0093] Based on the aforementioned gridded spatial topology, the strain component differences of the sensing nodes in the orthogonal directions are calculated, wherein the formula for calculating the strain component differences is as follows:

[0094]

[0095] In the formula, For sensor nodes in Difference in strain components in the direction For sensing nodes in Strain component values ​​in the direction, For directly adjacent sensor nodes in Strain component values ​​in the direction;

[0096] The strain component differences are combined according to the tensor arrangement rules to form the strain gradient tensor of the soft soil treatment area.

[0097] The strain change rate, which is derived from the timestamp-based spatial location analysis of the multidimensional strain data stream, includes:

[0098] Extract historical strain data within a preset time window from the multidimensional strain data stream;

[0099] The strain history dataset is reconstructed sequentially into a spatiotemporal cube data structure of the soft soil treatment area according to the timestamp order.

[0100] Based on the aforementioned spatiotemporal cube data structure, a first-order differential operation in the time dimension is performed on the spatial location coordinates to obtain the strain change between adjacent time stamps. The formula for calculating the strain change is as follows:

[0101]

[0102] In the formula, In the time dimension The strain change between two adjacent time stamps at a point. In the time dimension The Middle Strain value for each timestamp In the time dimension Point the first The strain value of each timestamp;

[0103] Based on the timestamp information in the spatiotemporal cube data structure, the strain change rate of the soft soil treatment area is calculated, wherein the formula for calculating the strain change rate is as follows:

[0104]

[0105] In the formula, In the time dimension The rate of change of strain on In the time dimension The strain change between two adjacent time stamps at a point. This represents the time interval between adjacent timestamps.

[0106] Specifically, the coordinate values ​​are extracted by the data parsing module and listed. Each coordinate value is associated with a corresponding sensing node to ensure that the spatial position of each node can be accurately identified in subsequent processing.

[0107] Furthermore, the spatial coordinates of all extracted sensor nodes are mapped to the three-dimensional spatial grid coordinate system corresponding to the soft soil treatment area according to the preset three-dimensional spatial grid division rules. This coordinate system takes a vertex of the soft soil treatment area as the origin and establishes coordinate axes along the length, width and height directions respectively. The size of the grid is set according to the distribution density of the sensor nodes. The coordinates of each sensor node are assigned to the corresponding grid cell. Through this mapping, a gridded spatial topology structure of the soft soil treatment area is formed. This structure clearly presents the distribution and mutual positional relationship of all sensor nodes in three-dimensional space.

[0108] Furthermore, based on the constructed gridded spatial topology, the directly adjacent sensor nodes of each sensor node in the three-dimensional orthogonal directions are determined. For each sensor node, the strain component value in the first orthogonal direction and the strain component value of the directly adjacent sensor node in that direction are obtained respectively. The strain component value of the current node is obtained by subtracting the strain component value of the adjacent node from the strain component value of the current node. The strain component difference of the node in the first orthogonal direction is calculated in the same way. The strain component difference in each direction is obtained by subtracting the strain component value of the adjacent node in the corresponding direction from the strain component value of the current node in the corresponding direction.

[0109] Furthermore, according to the tensor arrangement rules, the strain component differences of each sensing node in the three orthogonal directions are ordered and combined. Specifically, the strain component differences in the three directions are used as elements of the tensor and arranged sequentially in the order of the three orthogonal directions to form the strain gradient tensor element corresponding to the sensing node. The strain gradient tensor elements of all sensing nodes are integrated to finally form the strain gradient tensor of the entire soft soil treatment area. This tensor fully reflects the strain changes of each point in the soft soil treatment area in different orthogonal directions.

[0110] Specifically, all strain data within a preset time range are filtered from the multidimensional strain data stream. These data include the strain values ​​and corresponding time stamps of each sensing node at different times within the time range. The data filtering module extracts these data to form historical strain data within the preset time window. Each data entry corresponds to a specific spatial location coordinate and time stamp.

[0111] Furthermore, the extracted strain history data are arranged in chronological order according to the time markers, and combined with the spatial coordinates of each data point, a spatiotemporal cube data structure for the soft soil treatment area is constructed. This structure forms a four-dimensional framework with three spatial dimensions and a time dimension. The strain value of each spatial coordinate point under different time markers is accurately placed in the corresponding position of the structure, so that the data can reflect both the spatial distribution characteristics and the temporal variation pattern.

[0112] Furthermore, based on the constructed spatiotemporal cube data structure, for each spatial location coordinate point, all strain values ​​and corresponding time markers in the time dimension are extracted. The strain values ​​corresponding to two adjacent time markers are selected, and the strain value of the previous time marker is subtracted from the strain value of the latter time marker to obtain the strain change between the two adjacent time markers. The strain change between all adjacent time markers for each spatial location coordinate point is calculated in this way.

