A real-time monitoring and feedback system for water gate foundation treatment effect

Abstract

This invention relates to the field of monitoring technology in water conservancy and geotechnical engineering, specifically a real-time monitoring and feedback system for the treatment effect of sluice gate foundations. The system includes collecting signals such as pore water pressure, density, strength, and settlement; extracting features and change gradients segmented by time window; analyzing the evolution of the physical field in conjunction with spatial location; calculating and correlating multidimensional state response differences; identifying state abrupt changes to construct effect trajectories; matching process parameters to determine the treatment stage; and outputting control commands based on spatial evolution trends. This invention collects physical signals such as pore water pressure, density, strength, and settlement; extracts feature points and calculates response gradients; identifies the evolution trend of the physical field in conjunction with spatial location; constructs a state trend sequence and performs correlation analysis; identifies the treatment stage matching state abrupt changes and disturbance conditions; outputs control commands based on the evolutionary expansion trend; and links equipment to complete closed-loop regulation.

Description

Technical Field

[0001] This invention relates to the field of monitoring technology for water conservancy and geotechnical engineering, and in particular to a real-time monitoring and feedback system for the effect of sluice gate foundation treatment. Background Technology

[0002] With the continuous development of water conservancy infrastructure construction, the foundation stability of sluice gates, as important hydraulic structures, directly affects the safety and service life of the project. Under soft soil foundations or complex geological conditions, sluice gate foundations are prone to settlement, voids, or uneven deformation, seriously affecting structural safety. Therefore, real-time monitoring and dynamic feedback control of sluice gate foundation treatment effects have become a key technical requirement for improving the long-term service performance of sluice gates. However, existing technologies mostly focus on single-parameter monitoring or post-event risk prediction, lacking an integrated, real-time monitoring and closed-loop feedback mechanism for the entire foundation treatment process. A search revealed a method and system for predicting the foundation risk evolution of drainage sluice gates, published under publication number CN117094560B. By collecting water level, wave load, and settlement data, a steady-state function of water level settlement and a correlation function of wave load are constructed to predict the foundation settlement risk. However, the scheme mainly focuses on risk "prediction" based on historical data, rather than real-time perception of the physical state (such as pore water pressure, soil density, and reinforcement strength) during the foundation treatment process. At the same time, it does not establish a dynamic feedback control mechanism between monitoring results and foundation treatment measures, and cannot achieve the integration of "monitoring-assessment-control", making it difficult to support proactive intervention and optimization.

[0003] On the other hand, a device for monitoring settlement and displacement of sluice gate foundations, with publication number CN223607918U, was disclosed. This device, through a combination of a column, pointer, camera, and buzzer, enables real-time visual monitoring and over-limit alarms of foundation settlement and displacement. However, this device can only monitor the single indicator of macroscopic settlement and displacement, and cannot reflect the actual effectiveness of the internal mechanical state of the foundation, treatment effects (such as consolidation degree, bearing capacity improvement, etc.), or treatment processes (such as grouting, dynamic compaction, pile foundations, etc.). Furthermore, although it has an alarm function, it lacks a feedback control interface for linkage with the treatment system, making it impossible to convert monitoring information into control commands and hindering the realization of intelligent closed-loop management.

[0004] The aforementioned problems indicate that existing technologies for monitoring sluice gate foundations suffer from shortcomings such as limited monitoring dimensions, lack of quantitative evaluation of treatment effects, and absence of real-time feedback and control mechanisms. These deficiencies make it difficult to meet the intelligent management requirements of modern water conservancy projects for "knowable, assessable, and controllable" foundation treatment processes. Therefore, this invention provides a real-time monitoring and feedback system for sluice gate foundation treatment effects. This system aims to integrate multi-source sensor data, dynamically evaluate the effectiveness of foundation treatment, and establish a monitoring-feedback-control closed loop to achieve precise and intelligent management of the sluice gate foundation treatment process. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of existing technologies and propose a real-time monitoring and feedback system for the treatment effect of sluice gate foundations.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a real-time monitoring and feedback system for the foundation treatment effect of a sluice gate, the system comprising:

[0007] The multi-source sensor fusion module collects multi-dimensional physical signals during the sluice gate foundation treatment process, processes them in segments according to a preset time window, identifies key response feature points in each segment of the signal, locates the spatial distribution area of ​​the physical field based on the gradient of physical quantity changes and spatial location information between adjacent sampling points, extracts the spatial evolution rate, and constructs the pore water pressure dissipation trend, soil density growth trend, solidified body strength development trend, and settlement deformation accumulation trend.

[0008] The state coupling assessment module calculates the dynamic response rate difference for each time window based on the pore water pressure dissipation trend, soil compaction growth trend, solidified body strength development trend, and settlement deformation accumulation trend. After arranging them in chronological order, it performs cross-physical field correlation analysis to obtain multi-dimensional state coupling assessment results.

[0009] The treatment effect quantification module, based on the multi-dimensional state coupling evaluation results, identifies the time response intervals of state abrupt change points in continuous time windows, classifies the time response intervals, and performs offset amplitude accumulation by combining the state evolution intensity difference between adjacent time windows to construct the foundation treatment effect quantification trajectory.

[0010] The process stage discrimination module identifies continuous offset time windows based on the quantitative trajectory of the foundation treatment effect, and synchronously matches them with the corresponding foundation treatment process parameters. It then filters out time windows that meet the preset process disturbance conditions and generates processing stage identification content.

[0011] The control instruction generation module performs a spatial cross-comparison based on the processing stage identifier content and the expansion trend of the spatial evolution region of the ground physical field to determine whether it meets the preset control trigger conditions, and outputs the control instructions of the ground processing equipment according to the state evolution level.

