A method for synchronously visually tracking particle movement and seepage paths at soil contact surfaces.

Abstract

This invention discloses a method for synchronously visually tracking the particle movement and seepage path at soil contact surfaces, belonging to the field of visual tracking technology. The method includes the following steps: S1, acquiring data and establishing a storage repository; S2, extracting feature parameters and converting them into feature vectors; S3, setting weighting rules based on the data and using a model to evaluate the state and generate evaluation results; S4, combining preset tracking rules with the vectors and evaluation results to obtain a synchronous visual tracking scheme, and verifying and adjusting it; S5, distributing the scheme to the monitoring equipment of the direct shear tester for real-time tracking and data collection; S6, setting preset anomaly judgment criteria for comparative analysis, adding tags, and storing the data in the storage repository. This invention, by acquiring basic parameters of soil contact surface particles, avoids the fragmented data acquisition in traditional soil contact surface tracking, improving the accuracy of evaluation results and the adaptability of the tracking scheme.

Description

Technical Field

[0001] This invention relates, in particular, to a method for synchronously visually tracking the movement of particles and the seepage path at soil contact surfaces. Background Technology

[0002] In the era of experimental research that deeply integrates geotechnical engineering and seepage mechanics, the synchronous visualization and tracking of particle movement and seepage path at the soil contact surface, as a core technology for revealing the deterioration of soil mechanical properties and the mechanism of disaster incubation under seepage-shear coupling, has been widely applied in key areas such as water conservancy engineering seepage prevention structure testing, rail transit foundation stability assessment, and underground engineering surrounding rock seepage control. The synchronicity of its tracking data, the accuracy of particle-seepage correlation, and the adaptability of the direct shear test scenario are directly related to the accuracy of the study of soil contact surface action mechanism and the reliability of engineering protection scheme design.

[0003] The particle seepage collaborative tracking technology has demonstrated groundbreaking value in the field of soil contact surface test research due to its ability to integrate multi-physics field data and its advantages in visualization. By integrating core technical elements such as particle feature extraction, seepage field simulation, and multi-parameter synchronous acquisition, it constructs an integrated collaborative system of "data-model-tracking". It has become a key technical direction to break through the limitations of traditional single-physics field observation and improve the depth of soil contact surface test research, and has important scientific research reference significance and engineering application value.

[0004] However, existing methods for synchronously visually tracking particle movement and seepage paths at soil contact surfaces often only collect particle movement or seepage path data individually, lacking multi-dimensional collaborative design. They fail to integrate direct shear tester operating parameters with historical tracking data, resulting in low correlation of tracking data. This makes it difficult to meet the precise analysis needs of complex seepage-shear coupling scenarios. Furthermore, they lack a standardized mapping relationship between particle seepage characteristics and direct shear tester hardware capabilities, often using generic visualization parameters. This leads to poor compatibility between the tracking scheme and the direct shear tester's measurement accuracy and acquisition frequency, resulting in significant data redundancy and missing key information. Currently, no effective solutions have been proposed to address these technical problems. Summary of the Invention

[0005] To address the problems in related technologies, this invention proposes a method for synchronously visually tracking the movement of particles and the seepage path at the soil contact surface, thereby overcoming the aforementioned technical problems in existing related technologies.

[0006] To achieve the above objectives, the specific technical solution adopted by the present invention is as follows:

[0007] A method for synchronously visually tracking particle movement and seepage paths at soil contact surfaces includes the following steps:

[0008] S1. Obtain the soil contact surface particle foundation parameters, seepage medium property data and historical synchronous tracking data under the same working conditions that can be collected by the direct shear apparatus, and set up a soil particle seepage collaborative tracking data repository.

[0009] S2. Extract key feature parameters from the particle foundation parameters of the soil contact surface, and pre-set contact surface feature mapping rules that match the direct shear test scenario. Based on the contact surface feature mapping rules, transform the key feature parameters into standardized particle seepage collaborative feature vectors.

[0010] As a preferred embodiment, the extraction of key feature parameters from the particle foundation parameters of the soil contact surface, the pre-setting of contact surface feature mapping rules matching the direct shear test scenario, and the conversion of key feature parameters into standardized particle seepage collaborative feature vectors according to the contact surface feature mapping rules include the following steps:

[0011] S21. Extract the particle geometric and mechanical characteristic parameters, seepage medium physical characteristic parameters, and contact surface mechanical and structural characteristic parameters from the particle foundation parameters of the soil contact surface as three types of key characteristic parameters.