[0113] Furthermore, the time interval between two adjacent time markers is obtained from the spatiotemporal cube data structure. This interval is the difference between the time corresponding to the later time marker and the time corresponding to the earlier time marker. The strain change amount corresponding to the same set of adjacent time markers calculated earlier is divided by this time interval to obtain the strain change rate of the spatial coordinate point between these two adjacent time markers. In the same way, the strain change rate of each spatial coordinate point between all adjacent time markers is calculated, and finally the strain change rate data of the entire soft soil treatment area is formed.

[0114] In summary, by acquiring the spatial coordinates of the sensing nodes and mapping them to a three-dimensional grid coordinate system to form a gridded spatial topology, the spatial distribution and relationships of the nodes can be clearly defined. Based on this, the differences in strain components in orthogonal directions can be calculated and combined into a strain gradient tensor, which can systematically integrate strain differences in multiple directions, comprehensively characterize the spatial strain gradient distribution, provide a precise spatial feature basis for the subsequent generation of dynamic coupling feature matrices, improve the perception accuracy of complex strain states inside soft soil, and ensure the accuracy of subsequent risk marking and settlement cloud map construction.

[0115] In summary, by extracting historical strain data within a preset time window and reconstructing it into a spatiotemporal cube structure based on timestamps, the system can integrate temporal and spatial information and clearly present the strain state at each moment and location.

[0116] In summary, calculating the strain change and rate of change can accurately capture the dynamic trend of strain at each location over time, providing accurate time characteristics for the dynamic coupling feature matrix, improving the accuracy of understanding the strain time variation law of soft soil, and ensuring the accuracy of subsequent analysis in the time dimension.

[0117] S3. Perform feature cross-fusion of the strain gradient tensor and the strain change rate to obtain the dynamic coupling feature matrix of the soft soil treatment area;

[0118] In this embodiment of the invention, the step of performing feature cross-fusion of the strain gradient tensor and the strain change rate to obtain the dynamic coupling feature matrix of the soft soil treatment area includes:

[0119] Align the spatial coordinates of the strain gradient tensor with the spatial coordinates of the strain rate of change to obtain a coordinate mapping index table for the soft soil treatment area.

[0120] Based on the spatial coordinates of the strain gradient tensor, the elements of the strain gradient tensor are extracted to obtain the three-dimensional tensor elements of the strain gradient tensor.

[0121] Based on the spatial coordinates of the strain rate of change, a scalar value is extracted from the strain rate of change to obtain the scalar value of the strain rate of change.

[0122] The feature dimensions of the three-dimensional tensor elements and the scalar value of the rate of change are fused to obtain a four-dimensional fused feature vector of the soft soil treatment area.

[0123] The four-dimensional fused feature vectors are arranged in spatial topological order;

[0124] Based on the coordinate mapping index table, the arranged four-dimensional fused feature vectors are filled into the three-dimensional spatial matrix framework of the soft soil treatment area;

[0125] The dynamic coupling feature matrix of the soft soil treatment area is obtained by performing null value interpolation on the filled three-dimensional spatial matrix framework.

[0126] Specifically, all spatial coordinate points contained in the strain gradient tensor are extracted, and all spatial coordinate points contained in the strain rate of change are extracted simultaneously. The two sets of coordinate points are compared one by one to find coordinate points that completely overlap in spatial position. A unique correspondence is established for each pair of overlapping coordinate points, and these correspondences are recorded in order in a table to form a coordinate mapping index table for the soft soil treatment area. This table clearly shows the matching situation of spatial coordinate points in the strain gradient tensor and the strain rate of change.

[0127] Furthermore, using the spatial coordinates of the strain gradient tensor as a reference, the position of each coordinate point in the strain gradient tensor is located one by one. All tensor elements corresponding to the coordinate point are extracted from the position. These elements cover the strain component differences of the coordinate point in three orthogonal directions. These elements are combined in a preset order to form the three-dimensional tensor elements of the strain gradient tensor corresponding to the spatial coordinate point. The three-dimensional tensor elements of all spatial coordinate points are stored separately and associated with the corresponding coordinate points.

[0128] Furthermore, using the spatial coordinates of the strain rate of change as a reference, the position of each coordinate point in the strain rate of change is located one by one, and the strain rate of change value corresponding to that coordinate point is extracted from that position. This value is in scalar form and directly reflects the rate of change of strain with time at that coordinate point. The extracted value is used as the scalar value of the strain rate of change corresponding to that spatial coordinate point. The scalar values ​​of the rate of change of all spatial coordinate points are stored separately and associated with the corresponding coordinate points.