[0012] As a further aspect of the present invention, the trends of pore water pressure dissipation, soil density growth, solidified body strength development, and settlement deformation accumulation include a sequence of physical quantity characteristic points, gradient differences between adjacent sampling points, spatial distribution area boundaries, and physical field evolution rate indices. The multidimensional state coupling evaluation results include a dynamic response rate difference matrix, cross-physical field correlation coefficients, and state coupling difference values. The quantitative trajectory of the foundation treatment effect includes a state mutation point time index, response interval category labels, and offset amplitude superposition sequences. The treatment stage identification content includes continuous offset time window identifiers, corresponding time periods for process disturbances, and treatment effect abnormal stage numbers. The control instructions include spatial cross-matching results, state evolution level classifications, and equipment control instruction types.

[0013] As a further aspect of the present invention, the identification of state mutation points in a continuous time window refers to locating the time position where the rate of change exceeds a set mutation threshold by monitoring the change gradient of the multidimensional state coupling evaluation results in the time series.

[0014] As a further aspect of the present invention, the selection of time windows that meet the preset process disturbance conditions refers to extracting time windows that exceed the process disturbance reference value by performing difference analysis on the sequence of foundation treatment process parameters within a continuous offset time window.

[0015] As a further aspect of the present invention, the multi-source sensing fusion module includes:

[0016] The pore water pressure extraction submodule is deployed in an array of pore water pressure sensors at different depths of the foundation to acquire pore water pressure time series data during the processing. It is then segmented according to a preset time window, monitors the dissipation inflection point in each pressure sequence, extracts the pressure values ​​of adjacent sampling points and calculates the difference, and generates a pore water pressure gradient sequence.

[0017] The compaction and strength extraction submodule integrates a soil compaction sensor and a solidified body strength sensor to acquire compaction and strength data within the corresponding time window. It calls the spatial coordinate information of the sensor to locate the boundary of the effective sensing area, extracts the average value of physical quantities within the area, calculates the rate of change of physical quantities between adjacent time windows, and generates a joint evolution rate sequence of compaction and strength.

[0018] The settlement deformation extraction submodule is configured with a high-precision displacement sensor network to acquire settlement displacement data of the foundation surface and deep layers. Combined with timestamps and spatial coordinates, a three-dimensional settlement field model is constructed, the region with the maximum settlement gradient is extracted, and the settlement increment rate of adjacent time windows is calculated to generate a settlement deformation rate sequence.

[0019] The multi-field trend construction submodule calls the time index and spatial coordinates of the pore water pressure gradient sequence, the density-strength joint evolution rate sequence, and the settlement deformation rate sequence, matches the multidimensional data within the same time window and spatial region, performs spatiotemporal normalization processing on each physical quantity, establishes multi-physics field change curves under a unified spatiotemporal benchmark, and obtains the pore water pressure dissipation trend, soil density growth trend, solidified body strength development trend, and settlement deformation accumulation trend.

[0020] As a further aspect of the present invention, the state coupling evaluation module includes:

[0021] The dynamic rate construction submodule extracts the physical quantity values ​​of three consecutive sampling points in each segment according to the multi-physics field change curve, and subtracts the difference between the last two points and the first point in pairs to obtain the continuous dynamic response rate difference sequence of each segment of physical quantity, and obtains the multi-channel dynamic rate difference matrix.

[0022] The cross-field correlation analysis submodule calls the physical field data in the multi-channel dynamic rate difference matrix, pairs the rate differences in the same time window according to the time index, constructs a multi-dimensional feature vector set of the paired data, and performs clustering and aggregation based on the Euclidean distance or correlation coefficient between each vector to obtain state coupling data clusters.

[0023] The coupling evaluation generation submodule extracts the contribution weight of each physical field to the overall state according to the time series of each cluster in the state coupling data cluster, and plots the multi-field coupling trend curve in time order. It calculates the coupling difference value sequence between groups through the synergy and deviation of the trend lines, and establishes the multi-dimensional state coupling evaluation result.

[0024] As a further aspect of the present invention, the processing effect quantification module includes:

[0025] The mutation point identification submodule monitors the difference in coupling values ​​between adjacent time windows based on the continuous time series in the multidimensional state coupling evaluation results, and marks the time positions where the rate of change is greater than a set mutation threshold according to the first or second derivative of the change gradient, extracts the time windows between consecutive markers and aggregates the boundary index to obtain the state mutation point sequence.

[0026] The response interval classification submodule calculates the time response interval value between two points based on the index of each group of adjacent mutation points in the state mutation point sequence, and classifies all mutation groups into the corresponding category interval according to the interval length, counts the number of mutation segments and the average time span in each category, and obtains the response interval classification result set.

[0027] The quantization trajectory construction submodule calls the time index of each mutation group in the response interval classification result set, calculates the difference of the multidimensional state coupling value of the corresponding time window, performs amplitude superposition of the difference between adjacent segments in time order, summarizes the total difference of the classification and the trend line index, and establishes the quantization trajectory of the foundation treatment effect.

[0028] As a further aspect of the present invention, the process stage discrimination module includes:

[0029] The continuous offset detection submodule quantizes the continuous offset segments in the trajectory based on the foundation treatment effect, extracts the amplitude change gradient between adjacent segments by time index, judges the continuity according to the set offset duration threshold, aggregates and marks the index segments that meet the continuity condition, and obtains the continuous offset time window set.

[0030] The process disturbance correlation submodule calls the index of each time period in the continuous offset time window set, and synchronously obtains the operating parameter sequence of the ground treatment equipment within the corresponding time window, calculates the difference between any two adjacent sampling points in each parameter sequence, and determines whether it exceeds the process disturbance reference value, and obtains the disturbance correlation time window sequence.