[0012] S22. Preset classification mapping rules, dimension normalization mapping rules and synergistic correlation mapping rules. The classification mapping rules correspond the three key feature parameters to the particle motion influence dimension, seepage path influence dimension and synergistic correlation dimension respectively. The dimension normalization mapping rules unify the parameter dimensions. At the same time, the synergistic correlation mapping rules quantify the synergistic weights between parameters.

[0013] S23. Based on the three types of key feature parameters after mapping processing, construct an N-dimensional feature vector, perform consistency verification, add working condition labels and collaborative association weight labels to form a standardized particle seepage collaborative feature vector, and store it in the soil particle seepage collaborative tracking data repository.

[0014] As a preferred embodiment, the process of constructing an N-dimensional feature vector based on the three types of key feature parameters after mapping, performing consistency verification, adding working condition labels and collaborative association weight labels to form a standardized particle seepage collaborative feature vector, and storing it in the soil particle seepage collaborative tracking data repository includes the following steps:

[0015] S231. Arrange the particle motion influence dimension parameter, seepage path influence dimension parameter, and synergistic effect correlation dimension parameter in the three types of key feature parameters after mapping processing in the order of particles, seepage and synergy to construct an N-dimensional feature vector.

[0016] S232. Preset the measurement accuracy range of the direct shear apparatus parameters, compare the deviation values ​​of each parameter in the N-dimensional feature vector with the particle foundation parameters of the soil contact surface, and compare the deviation values ​​with the measurement accuracy range of the direct shear apparatus parameters. If the deviation value exceeds the measurement accuracy range of the direct shear apparatus parameters, return to S22 to readjust the feature mapping rules. If the deviation value is within the measurement accuracy range of the direct shear apparatus parameters, retain the N-dimensional feature vector unchanged.

[0017] S233. Generate working condition labels based on three types of key feature parameters, generate collaborative association weight labels by combining collaborative association weights, and associate the working condition labels and collaborative association weight labels with N-dimensional feature vectors.

[0018] S234. Define the labeled N-dimensional feature vector as a standardized particle seepage collaborative feature vector, and store it in the soil particle seepage collaborative tracking data repository according to the indexing rules of the direct shear instrument condition number and generation time in the historical synchronous tracking data of the same working condition.

[0019] S3. Based on the soil contact surface particle foundation parameters and seepage medium property data, and combined with the seepage shear coupling function of the direct shear apparatus, a multi-dimensional synergy evaluation weighting rule is set, and a DEM-LBM coupling architecture model calibrated by the direct shear apparatus is used to evaluate the synergy status of particle movement and seepage path, and generate synergy evaluation results.

[0020] As a preferred embodiment, the method of setting multi-dimensional synergistic evaluation weighting rules based on soil contact surface particle foundation parameters and seepage medium property data, combined with the seepage-shear coupling function of the direct shear apparatus, and using a DEM-LBM coupled architecture model calibrated under direct shear apparatus conditions to evaluate the synergistic state of particle movement and seepage path, and generating synergistic evaluation results includes the following steps:

[0021] S31. Based on the particle foundation parameters of the soil contact surface and the property data of the seepage medium, set the influence weight of particle movement, the influence weight of seepage path, and the synergistic correlation weight of the influence weight of particle movement and the influence weight of seepage path, and combine them to form a multi-dimensional synergistic evaluation weighting rule.

[0022] S32. Input the soil contact surface particle foundation parameters, seepage medium property data and multi-dimensional synergy evaluation weighting rules into the DEM-LBM coupled architecture model. Simulate the interaction process between particle movement and seepage field through the DEM-LBM coupled architecture model, and quantify the particle matching degree of particle displacement and seepage flow rate, the porosity and seepage path porosity correlation and the interface mechanical properties on the interface influence value of the synergy state.