[0129] Furthermore, for each pair of matching spatial coordinate points in the coordinate mapping index table, the three-dimensional tensor element corresponding to the coordinate point is combined with the scalar value of the rate of change. The combination method is to add a dimension to the three-dimensional tensor element to accommodate the scalar value of the rate of change, so that the original three-dimensional tensor element is expanded into a four-dimensional vector containing spatial strain gradient features and time strain rate of change features. Each matching coordinate point forms a four-dimensional fused feature vector of the corresponding soft soil treatment area in this way.

[0130] Furthermore, according to the arrangement order of spatial coordinate points in the gridded spatial topology of the soft soil treatment area, all four-dimensional fused feature vectors are sorted. During sorting, the position of the spatial coordinate points corresponding to each vector in the topology is taken into account to ensure that the four-dimensional fused feature vectors of adjacent spatial positions remain adjacent after sorting, so that the sorted vector sequence can reflect the spatial distribution characteristics of the soft soil treatment area.

[0131] Furthermore, based on the correspondence recorded in the coordinate mapping index table, the arranged four-dimensional fused feature vectors are placed into the corresponding positions in the three-dimensional spatial matrix frame of the soft soil treatment area. Each position in the three-dimensional spatial matrix frame corresponds to a spatial coordinate point in the soft soil treatment area. During placement, it is ensured that each four-dimensional fused feature vector accurately falls into the position of its corresponding coordinate point in the frame, forming a preliminary filled three-dimensional spatial matrix.

[0132] Furthermore, the filled three-dimensional spatial matrix frame is examined to identify blank positions that have not been filled by the four-dimensional fused feature vectors. For these blank positions, the average value of the four-dimensional fused feature vectors of their adjacent filled positions is used to fill them. When calculating the average value, the vectors of the nearest multiple filled positions around the blank position are selected, and the corresponding elements of these vectors are averaged to obtain the filling value of the blank position. After filling, there are no more blanks in the three-dimensional spatial matrix frame, forming a complete dynamic coupling feature matrix of the soft soil treatment area.

[0133] In summary, by aligning coordinates, extracting features, and fusing them, the spatial strain gradient and the temporal strain rate of change are integrated into a four-dimensional vector. After arrangement, filling, and null interpolation, a dynamic coupled feature matrix is ​​formed, which can accurately couple spatiotemporal features, providing complete and accurate feature data for subsequent analysis and improving the comprehensive perception of complex strain in soft soil.

[0134] S4. Mark the multi-level risk areas in the soft soil treatment area according to the gradient mutation characteristics in the dynamic coupling feature matrix;

[0135] In this embodiment of the invention, marking the multi-level risk areas in the soft soil treatment area based on the gradient abrupt change characteristics in the dynamic coupling feature matrix includes:

[0136] The dynamic coupling feature matrix is ​​spatially divided to obtain a grid array of the soft soil treatment area;

[0137] Identify the aggregation state of feature vector directions in the grid array;

[0138] When the aggregation state is non-uniform, the grid array is risk-marked to obtain the potential risk spatial distribution mark of the grid array;

[0139] By performing regional connectivity on the spatially adjacent potential risk spatial distribution markers, the initial risk area of ​​the soft soil treatment area is obtained;

[0140] In the initial risk area, based on the difference in aggregation intensity of the feature vector direction, the aggregation intensity of the feature vector direction is classified to obtain a multi-risk level area of ​​the soft soil treatment area;

[0141] By assigning corresponding risk level identifiers to the multi-risk level areas, a multi-level risk area distribution map of the soft soil treatment area is obtained.

[0142] Specifically, the dynamic coupling feature matrix is ​​uniformly divided according to a preset size. Each sub-region after division is a grid cell. All grid cells are arranged in an orderly manner to form a grid array of the soft soil treatment area. Each grid cell contains all the dynamic coupling feature vectors in the area, and the boundaries between grid cells are clear and non-overlapping, completely covering the entire soft soil treatment area.

[0143] Furthermore, the eigenvector directions within each grid cell of the grid array are analyzed one by one, and the number of eigenvectors in each direction is counted. By comparing the distribution of the number of eigenvectors in different directions, the clustering state of the eigenvector directions is determined. If most eigenvectors point in the same or similar directions, the clustering state is uniform; if the eigenvectors are distributed in multiple directions and the difference in the number of each direction is small, the clustering state is non-uniform.

[0144] Furthermore, when the eigenvector direction aggregation state of a grid cell is identified as non-uniform, a specific identifier is added to the grid cell to mark it as having potential risks. All the marked grid cells together constitute the potential risk spatial distribution marker of the grid array, and each marker corresponds precisely to a grid cell with potential risks.

[0145] Furthermore, all grid cells with potential risk spatial distribution markers in the grid array are traversed, and each marked grid cell is checked to see if there is a spatial adjacency relationship between it and other surrounding marked grid cells. For adjacent marked grid cells, they are merged into a continuous region. In this way, all spatially connected potential risk markers are integrated to form the initial risk region of the soft soil treatment area. Each initial risk region consists of multiple adjacent marked grid cells.