[0031] The stage identifier generation submodule filters out time windows that appear in the quantitative trajectory of the foundation treatment effect and are accompanied by process disturbances, based on the time window positions in the disturbance-related time window sequence. The filtered time windows are then structured and represented according to the start and end indices to establish the processing stage identifier content.

[0032] As a further aspect of the present invention, the control instruction generation module includes:

[0033] The physical field expansion judgment submodule, based on the time window index in the processing stage identifier content, calls the ground-based multiphysics spatial distribution model of the corresponding time frame, extracts the boundary contour of the state evolution region in each frame, calculates the boundary expansion direction and area increment of the region in adjacent frames, aggregates and judges the region expansion behavior in continuous time windows, and obtains the physical field expansion trend dataset.

[0034] The rule cross-filtering submodule performs three-dimensional spatial coordinate cross-judgment based on the spatial positioning information of the corresponding time window in the physical field expansion trend dataset and the processing stage identifier content, extracts the spatial volume range of the cross-region, compares it with the spatial overlap threshold in the preset control rules, filters the data segments that meet the conditions, and obtains the rule matching cross-region set.

[0035] The control instruction generation submodule calls the rules to match the state evolution level values ​​of the cross-regions, and retrieves the corresponding control instructions from the equipment control parameter library according to the rules based on the control strategy mapping relationship corresponding to the level values. It integrates the operation object, action type and execution parameters of the instructions to generate control instructions for the foundation treatment equipment.

[0036] As a further aspect of the present invention, the spatial overlap threshold refers to determining whether the degree of overlap between the physical field expansion region and the processing stage identification region in three-dimensional spatial coordinates reaches the cross-volume standard for triggering control commands.

[0037] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0038] In this invention, multi-dimensional physical signals such as pore water pressure, soil density, reinforcement strength, and settlement deformation are collected. Key feature points are extracted in segments and response gradients are calculated. Combined with the spatial location of the foundation, the evolution region and rate of change of the physical field are identified, and a spatiotemporal trend sequence of each physical quantity is established. The dynamic difference is calculated in chronological order and cross-physical quantity correlation is performed to obtain the state evolution trend and coupled change trajectory. Further, the processing stage is identified based on the state mutation and parameter disturbance conditions. After comparing the physical field expansion trend with the triggering conditions, control commands are output to adjust the processing equipment's response to the state change process, thus completing the linkage closed loop of information perception, state identification, and intervention command output. Attached Figure Description

[0039] Figure 1 This is a system flowchart of the present invention;

[0040] Figure 2 This is a flowchart illustrating the acquisition process of the multi-source sensing fusion module of the present invention.

[0041] Figure 3 This is a flowchart illustrating the acquisition process of the state coupling evaluation module of the present invention.

[0042] Figure 4 This is a flowchart illustrating the acquisition process of the quantification module for the processing effect of this invention.

[0043] Figure 5 This is a flowchart illustrating the process stage discrimination module of the present invention.

[0044] Figure 6 This is a flowchart illustrating the acquisition process of the control command generation module of the present invention. Detailed Implementation

[0045] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0046] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0047] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning.

[0048] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0049] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0050] Please see Figure 1 This invention provides a technical solution: a real-time monitoring and feedback system for the effect of sluice gate foundation treatment, the system comprising:

[0051] The multi-source sensor fusion module collects multi-dimensional physical signals of pore water pressure, soil density, reinforced body strength, and settlement deformation during the foundation treatment of the sluice gate. It performs segmented processing on each sensor signal according to a preset time window, identifies key response feature points in each signal segment, extracts the gradient of physical quantity changes between adjacent sampling points, synchronously acquires the spatial location information of the foundation corresponding to the time window, locates the spatial distribution area of ​​each physical field, and extracts the spatial evolution rate of the physical field within adjacent time windows. It constructs the trends of pore water pressure dissipation, soil density growth, reinforced body strength development, and settlement deformation accumulation.

[0052] The state coupling assessment module calculates the dynamic response rate difference of each physical quantity in each time window based on the pore water pressure dissipation trend, soil compaction growth trend, solidified body strength development trend and settlement deformation accumulation trend. The dynamic response rate difference is arranged in time order and cross-physical field correlation analysis is performed to obtain multi-dimensional state coupling assessment results.

[0053] The treatment effect quantification module identifies state abrupt change points in continuous time windows based on multi-dimensional state coupling evaluation results, calculates the time response interval between abrupt change points and classifies them according to response characteristics, and performs offset amplitude accumulation by combining the state evolution intensity difference between adjacent time windows to construct the foundation treatment effect quantification trajectory.

[0054] The process stage discrimination module identifies continuous offset time windows based on the quantitative trajectory of foundation treatment effect, performs synchronous matching with the foundation treatment process parameters within the continuous offset time window, filters time windows that meet preset process disturbance conditions, and generates treatment stage identification content.

[0055] The control instruction generation module performs spatial cross-comparison based on the processing stage identifier content and the expansion trend of the spatial evolution region of the ground physical field, determines whether the cross region meets the preset control trigger conditions, and outputs the control instructions of the ground processing equipment according to the state evolution level of the cross region.

[0056] The trends of pore water pressure dissipation, soil density growth, solidified body strength development, and cumulative settlement deformation include physical quantity characteristic point sequences, gradient differences between adjacent sampling points, spatial distribution area boundaries, and physical field evolution rate indices. The multidimensional state coupling assessment results include dynamic response rate difference matrix, cross-physical field correlation coefficients, and state coupling difference values. The quantitative trajectory of foundation treatment effect includes state change point time index, response interval category label, and offset amplitude superposition sequence. The treatment stage identification content includes continuous offset time window identifier, corresponding time period of process disturbance, and abnormal treatment effect stage number. The control instructions include spatial cross-matching results, state evolution level classification, and equipment control instruction type.