[0023] S33. Verify the consistency of particle matching degree, pore correlation and interface influence value, and integrate the verified particle matching degree, pore correlation and interface influence value to form the model simulation results;

[0024] As a preferred embodiment, the process of verifying the consistency of particle matching degree, porosity correlation, and interface influence values, and integrating the verified particle matching degree, porosity correlation, and interface influence values ​​to form the model simulation results includes the following steps:

[0025] S331, preset particle matching degree qualification threshold, pore association qualification threshold and interface influence value qualification threshold;

[0026] S332. Compare the particle matching degree with the qualified threshold of particle matching degree, the porosity correlation with the qualified threshold of porosity correlation, and the interface influence value with the qualified threshold of interface influence value. Determine whether the particle matching degree, porosity correlation, and interface influence value meet the qualified thresholds of particle matching degree, porosity correlation, and interface influence value. If any one or more of them do not meet the threshold requirements, return to S32 to adjust the simulation boundary conditions of the DEM-LBM coupled architecture model. If all of them meet the threshold requirements, the consistency verification is deemed to have passed.

[0027] S333. Following the dimensional order of particle matching degree, porosity correlation, and interface influence value, the verified particle matching degree, porosity correlation, and interface influence value are integrated into multidimensional structural data, and the model simulation timestamp and direct shear apparatus condition number are labeled to form the model simulation results.

[0028] S34. Extract the collaborative state characteristic indicators from the model simulation results, and generate a collaborative evaluation result that includes the collaborative matching index, the collaborative anomaly risk level, and the parameter correlation matrix.

[0029] As a preferred embodiment, the step of extracting the collaborative state characteristic indicators from the model simulation results and generating a collaborative evaluation result including a collaborative matching index, a collaborative anomaly risk level, and a parameter correlation matrix includes the following steps:

[0030] S341. Preset simulation result weights and matrix correlation thresholds, extract particle matching degree and porosity correlation from model simulation results, calculate the weighted average of particle matching degree and porosity correlation according to simulation result weights, and define it as the cooperative matching index;

[0031] S342. Based on the particle displacement, seepage flow rate and interface pressure data in the model simulation results, construct a parameter correlation matrix, count the proportion of elements in the parameter correlation matrix whose absolute value is less than the matrix correlation threshold, and generate a collaborative anomaly risk level based on the proportion of elements.

[0032] S343. Verify and correct the parameter correlation matrix with particle displacement, seepage flow rate and interface pressure data, and integrate the synergy matching index, synergy anomaly risk level and the verified and corrected parameter correlation matrix to generate synergy assessment results and associate them with the corresponding direct shear apparatus operating condition number.

[0033] S4. Preset multi-parameter synchronous tracking rules adapted to the hardware capabilities of the direct shear apparatus, adopt the particle seepage multi-parameter collaborative matching algorithm, combine the standardized particle seepage collaborative feature vector and collaborative evaluation results to obtain a synchronous visualization tracking scheme, and verify and optimize the accuracy of the synchronous visualization tracking scheme with historical synchronous tracking data;

[0034] As a preferred embodiment, the preset multi-parameter synchronous tracking rule adapted to the hardware capabilities of the direct shear apparatus adopts a particle seepage multi-parameter collaborative matching algorithm, combined with standardized particle seepage collaborative feature vectors and collaborative evaluation results, to obtain a synchronous visual tracking scheme. The synchronous visual tracking scheme is then verified and optimized against historical synchronous tracking data, including the following steps:

[0035] S41. Preset tracking parameter priority rules, data acquisition frequency adaptation rules, and visualization observation accuracy adaptation rules, and combine them to form multi-parameter synchronous tracking rules;

[0036] S42. Input the standardized particle seepage collaborative feature vector, collaborative evaluation results and multi-parameter synchronous tracking rules into the particle seepage multi-parameter collaborative matching algorithm. Quantify the fit between the feature vector and the evaluation results through the particle seepage multi-parameter collaborative matching algorithm to generate an initial visual tracking scheme that includes data acquisition parameters, visual observation optimization parameters and anomaly warning thresholds.

[0037] S43. Retrieve historical synchronous tracking data from the soil particle seepage collaborative tracking data repository, compare the initial visualization tracking scheme with the historical data, and calculate the scheme deviation value of the initial visualization tracking scheme in terms of tracking accuracy and collaborative matching degree.