[0146] Furthermore, within each initial risk region, the clustering intensity in the direction of the feature vector is calculated. The clustering intensity is measured by the proportion of the number of feature vectors in the same direction to the total number of feature vectors in the region. The higher the proportion, the stronger the clustering intensity. Based on the numerical range of the clustering intensity, the clustering intensity in the direction of the feature vector is divided into different categories, and each category corresponds to a risk level, thus obtaining the multi-risk level region of the soft soil treatment area.

[0147] Furthermore, a unique risk level identifier is assigned to each different risk level category in the multi-risk level area. The identifier uses different numbers or symbols to distinguish different risk levels. These identifiers are associated with the corresponding multi-risk level areas and marked on the spatial distribution map to form a multi-level risk area distribution map of the soft soil treatment area. This map clearly shows the spatial distribution of different risk level areas in the soft soil treatment area.

[0148] In summary, by dividing the grid array and identifying the clustering state of feature vectors, marking non-uniform regions as potential risks, connecting adjacent risk regions to form initial risk zones, and then classifying and assigning level identifiers according to clustering intensity, multi-level risk distribution can be accurately presented, providing accurate risk information for subsequent analysis and improving the efficiency of risk identification and the targeting of control.

[0149] S5. Based on the strain change rate corresponding to the multi-level risk area, superimpose it onto the spatial coordinates of the soft soil treatment area to obtain the settlement evolution cloud map of the soft soil treatment area.

[0150] In this embodiment of the invention, the step of superimposing the strain change rate corresponding to the multi-level risk area onto the spatial coordinates of the soft soil treatment area to obtain the settlement evolution cloud map of the soft soil treatment area includes:

[0151] Spatial location boundary information of the multi-level risk areas is extracted to obtain the spatial outline of the multi-level risk areas;

[0152] In the strain change rate, the spatial position of the spatial contour matching is located to obtain the coordinate mapping point of the multi-level risk area;

[0153] Based on the strain change rate corresponding to the coordinate mapping point, the monitoring feature values ​​of the multi-level risk area are integrated to obtain the exclusive change rate data of the multi-level risk area.

[0154] The specific rate of change data is registered with the three-dimensional spatial coordinate system of the soft soil treatment area to obtain the spatially registered rate of change data of the soft soil treatment area.

[0155] Based on the spatial registration change rate data, the spatial trend of the change rate of the spatial location is reconstructed to obtain the settlement trend characterization value of the soft soil treatment area.

[0156] Spatial morphological extrapolation was performed on the settlement trend characterization values ​​to obtain the continuous settlement change surface of the soft soil treatment area;

[0157] The continuous settlement change surface is visualized and rendered to obtain a settlement evolution cloud map of the soft soil treatment area.

[0158] Specifically, from the spatial distribution data of multi-level risk areas, the spatial coordinates of the edge of each area are extracted. These coordinates are connected in sequence to form closed lines. Each line accurately outlines the outer boundary of the multi-level risk area. All lines together constitute the spatial outline of the multi-level risk area. This outline fully presents the spatial range and shape of each multi-level risk area in the soft soil treatment area.

[0159] Furthermore, in the database of strain change rates, based on the spatial location range contained in the spatial contour of the multi-level risk area, coordinate points that match the spatial location within that range are searched one by one. These coordinate points belong to the area defined by the spatial contour and also have corresponding values ​​in the strain change rate record. These found coordinate points are determined as coordinate mapping points of the multi-level risk area, and each coordinate mapping point corresponds to a specific location within the spatial contour.

[0160] Furthermore, the strain rate of change values ​​corresponding to all coordinate mapping points are collected, and these values ​​are grouped according to multi-level risk areas. All strain rate of change values ​​within the same multi-level risk area are grouped together. The values ​​of each group are statistically integrated, and the average value of the group is calculated. This average value is used as the representative value of the strain rate of change of the entire multi-level risk area, thus obtaining the exclusive rate of change data for the multi-level risk area. Each data point is associated with the corresponding multi-level risk area.

[0161] Furthermore, the spatial location of the multi-level risk area corresponding to the exclusive rate of change data is aligned with the three-dimensional spatial coordinate system of the soft soil treatment area. By adjusting the position parameters of the multi-level risk area in the coordinate system, the spatial location of the exclusive rate of change data is made to match the coordinate scale in the coordinate system, ensuring that each exclusive rate of change data can accurately correspond to a specific position in the three-dimensional spatial coordinate system. After alignment, spatial registration rate of change data of the soft soil treatment area is formed.