[0057] Please see Figure 2 The multi-source sensor fusion module includes:

[0058] The pore water pressure extraction submodule is deployed in an array of pore water pressure sensors at different depths of the foundation to acquire pore water pressure time series data during the processing. It is then segmented according to a preset time window, monitors the dissipation inflection point in each pressure sequence, extracts the pressure values ​​of adjacent sampling points and calculates the difference, and generates a pore water pressure gradient sequence.

[0059] First, the pore water pressure extraction submodule is activated to perform data acquisition and processing on pore water pressure sensor arrays deployed at different depths in the foundation. The sensor arrays are vertically positioned at depths of 5.0 meters, 10.0 meters, and 15.0 meters, with a sampling frequency of 1 Hz. The pore water pressure time-series data for the entire processing process is acquired, and the continuous data stream is segmented according to a preset time window of 600 seconds (i.e., 10 minutes). Within each independent time window, the second derivative of the pressure sequence is calculated, and the dissipation inflection point in that segment of the pressure sequence is monitored. The specific criterion is that the second derivative crosses zero and the corresponding first derivative is negative. For adjacent sampling points, such as time points... and The pressure values ​​of the two points are extracted separately, and the pressure difference between the two points is calculated by subtraction. Then, the pressure difference is divided by the spacing value of the sensors in the vertical direction (5.0 meters) to generate a pore water pressure gradient sequence.

[0060] The compaction and strength extraction submodule integrates a soil compaction sensor and a solidified body strength sensor to acquire compaction and strength data within the corresponding time window. It calls the spatial coordinate information of the sensor to locate the boundary of the effective sensing area, extracts the average value of physical quantities within the area, calculates the rate of change of physical quantities between adjacent time windows, and generates a joint evolution rate sequence of compaction and strength.

[0061] This submodule integrates a static cone penetration test (CPT) sensor and a solidified soil strength sensor. Within the same time window, it acquires cone tip resistance data of the soil as a measure of compaction, and sidewall skin friction data as a measure of strength. It uses the sensor's preset spatial coordinates, defining a sphere with a radius of 1.5 meters centered on the sensor probe as the effective sensing area boundary. It iterates through all physical quantity sampling data within this area, using an arithmetic mean method to extract the average compaction and strength values. For example, within the current time window... Extract the mean density value within. In the next time window Mean extraction The mean values ​​of adjacent time windows are differentially calculated, and the difference is divided by the time window length of 600 seconds to generate a density-intensity joint evolution rate sequence.

[0062] The settlement deformation extraction submodule is configured with a high-precision displacement sensor network to acquire settlement displacement data of the foundation surface and deep layers. Combined with timestamps and spatial coordinates, a three-dimensional settlement field model is constructed, the region with the maximum settlement gradient is extracted, and the settlement increment rate of adjacent time windows is calculated to generate a settlement deformation rate sequence.

[0063] Based on a high-precision fiber Bragg grating displacement sensor network, real-time settlement displacement data of surface and deep stratified settlement markers of the foundation are acquired. A three-dimensional settlement field model is constructed by combining the data timestamps and three-dimensional spatial coordinates. In the model, the region with the largest settlement gradient is identified by calculating the partial derivative of the settlement with respect to the spatial coordinates. Specifically, this involves calculating... direction and The square root of the sum of squares of the directional partial derivatives is used to lock the region where the calculated result exceeds the set gradient benchmark value of 0.05 as the maximum deformation region. Within this region, the settlement increment between adjacent time windows is calculated, and this increment is divided by the time interval to generate a settlement deformation rate sequence.

[0064] The multi-field trend construction submodule calls the time index and spatial coordinates of the pore water pressure gradient sequence, the density-strength joint evolution rate sequence, and the settlement deformation rate sequence, matches multidimensional data within the same time window and spatial region, performs spatiotemporal normalization processing on each physical quantity, establishes multi-physics field change curves under a unified spatiotemporal benchmark, and obtains the pore water pressure dissipation trend, soil density growth trend, solidified body strength development trend, and settlement deformation accumulation trend.

[0065] The generated pore water pressure gradient sequence, density-strength joint evolution rate sequence, and settlement deformation rate sequence are invoked. Using time index and spatial coordinates as primary keys, precise matching is performed within the same time window (e.g., the [number]th time window). (Single window) and multidimensional data within the same spatial region (e.g., at a depth of 10 meters). To address the issue of inconsistent dimensions among physical quantities, spatiotemporal normalization is performed on each sequence. The normalization process employs a range standardization method: subtracting the minimum value of a physical quantity within a historical period from its current value, and then dividing the difference by the difference between the historical maximum and minimum values, maps all physical quantities to a dimensionless interval of 0 to 1. Through this operation, multiphysics field variation curves under a unified spatiotemporal benchmark are established, ultimately yielding trends in pore water pressure dissipation, soil density growth, solidified body strength development, and cumulative settlement deformation.

[0066] Table 1: Normalized Multiphysics Data of Monitoring Points

[0067]

[0068] As shown in Table 1, the above processing procedure transforms multi-source sensor data at a depth of 10 meters across three consecutive time windows into standardized trend values. The data shows that as time progresses (from... to The normalized value of the pore water pressure gradient shows a decreasing trend, the normalized value of the compaction rate shows an increasing trend, and the normalized value of the settlement rate gradually decreases. This set of data accurately reflects the consolidation process of the foundation under vacuum preloading.

[0069] Please see Figure 3 The state coupling evaluation module includes:

[0070] The dynamic rate construction submodule, based on the multiphysics field change curve, extracts the physical quantity values ​​of three consecutive sampling points in each segment according to the time window, subtracts the difference between the last two points and the first point in pairs to obtain the continuous dynamic response rate difference sequence of each segment of physical quantity, and obtains the multi-channel dynamic rate difference matrix.