[0038] As a preferred embodiment, the step of retrieving historical synchronous tracking data from the soil particle seepage collaborative tracking data repository, comparing the initial visual tracking scheme with the historical data, and calculating the scheme deviation value of the initial visual tracking scheme in terms of tracking accuracy and collaborative matching degree includes the following steps:

[0039] S431. Filter the soil particle seepage collaborative tracking data in the soil particle seepage collaborative tracking data repository according to the working condition number and timestamp index rules to select historical synchronous tracking data, and integrate the historical synchronous tracking data to form a benchmark dataset.

[0040] S432. Construct a deviation calculation model based on the benchmark dataset in different dimensions, and input the initial visual tracking scheme and historical data into the deviation calculation model to calculate the tracking accuracy deviation value and the collaborative matching degree deviation value respectively.

[0041] S433. Verify and adjust the tracking accuracy deviation value and the collaborative matching degree deviation value, and output the verified and adjusted tracking accuracy deviation value and collaborative matching degree deviation value as the initial visual tracking scheme's scheme deviation value in terms of tracking accuracy and collaborative matching degree.

[0042] S44. Based on the deviation value of the scheme and the risk level of collaborative anomaly in the collaborative assessment results, adjust the collection frequency, visualization observation optimization parameters and early warning threshold in the initial scheme to form a synchronous visualization tracking scheme.

[0043] As a preferred option, the process of adjusting the initial sampling frequency, visualization observation optimization parameters, and early warning threshold based on the scheme deviation value and the risk level of collaboration anomaly in the collaboration assessment results to form a synchronous visualization tracking scheme includes the following steps:

[0044] S441. Preset the scheme deviation threshold to match the accuracy of the direct shearing instrument, compare the scheme deviation value with the scheme deviation threshold, and determine whether the deviation exceeds the standard based on the scheme deviation comparison result, and form a preliminary adjustment scheme.

[0045] S442. Re-compare the preliminary adjustment plan with historical synchronous tracking data and calculate the advanced deviation value.

[0046] If the advanced deviation value is less than or equal to the scheme deviation threshold, the initial adjustment scheme is defined as the synchronous visual tracking scheme.

[0047] If the advanced deviation value is still greater than the scheme deviation threshold, return to S441 to re-execute the adjustment strategy until the advanced deviation value is less than or equal to the scheme deviation threshold.

[0048] S5. The verified and adjusted synchronous visualization tracking scheme is sent to the monitoring equipment of the direct shear machine to track the coordinated state of particle movement and seepage path at the soil contact surface in real time, and collect particle displacement rate, seepage flow rate, visualization image frame data and interface pressure change data.

[0049] S6. Preset tracking accuracy thresholds and anomaly judgment criteria that match the accuracy of the direct shear apparatus. Based on the tracking accuracy thresholds and anomaly judgment criteria, compare and analyze historical synchronous tracking data with real-time acquired data, add tags according to the comparison results, and store them in the soil particle seepage collaborative tracking data repository.

[0050] As a preferred embodiment, the preset tracking accuracy threshold and anomaly judgment criteria that match the accuracy of the direct shear apparatus, and the comparison and analysis of historical synchronous tracking data with real-time acquired data based on the tracking accuracy threshold and anomaly judgment criteria, and the addition of tags according to the comparison results, and the storage in the soil particle seepage collaborative tracking data repository, include the following steps:

[0051] S61. Based on the parameter measurement accuracy of the direct shear apparatus, preset multi-dimensional tracking accuracy thresholds, and based on the seepage shear coupling test scenario of the direct shear apparatus, preset anomaly judgment criteria.

[0052] S62. Compare the real-time collected data with the historical synchronous tracking data under the same working conditions point by point, calculate the relative error value of each dimension of data, and compare and judge the relative error value with the tracking accuracy threshold and the anomaly judgment standard.

[0053] S63. Add tags to the real-time collected data based on the comparison and judgment results, and record the real-time collected data when an anomaly occurs;

[0054] S64. Store the tagged real-time collected data into the soil particle seepage collaborative tracking data repository according to the indexing rules of timestamp and working condition number.