[0162] Furthermore, based on the spatial registration change rate data, the correlation between the change rate of each spatial location and the change rate of adjacent locations is analyzed. Based on the change rate values ​​of adjacent locations, the change rate trend of locations not directly measured is inferred. In this way, the gaps in the change rate data in space are filled, and a continuous change rate trend covering the entire soft soil treatment area is constructed, with each spatial location having a corresponding settlement trend characterization value.

[0163] Furthermore, based on the settlement trend characterization values, the overall shape of settlement changes in the soft soil treatment area is depicted. Starting from the spatial location of the existing characterization values, the model is extended to the surrounding area to determine the connection relationship of settlement trends at different locations, so that the settlement trends at each location form a continuous whole. Finally, a continuous settlement change surface that can reflect the settlement changes in the entire soft soil treatment area is constructed. This surface has a smooth transition and no obvious breaks.

[0164] Furthermore, the continuous settlement change surface is rendered with color. According to the magnitude of the settlement trend characterization value at different positions on the surface, the corresponding color is assigned. Areas with larger characterization values ​​use colors with obvious differences, while areas with similar characterization values ​​use a gradient color transition. The strength and distribution of the settlement trend are intuitively displayed through the depth and difference of the colors. The image formed after rendering is the settlement evolution cloud map of the soft soil treatment area.

[0165] In summary, by extracting the spatial contours of multi-level risk areas, locating and matching coordinate points, and integrating exclusive rate of change data, a continuous settlement surface is formed through registration, trend reconstruction, and morphological deduction. After visualization rendering, a settlement evolution cloud map is obtained, which can intuitively present the relationship between settlement trends and risks, providing a clear basis for subsequent strategy formulation.

[0166] S6. The settlement evolution cloud map is dynamically used to generate a monitoring parameter adjustment strategy for the soft soil treatment area through fuzzy control rules, and the monitoring parameter adjustment strategy is output to the execution terminal of the construction machinery.

[0167] In this embodiment of the invention, the step of dynamically generating a monitoring parameter adjustment strategy for the soft soil treatment area using fuzzy control rules from the settlement evolution cloud map, and outputting the monitoring parameter adjustment strategy to the execution terminal of the construction machinery, includes:

[0168] Based on the regional colorimetric characteristics of the settlement evolution cloud map, the risk level of the settlement evolution cloud map is extracted to obtain the colorimetric distribution area of ​​the soft soil treatment area;

[0169] Based on the mapping relationship between the regional chromaticity characteristics and the settlement trend in the settlement evolution cloud map, the settlement trend intensity is matched for the chromaticity distribution area to obtain the settlement trend intensity classification map of the soft soil treatment area.

[0170] By matching and mapping the regional features in the settlement trend intensity grading map with the fuzzy control rule base, a preliminary parameter adjustment scheme for the soft soil treatment area is obtained.

[0171] Based on the constraint library of construction machinery, the preliminary parameter adjustment scheme is adapted to engineering constraints to obtain the monitoring parameter adjustment strategy for the soft soil treatment area.

[0172] The step of dynamically generating a monitoring parameter adjustment strategy for the soft soil treatment area using fuzzy control rules from the settlement evolution cloud map, and outputting the monitoring parameter adjustment strategy to the execution terminal of the construction machinery, includes:

[0173] Based on the communication protocol library of the construction machinery, the monitoring parameter adjustment strategy of the soft soil treatment area is encoded using engineering protocol to obtain the control command of the soft soil treatment area.

[0174] Based on industrial communication protocols, control commands that add spatial positioning identifiers and timestamp identifiers are encapsulated to obtain data packets of identifier commands;

[0175] The data packet is distributed to the corresponding construction machinery execution terminal.

[0176] Specifically, the color types, shades, and distribution ranges of different areas in the settlement evolution cloud map are observed. These color features together constitute the regional color characteristics. According to the preset color and risk level correspondence standard, areas with the same or similar colors in the cloud map are divided into the same group. Each group of areas represents a risk level. Through this division, the color distribution area of ​​the soft soil treatment area is obtained. Each color distribution area has a clear boundary and corresponding color characteristics.

[0177] Furthermore, a pre-established mapping table between regional chromaticity characteristics and settlement trends is invoked. This table records the settlement trend type and range corresponding to each chromaticity characteristic. The chromaticity characteristics of the chromaticity distribution area are compared one by one with the entries in the table to find a perfectly matching mapping relationship. Based on this mapping relationship, the settlement trend intensity corresponding to each chromaticity distribution area is determined. The settlement trend intensity of all areas is marked on the map according to spatial location to form a settlement trend intensity grading map of the soft soil treatment area.

[0178] Furthermore, a fuzzy control rule base is established, which contains parameter adjustment principles and directions corresponding to different settlement trend intensity levels. The characteristics of each region in the settlement trend intensity classification map, including the settlement trend intensity level and the region location, are matched with the rules in the rule base to find the rule entry that best matches the characteristics of the region. Based on the matched rule entry, parameter adjustment suggestions for the region are generated. The adjustment suggestions for all regions are integrated to form a preliminary parameter adjustment scheme for the soft soil treatment area.