[0071] The work is carried out based on the multiphysics field variation curves output by the multiphysics trend construction submodule. The physical quantity values ​​of three consecutive sampling points within each time window are extracted sequentially. Taking the pore water pressure gradient normalized data in Table 1 above as an example, the time window is extracted... The corresponding values ​​are 0.85, 0.82, and 0.79. Perform pairwise subtraction: subtract the first point's 0.85 from the second point's 0.82 to obtain the first-stage difference of -0.03; subtract the second point's 0.82 from the third point's 0.79 to obtain the second-stage difference of -0.03. Arrange these differences sequentially to obtain a continuous dynamic response rate difference sequence for this physical quantity. Perform the same operation on the density and settlement deformation data, and finally combine them to obtain a multi-channel dynamic rate difference matrix.

[0072] The cross-field correlation analysis submodule calls the physical field data in the multi-channel dynamic rate difference matrix, pairs the rate differences in the same time window according to the time index, constructs a multi-dimensional feature vector set of the paired data, and performs clustering and aggregation based on the Euclidean distance or correlation coefficient between the vectors to obtain state-coupled data clusters.

[0073] The system retrieves the physical field data from the multi-channel dynamic rate difference matrix. Rate differences within the same time window are paired according to their time index, and the paired data are used to construct a multi-dimensional feature vector set. For example, for a time window... to The process of change is used to construct vectors containing pore pressure difference (-0.03), density difference (0.06), and settlement difference (-0.03). Clustering is performed based on the Euclidean distance between each vector. During this process, a cluster center distance threshold of 0.05 is set. The threshold setting process is as follows: Select feature vector datasets from the stable consolidation stage of historical similar foundation treatment projects, calculate the average distance and standard deviation between all vector points, and set the threshold as the absolute value of the average distance minus twice the standard deviation. Actual measurements show this value to be 0.05. Calculate the Euclidean distance between the current feature vector and each existing cluster center. If the distance is less than 0.05, the vector is assigned to the corresponding state-coupled data cluster.

[0074] The coupling evaluation generation submodule extracts the contribution weight of each physical field to the overall state according to the time series of each cluster in the state coupling data cluster, and plots the multi-field coupling trend curve in time order. It calculates the coupling difference value sequence between groups through the synergy and deviation of the trend lines, and establishes the multi-dimensional state coupling evaluation result.

[0075] Based on the time series of each cluster in the state-coupled data cluster, weight allocation and coupling value calculation are performed. The entropy method is used to extract the contribution weight of each physical field to the overall state. First, a feature matrix is ​​constructed, and the entropy value of each physical quantity index is calculated. Then, the difference coefficient is calculated based on the entropy value; the larger the difference coefficient, the higher the corresponding weight. In this embodiment, the calculated weights are 0.4 for pore water pressure, 0.35 for density, and 0.25 for settlement deformation. Multi-field coupling trend curves are plotted in chronological order using a linear weighting formula. Subsequently, the coupling difference value is determined by calculating the synergy of the trend lines (characterized by Pearson correlation coefficient) and the deviation (characterized by root mean square error). A balance coefficient is set. It is 0.6. The coefficient of variation is 0.4. The calculation formula is: the difference value equals 0.6 multiplied by (1 minus the correlation coefficient) and the sum of 0.4 multiplied by the root mean square error. Substituting the current data into the calculation, if the correlation coefficient is 0.92 and the root mean square error is 0.04, the calculated inter-group coupling difference value is 0.064. This result is less than the preset abnormal decoupling benchmark value of 0.2, establishing a multi-dimensional state coupling assessment result, indicating that the changes in the various physical fields of the current foundation are highly coordinated and in a normal consolidation state.

[0076] Please see Figure 4 The processing effect quantification module includes:

[0077] The mutation point identification submodule monitors the difference in coupling values ​​between adjacent time windows based on the continuous time series in the multidimensional state coupling evaluation results, and marks the time positions where the rate of change is greater than the set mutation threshold according to the first or second derivative of the change gradient. It extracts the time windows between consecutive markers and aggregates the boundary index to obtain the state mutation point sequence.

[0078] Based on the continuous time series from the multidimensional state coupling assessment results, i.e., the coupling difference value sequence calculated above, the differences in coupling values ​​between adjacent time windows are monitored. The first derivative of the coupling difference value sequence is calculated to reflect its change gradient. In this step, the mutation threshold is set to 0.05. The specific basis for setting this threshold is: collecting coupling difference value data for 24 consecutive hours after the foundation treatment enters the stabilization period, calculating the standard deviation of the first derivative of this dataset, and the result is 0.015. Following statistical principles, the mutation threshold is set to three times the standard deviation (approximately 0.05) to filter out 99.7% of random background noise. Traversing the entire sequence, the time positions where the absolute value of the change gradient is greater than 0.05 are marked, the time windows between consecutive markers are extracted, and the boundary indices are aggregated to obtain the state mutation point sequence.

[0079] The response interval classification submodule calculates the time response interval between two points based on the index of each group of adjacent mutation points in the state mutation point sequence, and classifies all mutation groups into the corresponding category interval according to the interval length. It also counts the number of mutation segments and the average time span in each category to obtain the response interval classification result set.

[0080] Based on the sequence of state mutation points, extract the index value of each group of adjacent mutation points. For example, if the previous mutation point occurs in the 20th time window and the next mutation point occurs in the 35th time window, calculate the time response interval between the two points, i.e., 15 time windows, corresponding to 150 minutes of actual time. Assign all mutation groups to corresponding category intervals according to the interval length: intervals of 0 to 30 minutes are assigned to the short-frequency response interval, intervals of 30 to 120 minutes to the medium-frequency response interval, and intervals greater than 120 minutes to the long-frequency response interval. Count the number of mutation segments and the average time span in each category to obtain the response interval classification result set.