[0055] The beneficial effects of this invention are as follows:

[0056] 1. This invention acquires soil contact surface particle basic parameters, seepage medium property data, and historical synchronous tracking data, and constructs a soil particle seepage collaborative tracking data repository. It combines a DEM-LBM coupled architecture model calibrated under direct shear test conditions with multi-dimensional collaborative evaluation weighting rules to achieve accurate assessment of particle seepage collaborative status. Simultaneously, it employs collaborative correlation mapping rules to optimize parameter collaborative weights and a multi-parameter collaborative matching algorithm for particle seepage to generate a synchronous visual tracking scheme. This avoids the problems of fragmented data acquisition, neglect of direct shear test hardware differences and operating conditions in traditional soil contact surface tracking, and poor adaptability of tracking schemes to particle seepage characteristics. It reduces the risk of wasted observation resources, low test execution efficiency, or substandard data accuracy due to blind design. Furthermore, by constructing standardized particle seepage collaborative feature vectors and verifying direct shear test accuracy deviations, it strengthens the expression of key dimension features, ensuring that the collaborative evaluation logic aligns with the seepage-shear coupling test principle, avoiding interference from invalid parameters, and improving the accuracy of evaluation results and the adaptability of the tracking scheme.

[0057] 2. This invention achieves real-time monitoring and tagged storage of experimental data by comparing the deviation between the synchronous visualization tracking scheme and historical synchronous tracking data, and by dynamically adjusting and optimizing the scheme. It combines preset tracking accuracy thresholds and anomaly judgment criteria to solve the pain points of traditional tracking schemes, which lack historical data support and closed-loop iterative optimization mechanisms. This reduces the cost of repeated debugging and the iteration cycle of the scheme, and improves the reliability and scalability of synchronous tracking of particle seepage. At the same time, through DEM-LBM model consistency verification and direct shear instrument monitoring equipment resource adaptation process, it ensures the model and hardware readiness before the test execution, avoids tracking interruption caused by inaccurate model boundary conditions or mismatched equipment parameters, and ensures the stability and effectiveness of experimental data acquisition. Attached Figure Description

[0058] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0059] Figure 1 This is a flowchart of a method for synchronously visually tracking the movement of particles and the seepage path at the soil contact surface according to an embodiment of the present invention.

[0060] Figure 2 This is a diagram of an electro-hydraulic servo dual-channel dynamic compression-shear loading test device for a synchronous visualization tracking method of particle movement and seepage path at the soil contact surface according to an embodiment of the present invention.

[0061] Figure 3 This is a system structure diagram of a synchronous visualization tracking method for particle movement and seepage path at the soil contact surface according to an embodiment of the present invention. Detailed Implementation

[0062] The specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and are not intended to limit the scope of the invention.

[0063] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0064] The present invention will now be further described in conjunction with the accompanying drawings and specific embodiments, such as... Figures 1-3 As shown, the synchronous visualization and tracking method for particle movement and seepage path at the soil contact surface according to an embodiment of the present invention includes the following steps:

[0065] S1. Obtain the soil contact surface particle foundation parameters, seepage medium property data and historical synchronous tracking data under the same working conditions that can be collected by the direct shear apparatus, and set up a soil particle seepage collaborative tracking data repository.

[0066] Specifically, the particle basic parameters of the soil contact surface are collected as follows: Extracted from the detection module of the direct shear tester, the geometric characteristics of the particles (e.g., particle size distribution of coastal sandy / cohesive soil, maximum particle size ≤ 60mm, compatible with a shear box size of 300mm × 300mm × 150mm, meeting the boundary effect control standard of "direct shear box length ≥ 35 times particle size") are captured using a high-speed CCD camera (maximum frame rate 79 frames / s, equipped with an Edmund telecentric lens, resolution 0.01mm / pixel). The interparticle friction angle (e.g., 32° for coastal sand particles) and compressive strength are collected using particle mechanics sensors. Combined with the direct shear tester's normal / tangential force sensors (maximum normal force 200kN, maximum tangential force 150kN, spoke structure, overload capacity 150%, accuracy ±1%FS), the relevant particle stress parameters (e.g., density 1.6g / cm³) are indirectly derived. 3 (Calculated by "soil sample weight - volume ratio") to ensure that the data accurately reflects the actual mechanical state of the soil-structure contact surface.