[0179] Furthermore, constraints on construction machinery are collected and a constraint database is established. These constraints include the maximum operating range of the machinery, the range of adjustable parameters, and operating restrictions under different geological conditions. The adjustments in the preliminary parameter adjustment plan are compared with the entries in the constraint database to check whether the adjustments conform to the actual capabilities and operating restrictions of the machinery. Adjustments that do not conform to the constraints are modified to meet the operating requirements of the machinery. The resulting plan after adaptation is the monitoring parameter adjustment strategy for the soft soil treatment area.

[0180] Specifically, based on the communication protocol library of the construction machinery, the protocol format matching the construction machinery is retrieved from it, and the contents of the monitoring parameter adjustment strategy of the soft soil treatment area are converted and filled according to the fields, order and data types required by the protocol format to complete the engineering protocol encoding of the monitoring parameter adjustment strategy, and finally the control command of the soft soil treatment area is obtained.

[0181] Furthermore, after obtaining the control command, the specific spatial location of the soft soil treatment area corresponding to the control command is determined, a unique spatial positioning identifier is generated and added to the control command; at the same time, the precise time of generating the control command is recorded, a unique timestamp identifier is generated and added to the control command, and then, according to the data packet structure specified by the industrial communication protocol, the control command with added spatial positioning identifier and timestamp identifier is used as the data part, and necessary information such as protocol header and check code is added and integrated and encapsulated to obtain the data packet of the identifier command.

[0182] Furthermore, through the communication network between the construction machinery execution terminal and the control center, the identification information of each construction machinery execution terminal is queried, the spatial positioning identifier in the data packet is matched with the work area responsible for each construction machinery execution terminal, the corresponding construction machinery execution terminal is found, and then the data packet is sent to the execution terminal through the communication network to complete the distribution of the data packet.

[0183] In summary, risk levels are extracted by using the color features of settlement evolution cloud maps, settlement trend intensity is matched and a preliminary plan is generated by combining fuzzy control rules, and monitoring parameter adjustment strategies are obtained by adapting to construction machinery constraints. This can dynamically generate highly targeted strategies and improve the accuracy of construction control.

[0184] In summary, control commands are obtained by encoding monitoring parameters and adjusting strategies based on the construction machinery communication protocol library. After adding identifiers, the commands are encapsulated into data packets according to industrial protocols and distributed to the corresponding execution terminals, ensuring accurate transmission and execution of commands and improving the timeliness and effectiveness of construction control.

[0185] like Figure 2 The diagram shown is a functional block diagram of a real-time settlement monitoring system for soft soil treatment engineering provided in an embodiment of the present invention.

[0186] The real-time settlement monitoring system 100 for soft soil treatment projects described in this invention can be installed in an electronic device. Depending on the functions implemented, the real-time settlement monitoring system 100 for soft soil treatment projects may include a multi-dimensional strain monitoring module 101, a strain field analysis module 102, a strain field analysis module 103, a multi-level risk marking module 104, a settlement cloud map construction module 105, and a strategy output execution module 106. The module described in this invention can also be called a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and are stored in the memory of the electronic device.

[0187] In this embodiment, the functions of each module / unit are as follows:

[0188] The multidimensional strain monitoring module 101 is used to collect multidimensional strain data streams inside the soft soil in real time through a continuous monitoring network deployed in the soft soil treatment area.

[0189] The strain field analysis module 102 is used to extract the strain gradient tensor of the spatial position coordinates in the spatial gradient field in the multidimensional strain data stream, and analyze the strain change rate of the spatial position based on the timestamp in the multidimensional strain data stream.

[0190] The dynamic coupling matrix generation module 103 is used to perform feature cross-fusion of the strain gradient tensor and the strain change rate to obtain the dynamic coupling feature matrix of the soft soil treatment area.

[0191] The multi-level risk marking module 104 is used to mark multi-level risk areas in the soft soil treatment area according to the gradient mutation characteristics in the dynamic coupling feature matrix.

[0192] The settlement cloud map construction module 105 is used to superimpose the strain change rate corresponding to the multi-level risk area onto the spatial coordinates of the soft soil treatment area to obtain the settlement evolution cloud map of the soft soil treatment area.

[0193] The strategy output execution module 106 is used to dynamically generate a monitoring parameter adjustment strategy for the soft soil treatment area from the settlement evolution cloud map through fuzzy control rules, and output the monitoring parameter adjustment strategy to the execution terminal of the construction machinery.