[0081] The quantization trajectory construction submodule calls the time index of each mutation group in the response interval classification result set, calculates the difference of the multidimensional state coupling value of the corresponding time window, performs amplitude superposition of the difference between adjacent segments in time order, summarizes the total difference of the classification and the trend line index, and establishes the quantization trajectory of the foundation treatment effect.

[0082] The time index of each mutation group in the response interval classification result set is invoked. The difference between the multidimensional state coupling values ​​of the corresponding time window is calculated, i.e., the coupling value at the end of the segment is subtracted from the coupling value at the beginning of the segment. The differences between adjacent segments are then summed in chronological order. Taking the long-frequency response interval as an example, the total difference of all mutation segments within this interval is summarized. Assuming there are 5 mutation segments in the long-frequency interval, with differences of 0.5, 0.6, 0.7, 0.65, and 0.7 respectively, the total summation is calculated to be 3.15. Combining this total with the time trend line index, a quantitative trajectory of the foundation treatment effect is established. The monotonically increasing characteristic of this trajectory reflects the cumulative effect of the foundation treatment. The large proportion of the total difference in the long-frequency interval indicates that the treatment effect is mainly contributed by long-term stable consolidation rather than short-term disturbances.

[0083] Please see Figure 5 The process stage discrimination module includes:

[0084] The continuous offset detection submodule quantizes continuous offset segments in the trajectory based on the foundation treatment effect, extracts the amplitude change gradient between adjacent segments by time index, judges the continuity based on the set offset duration threshold, aggregates and marks index segments that meet the continuity condition, and obtains a continuous offset time window set.

[0085] The analysis focuses on continuous offset segments in the quantized trajectory based on the foundation treatment effect. The amplitude change gradient between adjacent segments is extracted by time index. An offset duration threshold of 6 is set, corresponding to 6 consecutive time windows (i.e., 1 hour). The continuity of the trajectory gradient is judged: if the trajectory gradient maintains the same sign (both positive growth and both negative decay) within 6 consecutive time windows, and the absolute value of the gradient is greater than 0.01, then the continuity condition is met. Index segments that meet the condition are aggregated and marked, for example, the aggregated index interval [100, 106], to obtain the set of continuous offset time windows.

[0086] The process disturbance correlation submodule calls the index of each time period in the continuous offset time window set, and synchronously obtains the operating parameter sequence of the ground treatment equipment within the corresponding time window. It calculates the difference between any two adjacent sampling points in each parameter sequence and determines whether it exceeds the process disturbance reference value, and obtains the disturbance correlation time window sequence.

[0087] The system retrieves the index for each time period within a continuous offset time window and simultaneously obtains the sequence of operating parameters for the ground treatment equipment within the corresponding time window via the industrial bus. It also obtains the vacuum pressure value sequence of the vacuum pump. The system calculates the difference between any two adjacent sampling points in this parameter sequence. A process disturbance baseline value of 2.0 kPa is set. This baseline value is set according to the equipment's factory technical specifications, with a rated working pressure of 80 kPa and an allowable fluctuation range of ±2.5%, i.e., 2.0 kPa. The system iterates through the parameter difference sequence to determine if any absolute difference exceeds 2.0 kPa. Within the index interval [100, 106], a sudden drop in vacuum from 80 kPa to 75 kPa is detected, with a difference of 5 kPa, exceeding the process disturbance baseline value. Therefore, this segment is locked, and the disturbance-related time window sequence is obtained.

[0088] The stage identifier generation submodule filters out time windows that appear in the quantitative trajectory of foundation treatment effect and are accompanied by process disturbances, based on the time window position in the disturbance-related time window sequence. The filtered time windows are then structured according to the start and end indices to establish the treatment stage identifier content.

[0089] Based on the time window position in the disturbance-related time window sequence, time windows exhibiting continuous offsets in the quantitative trajectory of foundation treatment effects, accompanied by process disturbances, are selected. Logical judgment is executed: if a time window shows both significant decay of the quantitative trajectory (e.g., continuously negative gradients) and abnormal fluctuations in equipment parameters (e.g., vacuum level drops), the window is marked as "Effectiveness Decline Stage Caused by Equipment Failure." Conversely, if the quantitative trajectory shows significant changes (e.g., continuous and drastic gradient fluctuations) but equipment parameters operate stably (fluctuations within 2.0 kPa), it is marked as "Soil Structure Abrupt Change Stage." The selected and judged time windows are structurally represented by start and end indices, establishing a processing stage identifier that includes stage type, start and end times, and associated parameters.

[0090] Please see Figure 6 The control instruction generation module includes:

[0091] The physical field expansion judgment submodule calls the ground-based multiphysics spatial distribution model of the corresponding time frame based on the time window index in the processing stage identifier, extracts the boundary contour of the state evolution region in each frame, calculates the boundary expansion direction and area increment of the region in adjacent frames, aggregates and judges the regional expansion behavior in continuous time windows, and obtains the physical field expansion trend dataset.

[0092] Based on the time window index in the processing stage identifier, the ground-based multiphysics spatial distribution model for the corresponding time frame is invoked. The boundary contours of the state evolution region in each frame are extracted; for example, contour lines are extracted for regions where pore water pressure decreases by more than 10%. The boundary expansion direction and area increment of this region in adjacent frames are calculated. Aggregate and judge the regional expansion behavior within consecutive time windows. If the area increment is positive for five consecutive time windows, the physical field is determined to be in an expansion state, and a physical field expansion trend dataset is obtained.

[0093] The rule cross-filtering submodule performs three-dimensional spatial coordinate cross-judgment based on the spatial positioning information of the corresponding time window in the physical field expansion trend dataset and the processing stage identifier content, extracts the spatial volume range of the cross-region, compares it with the spatial overlap threshold in the preset control rules, filters the data segments that meet the conditions, and obtains the rule matching cross-region set.