[0067] Data acquisition of seepage medium properties: Extracted from the water pressure seepage control system of the direct shear apparatus. This system includes a pressure pump, control valve, and sponge dispersion device (the inlet sponge pore size is ≤1mm to avoid seepage path). Seepage water pressure is collected by a pressure sensor (0-1MPa, higher than the household tap water pressure by 0.45MPa, with stable seepage pressure fluctuation ≤±0.05MPa); seepage flow rate (unit cm / s, such as 0.45-0.5cm / s in the coastal sand test) is collected by an inlet / outlet flow meter (accuracy ±5%); pore water pressure (e.g., 80-90kPa) is collected by a pore water pressure gauge (accuracy ±2%FS) installed inside the soil sample; the water content of the seepage medium (e.g., 15% water content in coastal sand) and permeability coefficient are recorded (calculated through the "seepage water pressure-flow rate relationship", which meets the requirements of groundwater level fluctuation scenarios in coastal areas).

[0068] Historical synchronous tracking data acquisition under the same working conditions: exported from the direct shear tester computer measurement and control system (ASUS industrial computer, 32-bit Win7 system, equipped with two-channel measurement and control card) and test logs, including particle displacement rate (mm / min, average 0.12mm / min), seepage flow rate change curve, interface pressure (kPa, average 120kPa) and visualization image frame data (TIFF format, each frame associated with a timestamp); the selection criteria must meet the following requirements: "72h continuous operation without failure, seepage sealing performance of 1MPa pressure holding for 30min without leakage (the shear box connection uses fluororubber O-rings, sealing gasket thickness 2.5mm), data integrity ≥95%", and cover the high seepage pressure-dynamic load conditions simulated by typhoons (such as 2Hz normal sinusoidal wave load, amplitude 50kPa).

[0069] The soil particle seepage collaborative tracking data repository is configured as follows: It adopts a "local storage (1TB industrial hard drive) + cloud backup (Alibaba Cloud, hourly incremental backup + daily full backup)" architecture, divided into three core modules:

[0070] Particle parameter module: Stores particle geometry (particle size, particle size distribution) and mechanical data (friction angle, density);

[0071] Seepage attribute module: Stores seepage pressure, flow rate, pore water pressure data and permeability coefficient;

[0072] Historical Scheme Module: Stores the same working condition tracking scheme (collection frequency, early warning threshold) and test results (displacement field cloud map, seepage path image).

[0073] Establish a multi-dimensional data index: associate data by “working condition number (e.g., coastal sand-C30 concrete-1MPa seepage pressure-2Hz normal dynamic load) + test timestamp (YYYYMMDD-HHMMSS) + equipment number (e.g., ZC20250904)”, and support quick filtering by soil sample type (sandy soil / cohesive soil), seepage pressure level (low / medium / high), and load frequency (0.01-5Hz).

[0074] Access control is set up: differentiate between three levels of permissions: "administrator, experimenter (responsible for data reading / scheme retrieval), and auditor (responsible for log viewing)". Data security is ensured through the password verification mechanism of the direct shear tester control system (password is an 8-character combination of letters and numbers); at the same time, a data interface is reserved to support the reuse of historical data and deviation comparison when the scheme is generated in the future.

[0075] S2. Extract key feature parameters from the particle foundation parameters of the soil contact surface, and pre-set contact surface feature mapping rules that match the direct shear test scenario. Based on the contact surface feature mapping rules, transform the key feature parameters into standardized particle seepage collaborative feature vectors.

[0076] As a preferred embodiment, the extraction of key feature parameters from the particle foundation parameters of the soil contact surface, the pre-setting of contact surface feature mapping rules matching the direct shear test scenario, and the conversion of key feature parameters into standardized particle seepage collaborative feature vectors according to the contact surface feature mapping rules include the following steps:

[0077] S21. Extract the particle geometric and mechanical characteristic parameters, seepage medium physical characteristic parameters, and contact surface mechanical and structural characteristic parameters from the particle foundation parameters of the soil contact surface as three types of key characteristic parameters.

[0078] Specifically, we will focus on the "core dimensions affecting the synergy between particle movement and seepage" from the basic particle parameters:

[0079] Particle geometry and mechanics dimensions: Extracting "particle size distribution (focusing on the 35-fold particle size threshold, such as coastal sand with a maximum particle size of 5mm, where 35-fold particle size 210mm < shear box length 300mm, meeting the matching requirements), interparticle friction angle (32°), and particle density (1.6g / cm³)." 3 "(Probability of flow path continuity)";

[0080] The dimensions associated with the seepage medium are: "seepage pressure (1MPa), seepage flow rate (0.45cm / s), and porosity (0.35, measured by 'soil sample pore volume / total volume', to meet the pore simulation requirements of the DEM-LBM model)".