[0194] In the several embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0195] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0196] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0197] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0198] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0199] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A soft soil treatment engineering settlement real-time monitoring method, characterized in that, The method comprises: S1, collecting a multi-dimensional strain data stream inside soft soil in real time through a continuous monitoring network deployed in a soft soil treatment area, comprising: receiving a light signal reflected by the soft soil treatment area; performing electrical signal conversion on the light signal to obtain a multi-dimensional strain data stream inside the soft soil; S2, extracting a strain gradient tensor of a spatial position coordinate in a spatial gradient field in the multi-dimensional strain data stream, and analyzing a strain change rate of a spatial position based on a time stamp in the multi-dimensional strain data stream, comprising: obtaining a spatial position coordinate of a sensing node in the multi-dimensional strain data stream; mapping the spatial position coordinate to a three-dimensional spatial grid coordinate system of the soft soil treatment area to obtain a grid spatial topology structure of the soft soil treatment area; based on the grid spatial topology structure, calculating a strain component difference value of the sensing node in an orthogonal direction, wherein the strain component difference value calculation formula is as follows: ； wherein is the difference between the strain component values of the sensor nodes in direction, is the strain component value of the sensor node in direction, is the strain component value of the directly adjacent sensor node in direction; combining the strain component difference value into a strain gradient tensor of the soft soil treatment area according to a tensor arrangement rule; S3, cross-fusing the strain gradient tensor and the strain change rate to obtain a dynamic coupling feature matrix of the soft soil treatment area; S4, marking a multi-level risk area in the soft soil treatment area according to a gradient mutation feature in the dynamic coupling feature matrix; S5, superimposing a strain change rate corresponding to the multi-level risk area to a spatial coordinate of the soft soil treatment area to obtain a settlement evolution cloud chart of the soft soil treatment area; S6, dynamically generating a monitoring parameter adjustment strategy of the soft soil treatment area through a fuzzy control rule based on the settlement evolution cloud chart, and outputting the monitoring parameter adjustment strategy to an execution terminal of a construction machine.

2. The soft soil treatment work settlement real-time monitoring method according to claim 1, wherein The method comprises: extracting strain historical data in a preset time window from the multi-dimensional strain data stream; reconstructing a set of the strain historical data in a time stamp order into a space-time cubic data structure of the soft soil treatment area; based on the space-time cubic data structure, performing a first-order differential operation on a spatial position coordinate point in a time dimension to obtain a strain change amount of adjacent time stamps, wherein the strain change amount calculation formula is as follows: ； In the formula, In the time dimension The strain change between two adjacent time stamps at a point. In the time dimension The Middle Strain value for each timestamp In the time dimension Point the first The strain value of each timestamp; based on the time stamp information in the space-time cubic data structure, calculating a strain change rate of the soft soil treatment area, wherein the strain change rate calculation formula is as follows: ； wherein is the rate of change of strain in the time dimension , is the rate of change of strain in the time dimension , is the time interval between adjacent time stamps.

3. The soft soil treatment work settlement real-time monitoring method according to claim 1, wherein The method comprises: aligning a spatial coordinate point of the strain gradient tensor and a spatial coordinate point of the strain change rate to obtain a coordinate mapping index table of the soft soil treatment area; based on the spatial coordinate point of the strain gradient tensor, extracting an element of the strain gradient tensor to obtain a three-dimensional tensor element of the strain gradient tensor; based on the spatial coordinate point of the strain change rate, extracting a scalar value of the strain change rate to obtain a change rate scalar value of the strain change rate; characteristic dimension fusion is performed on the three-dimensional tensor element and the rate of change scalar value to obtain a four-dimensional fused feature vector of the soft soil treatment area; the four-dimensional fused feature vector is arranged in a spatial topological order; the arranged four-dimensional fused feature vector is filled into a three-dimensional space matrix framework of the soft soil treatment area based on the coordinate mapping index table; null value interpolation is performed on the filled three-dimensional space matrix framework to obtain a dynamic coupling feature matrix of the soft soil treatment area.

4. The soft soil treatment work settlement real-time monitoring method according to claim 1, wherein The multi-level risk area in the soft soil treatment area is marked according to the gradient mutation feature in the dynamic coupling feature matrix, including: spatial region division is performed on the dynamic coupling feature matrix to obtain a grid array of the soft soil treatment area; an aggregation state of the feature vector direction in the grid array is identified; when the aggregation state presents a non-uniform state, the grid array is marked for risk to obtain a potential risk spatial distribution mark of the grid array; spatially adjacent potential risk spatial distribution marks are regionally connected to obtain an initial risk area of the soft soil treatment area; in the initial risk area, the feature vector direction is classified according to the aggregation intensity difference of the feature vector direction to obtain a multi-risk level area of the soft soil treatment area; a corresponding risk level identifier is assigned to the multi-risk level area to obtain a multi-level risk area distribution map of the soft soil treatment area.