[0094] Based on the spatial positioning information of the corresponding time window in the physical field expansion trend dataset and the processing stage identifier, a three-dimensional spatial coordinate cross-judgment is performed. The spatial volume range of the cross-region is extracted. The spatial overlap threshold in the preset control rules is 50 cubic meters. This threshold is set based on the effective influence volume of a single reinforcement unit (such as a drainage board), which is calculated as the square of the drainage board spacing (approximately 1.0 meter) multiplied by half the depth (approximately 10 meters), rounded down to set as the safety threshold. The calculated cross-volume is compared with 50 cubic meters, and data segments with volumes greater than 50 cubic meters are selected to obtain the rule-matching cross-region set.

[0095] The control instruction generation submodule calls the rule to match the state evolution level value of the cross area, and retrieves the corresponding control instruction from the equipment control parameter library according to the control strategy mapping relationship corresponding to the level value. It integrates the operation object, action type and execution parameters of the instruction to generate the control instruction of the foundation treatment equipment.

[0096] The rules are invoked to match the state evolution level values ​​of the cross-regions. These level values ​​are based on the numerical ranges of the aforementioned normalized physical quantities: 0 to 0.25 is Level I (ineffective), 0.25 to 0.5 is Level II (slow-acting), 0.5 to 0.75 is Level III (normal consolidation), and 0.75 to 1.0 is Level IV (overspeed / clogging risk). Based on the mapping relationship of the control strategies corresponding to the level values, the corresponding control instructions are retrieved from the equipment control parameter library. In this embodiment, the state evolution level value of the cross-region is identified as Level II (slow-acting), and it is in the "normal pressurization stage." The integrated instructions are: the operation object is selected as "vacuum pump group A," the action type is selected as "frequency conversion pressurization," and the execution parameter is set to "increase power to 90%." Finally, a specific ground treatment equipment control instruction is generated: "vacuum pump group A - execute frequency conversion pressurization - target power 90%," and executed through the control interface.

[0097] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A real-time monitoring and feedback system for the foundation treatment effect of a sluice gate, characterized in that, The system includes: The multi-source sensor fusion module collects multi-dimensional physical signals during the sluice gate foundation treatment process, processes them in segments according to a preset time window, identifies key response feature points in each segment of the signal, locates the spatial distribution area of ​​the physical field based on the gradient of physical quantity changes and spatial location information between adjacent sampling points, extracts the spatial evolution rate, and constructs the pore water pressure dissipation trend, soil density growth trend, solidified body strength development trend, and settlement deformation accumulation trend. The state coupling assessment module calculates the dynamic response rate difference for each time window based on the pore water pressure dissipation trend, soil compaction growth trend, solidified body strength development trend, and settlement deformation accumulation trend. After arranging them in chronological order, it performs cross-physical field correlation analysis to obtain multi-dimensional state coupling assessment results. The treatment effect quantification module, based on the multi-dimensional state coupling evaluation results, identifies the time response intervals of state abrupt change points in continuous time windows, classifies the time response intervals, and performs offset amplitude accumulation by combining the state evolution intensity difference between adjacent time windows to construct the foundation treatment effect quantification trajectory. The process stage discrimination module identifies continuous offset time windows based on the quantitative trajectory of the foundation treatment effect, and synchronously matches them with the corresponding foundation treatment process parameters. It then filters out time windows that meet the preset process disturbance conditions and generates processing stage identification content. The control instruction generation module performs a spatial cross-comparison based on the processing stage identifier content and the expansion trend of the spatial evolution region of the ground physical field to determine whether it meets the preset control trigger conditions, and outputs the control instructions of the ground processing equipment according to the state evolution level.

2. The real-time monitoring and feedback system for the foundation treatment effect of a sluice gate according to claim 1, characterized in that: The trends of pore water pressure dissipation, soil density growth, solidified body strength development, and cumulative settlement deformation include a sequence of physical quantity characteristic points, gradient differences between adjacent sampling points, spatial distribution area boundaries, and physical field evolution rate indices. The multidimensional state coupling evaluation results include a dynamic response rate difference matrix, cross-physical field correlation coefficients, and state coupling difference values. The quantitative trajectory of foundation treatment effect includes a state change point time index, response interval category labels, and offset amplitude superposition sequence. The treatment stage identification content includes continuous offset time window identifiers, corresponding time periods for process disturbances, and treatment effect abnormal stage numbers. The control instructions include spatial cross-matching results, state evolution level classification, and equipment control instruction types.

3. The real-time monitoring and feedback system for the foundation treatment effect of a sluice gate according to claim 1, characterized in that, The identification of state mutation points in a continuous time window refers to locating the time position where the rate of change exceeds a set mutation threshold by monitoring the change gradient of the multidimensional state coupling evaluation results in the time series.

4. The real-time monitoring and feedback system for the foundation treatment effect of a sluice gate according to claim 1, characterized in that, The time window for screening that meets the preset process disturbance conditions refers to the time window that exceeds the process disturbance benchmark value by performing difference analysis on the sequence of foundation treatment process parameters within a continuous offset time window.