[0081] Contact surface coupling dimension: Extract “soil-structure contact surface cohesion (15kPa), internal friction angle (28°, structural specimen is C30 concrete, size 300mm×300mm×150mm, placed in lower shear box), contact surface displacement discontinuity (0.005mm, collected by a wire-type tangential displacement sensor of the direct shear tester, resolution 0.001mm, reflecting the development state of the shear band at the contact surface)”, to ensure that the extracted parameters cover the coupling correlation points of “particle-seepage-contact surface”.

[0082] S22. Preset classification mapping rules, dimension normalization mapping rules and synergistic correlation mapping rules. The classification mapping rules correspond the three key feature parameters to the particle motion influence dimension, seepage path influence dimension and synergistic correlation dimension respectively. The dimension normalization mapping rules unify the parameter dimensions. At the same time, the synergistic correlation mapping rules quantify the synergistic weights between parameters.

[0083] Specifically, the classification mapping rules are as follows: key feature parameters are classified into "particle motion influence dimensions (particle size, friction angle, density), seepage path influence dimensions (seepage pressure, flow rate, porosity), and synergistic effect correlation dimensions (contact surface cohesion, internal friction angle, displacement discontinuity)". Among them, the particle motion dimension parameters correspond to the tangential loading system of the direct shear apparatus (shear rate 0.00025-5mm / min), the seepage path dimension parameters correspond to the seepage control system (pressure 0-1MPa), and the synergistic dimension parameters correspond to the "seepage-shear coupling" functional module (supporting the synchronous application of seepage and dynamic load).

[0084] Dimensional normalization mapping rules: Based on the hardware accuracy of the direct shear apparatus, particle size is normalized according to "actual particle size / 60mm (maximum particle size)" (e.g., 2mm-0.033, 5mm-0.083), seepage pressure is normalized according to "actual water pressure / 1MPa (maximum seepage pressure)" (1Mpa-1.0), and porosity is normalized according to "actual porosity / 1.0" (0.3-0.35). This ensures that the normalized parameters are all within the range of [0,1] and are compatible with the measurement accuracy of the direct shear apparatus parameters (normal pressure ±1%FS, displacement ±0.5%FS), avoiding interference from dimensional differences in collaborative evaluation.

[0085] Synergistic correlation mapping rules: Based on the weight allocation of the test scenario, under normal working conditions, the weight of particle motion dimension is 0.3, the weight of seepage path dimension is 0.4, and the weight of synergistic correlation dimension is 0.3; when simulating the "high seepage pressure plus dynamic load" working condition caused by a typhoon, the weight of seepage path dimension is increased to 0.45, the weight of synergistic correlation dimension is increased to 0.35, and the weight of particle motion dimension is reduced to 0.2 (the adjustment basis is "the contribution rate of seepage to the attenuation of contact surface strength under this working condition is >45%)); the rationality of the weights is verified by "historical test data under the same working condition". If the correlation between seepage pressure fluctuation and contact surface strength attenuation is ≥0.8 (calculated by Pearson correlation analysis), the weights are retained; otherwise, the "analytic hierarchy process (AHP)" is used to iterate and optimize again to ensure that the consistency test CR <0.1 (e.g., CR=0.08 in this embodiment, the weights are valid).

[0086] S23. Based on the three types of key feature parameters after mapping processing, construct an N-dimensional feature vector, perform consistency verification, add working condition labels and collaborative association weight labels to form a standardized particle seepage collaborative feature vector, and store it in the soil particle seepage collaborative tracking data repository.

[0087] As a preferred embodiment, the process of constructing an N-dimensional feature vector based on the three types of key feature parameters after mapping, performing consistency verification, adding working condition labels and collaborative association weight labels to form a standardized particle seepage collaborative feature vector, and storing it in the soil particle seepage collaborative tracking data repository includes the following steps:

[0088] S231. Arrange the particle motion influence dimension parameter, seepage path influence dimension parameter, and synergistic effect correlation dimension parameter in the three types of key feature parameters after mapping processing in the order of particles, seepage and synergy to construct an N-dimensional feature vector.