5. The soft soil treatment work settlement real-time monitoring method according to claim 1, wherein, The strain rate corresponding to the multi-level risk area is superimposed into the spatial coordinates of the soft soil treatment area to obtain a settlement evolution cloud chart of the soft soil treatment area, including: spatial position boundary information of the multi-level risk area is extracted to obtain a spatial profile of the multi-level risk area; a spatial position matched with the spatial profile is located in the strain rate to obtain a coordinate mapping point of the multi-level risk area; monitoring feature values of the multi-level risk area are integrated based on the strain rate corresponding to the coordinate mapping point to obtain exclusive rate data of the multi-level risk area; the exclusive rate data is positionally registered with a three-dimensional space coordinate system of the soft soil treatment area to obtain spatially registered rate data of the soft soil treatment area; a spatial trend of the rate of the spatial position is reconstructed based on the spatially registered rate data to obtain a settlement trend representation value of the soft soil treatment area; a spatial form of the settlement trend representation value is deduced to obtain a continuous settlement change surface of the soft soil treatment area; the continuous settlement change surface is subjected to visual rendering processing to obtain a settlement evolution cloud chart of the soft soil treatment area.

6. The soft soil treatment work settlement real-time monitoring method according to claim 1, wherein, The settlement evolution cloud chart is used to dynamically generate a monitoring parameter adjustment strategy of the soft soil treatment area through fuzzy control rules, including: risk level extraction is performed on the settlement evolution cloud chart according to the area chroma feature of the settlement evolution cloud chart to obtain a chroma distribution area of the soft soil treatment area; Based on the mapping relationship between the regional chrominance characteristics and the settlement trend in the settlement evolution cloud chart, the settlement trend intensity of the chrominance distribution region is matched to obtain a settlement trend intensity classification chart of the soft soil treatment region; The regional characteristics in the settlement trend intensity classification chart are matched and mapped with the fuzzy control rule base to obtain a preliminary parameter adjustment scheme of the soft soil treatment region; Based on the constraint condition base of the construction machinery, the preliminary parameter adjustment scheme is adapted to engineering constraints to obtain a monitoring parameter adjustment strategy of the soft soil treatment region.

7. The soft soil treatment work settlement real-time monitoring method according to claim 1, wherein The monitoring parameter adjustment strategy is output to the execution terminal of the construction machinery, including: Based on the communication protocol base of the construction machinery, the monitoring parameter adjustment strategy of the soft soil treatment region is encoded into an engineering protocol to obtain a control instruction of the soft soil treatment region; Based on the industrial communication protocol, the control instruction added with spatial positioning identification and timestamp identification is encapsulated to obtain a data packet of the identification instruction; The data packet is distributed to the corresponding construction machinery execution terminal.

8. A soft soil treatment engineering settlement real-time monitoring system, characterized in that, The system includes: A multi-dimensional strain monitoring module for real-time collection of multi-dimensional strain data streams inside the soft soil through a continuous monitoring network deployed in the soft soil treatment region, including: Receiving light signals reflected by the soft soil treatment region; Converting the light signals into electrical signals to obtain multi-dimensional strain data streams inside the soft soil; A strain field analysis module for extracting strain gradient tensors of spatial position coordinates in the spatial gradient field from the multi-dimensional strain data streams, and analyzing strain change rates of spatial positions based on timestamps in the multi-dimensional strain data streams, including: Obtaining spatial position coordinates of sensing nodes in the multi-dimensional strain data streams; Mapping the spatial position coordinates to a three-dimensional spatial grid coordinate system of the soft soil treatment region to obtain a grid-based spatial topology structure of the soft soil treatment region; Based on the grid-based spatial topology structure, calculating strain component difference values of the sensing nodes in orthogonal directions, wherein the strain component difference value calculation formula is as follows: ； wherein is the difference between the strain component values of the sensor nodes in direction, is the strain component value of the sensor node in direction, is the strain component value of the directly adjacent sensor node in direction. Combining the strain component difference values into strain gradient tensors of the soft soil treatment region according to the tensor arrangement rule; A dynamic coupling matrix generation module for cross-fusing the strain gradient tensors and the strain change rates to obtain a dynamic coupling feature matrix of the soft soil treatment region; A multi-level risk marking module for marking multi-level risk regions in the soft soil treatment region according to gradient mutation features in the dynamic coupling feature matrix; A settlement cloud chart construction module for superimposing strain change rates corresponding to the multi-level risk regions into spatial coordinates of the soft soil treatment region to obtain a settlement evolution cloud chart of the soft soil treatment region; A strategy output execution module for dynamically generating a monitoring parameter adjustment strategy of the soft soil treatment region through fuzzy control rules based on the settlement evolution cloud chart, and outputting the monitoring parameter adjustment strategy to an execution terminal of the construction machinery.

Images