5. The real-time monitoring and feedback system for the foundation treatment effect of a sluice gate according to claim 1, characterized in that, The multi-source sensing fusion module includes: The pore water pressure extraction submodule is deployed in an array of pore water pressure sensors at different depths of the foundation to acquire pore water pressure time series data during the processing. It is then segmented according to a preset time window, monitors the dissipation inflection point in each pressure sequence, extracts the pressure values ​​of adjacent sampling points and calculates the difference, and generates a pore water pressure gradient sequence. The compaction and strength extraction submodule integrates a soil compaction sensor and a solidified body strength sensor to acquire compaction and strength data within the corresponding time window. It calls the spatial coordinate information of the sensor to locate the boundary of the effective sensing area, extracts the average value of physical quantities within the area, calculates the rate of change of physical quantities between adjacent time windows, and generates a joint evolution rate sequence of compaction and strength. The settlement deformation extraction submodule is configured with a high-precision displacement sensor network to acquire settlement displacement data of the foundation surface and deep layers. Combined with timestamps and spatial coordinates, a three-dimensional settlement field model is constructed, the region with the maximum settlement gradient is extracted, and the settlement increment rate of adjacent time windows is calculated to generate a settlement deformation rate sequence. The multi-field trend construction submodule calls the time index and spatial coordinates of the pore water pressure gradient sequence, the density-strength joint evolution rate sequence, and the settlement deformation rate sequence, matches the multidimensional data within the same time window and spatial region, performs spatiotemporal normalization processing on each physical quantity, establishes multi-physics field change curves under a unified spatiotemporal benchmark, and obtains the pore water pressure dissipation trend, soil density growth trend, solidified body strength development trend, and settlement deformation accumulation trend.

6. The real-time monitoring and feedback system for the foundation treatment effect of a sluice gate according to claim 1, characterized in that, The state coupling evaluation module includes: The dynamic rate construction submodule extracts the physical quantity values ​​of three consecutive sampling points in each segment according to the multi-physics field change curve, and subtracts the difference between the last two points and the first point in pairs to obtain the continuous dynamic response rate difference sequence of each segment of physical quantity, and obtains the multi-channel dynamic rate difference matrix. The cross-field correlation analysis submodule calls the physical field data in the multi-channel dynamic rate difference matrix, pairs the rate differences in the same time window according to the time index, constructs a multi-dimensional feature vector set of the paired data, and performs clustering and aggregation based on the Euclidean distance or correlation coefficient between each vector to obtain state coupling data clusters. The coupling evaluation generation submodule extracts the contribution weight of each physical field to the overall state according to the time series of each cluster in the state coupling data cluster, and plots the multi-field coupling trend curve in time order. It calculates the coupling difference value sequence between groups through the synergy and deviation of the trend lines, and establishes the multi-dimensional state coupling evaluation result.

7. The real-time monitoring and feedback system for the foundation treatment effect of a sluice gate according to claim 1, characterized in that, The processing effect quantification module includes: The mutation point identification submodule monitors the difference in coupling values ​​between adjacent time windows based on the continuous time series in the multidimensional state coupling evaluation results, and marks the time positions where the rate of change is greater than a set mutation threshold according to the first or second derivative of the change gradient, extracts the time windows between consecutive markers and aggregates the boundary index to obtain the state mutation point sequence. The response interval classification submodule calculates the time response interval value between two points based on the index of each group of adjacent mutation points in the state mutation point sequence, and classifies all mutation groups into the corresponding category interval according to the interval length, counts the number of mutation segments and the average time span in each category, and obtains the response interval classification result set. The quantization trajectory construction submodule calls the time index of each mutation group in the response interval classification result set, calculates the difference of the multidimensional state coupling value of the corresponding time window, performs amplitude superposition of the difference between adjacent segments in time order, summarizes the total difference of the classification and the trend line index, and establishes the quantization trajectory of the foundation treatment effect.

8. The real-time monitoring and feedback system for the foundation treatment effect of a sluice gate according to claim 1, characterized in that, The process stage discrimination module includes: The continuous offset detection submodule quantizes the continuous offset segments in the trajectory based on the foundation treatment effect, extracts the amplitude change gradient between adjacent segments by time index, judges the continuity according to the set offset duration threshold, aggregates and marks the index segments that meet the continuity condition, and obtains the continuous offset time window set. The process disturbance correlation submodule calls the index of each time period in the continuous offset time window set, and synchronously obtains the operating parameter sequence of the ground treatment equipment within the corresponding time window, calculates the difference between any two adjacent sampling points in each parameter sequence, and determines whether it exceeds the process disturbance reference value, and obtains the disturbance correlation time window sequence. The stage identifier generation submodule filters out time windows that appear in the quantitative trajectory of the foundation treatment effect and are accompanied by process disturbances, based on the time window positions in the disturbance-related time window sequence. The filtered time windows are then structured and represented according to the start and end indices to establish the processing stage identifier content.

9. The real-time monitoring and feedback system for the foundation treatment effect of a sluice gate according to claim 1, characterized in that, The control command generation module includes: The physical field expansion judgment submodule, based on the time window index in the processing stage identifier content, calls the ground-based multiphysics spatial distribution model of the corresponding time frame, extracts the boundary contour of the state evolution region in each frame, calculates the boundary expansion direction and area increment of the region in adjacent frames, aggregates and judges the region expansion behavior in continuous time windows, and obtains the physical field expansion trend dataset. The rule cross-filtering submodule performs three-dimensional spatial coordinate cross-judgment based on the spatial positioning information of the corresponding time window in the physical field expansion trend dataset and the processing stage identifier content, extracts the spatial volume range of the cross-region, compares it with the spatial overlap threshold in the preset control rules, filters the data segments that meet the conditions, and obtains the rule matching cross-region set. The control instruction generation submodule calls the rules to match the state evolution level values ​​of the cross-regions, and retrieves the corresponding control instructions from the equipment control parameter library according to the rules based on the control strategy mapping relationship corresponding to the level values. It integrates the operation object, action type and execution parameters of the instructions to generate control instructions for the foundation treatment equipment.

10. The real-time monitoring and feedback system for the foundation treatment effect of a sluice gate according to claim 9, characterized in that, The spatial overlap threshold refers to whether the degree of overlap between the physical field expansion region and the processing stage identification region in three-dimensional spatial coordinates reaches the cross-volume standard for triggering control commands.

Images