[0089] Specifically, according to the "dimensional parameters affecting particle motion (3 parameters: normalized particle size 0.083, friction angle 32°, density 1.6 g / cm³)", 3The parameters affecting the seepage path (3 parameters: normalized seepage pressure 1.0, seepage flow rate 0.45 cm / s, porosity 0.35) and the parameters related to the synergistic effect (3 parameters: cohesion 15 kPa, internal friction angle 28°, displacement discontinuity 0.005 mm) are arranged in the following order to form a 9-dimensional feature vector (the vector dimensions can be flexibly increased or decreased according to the test conditions, such as adding the "clay content" parameter to expand it to 10 dimensions for cohesive soil tests).

[0090] S232. Preset the measurement accuracy range of the direct shear apparatus parameters (normal force ±1%FS, tangential displacement ±0.5%FS, seepage pressure ±0.05MPa), compare the deviation values ​​of each parameter in the N-dimensional feature vector with the particle foundation parameters of the soil contact surface (e.g., normalized particle size deviation 0.002, seepage pressure deviation 0.02MPa), and compare the deviation values ​​with the measurement accuracy range of the direct shear apparatus parameters. If the deviation value exceeds the measurement accuracy range of the direct shear apparatus parameters (e.g., seepage pressure deviation 0.06MPa > ±0.05MPa), return to S22 to readjust the feature mapping rules (e.g., correct the dimensionless normalization coefficient). If the deviation value is within the measurement accuracy range of the direct shear apparatus parameters (e.g., in this embodiment, the deviation of each parameter is ≤ ±0.03), then retain the N-dimensional feature vector unchanged.

[0091] S233. Generate working condition labels based on three types of key feature parameters, generate collaborative association weight labels by combining collaborative association weights, and associate the working condition labels and collaborative association weight labels with N-dimensional feature vectors.

[0092] Specifically, the working condition label is generated based on "soil sample type (sandy soil) + seepage pressure level (high, 1MPa) + load type (dynamic, 2Hz sine wave)", with the format "coastal sand - high seepage pressure - 2Hz dynamic load"; the collaborative association weight label is generated based on the weight determined by S22, with the format "particle 0.2 - seepage 0.45 - collaborative 0.35"; the two types of labels are bound to the 9-dimensional feature vector to form a labeled feature vector.

[0093] S234. Define the labeled N-dimensional feature vector as a standardized particle seepage collaborative feature vector, and store it in the soil particle seepage collaborative tracking data repository according to the indexing rules of the direct shear instrument condition number and generation time in the historical synchronous tracking data of the same working condition.

[0094] Specifically, according to the indexing rule of "working condition number (coastal sand-C30 concrete-1MPa seepage pressure-2Hz normal dynamic load) + generation time (20251015-143000)", the standardized feature vector is stored in the "collaborative feature subdirectory" under the particle parameter module, and the storage log (including storage person, storage time, data size) is recorded to facilitate subsequent retrieval and traceability.

[0095] S3. Based on the soil contact surface particle foundation parameters and seepage medium property data, and combined with the seepage shear coupling function of the direct shear apparatus, a multi-dimensional synergy evaluation weighting rule is set, and a DEM-LBM coupling architecture model calibrated by the direct shear apparatus is used to evaluate the synergy status of particle movement and seepage path, and generate synergy evaluation results.

[0096] As a preferred embodiment, the method of setting multi-dimensional synergistic evaluation weighting rules based on soil contact surface particle foundation parameters and seepage medium property data, combined with the seepage-shear coupling function of the direct shear apparatus, and using a DEM-LBM coupled architecture model calibrated under direct shear apparatus conditions to evaluate the synergistic state of particle movement and seepage path, and generating synergistic evaluation results includes the following steps:

[0097] S31. Based on the particle foundation parameters of the soil contact surface and the property data of the seepage medium, set the influence weight of particle movement, the influence weight of seepage path, and the synergistic correlation weight of the influence weight of particle movement and the influence weight of seepage path, and combine them to form a multi-dimensional synergistic evaluation weighting rule.

[0098] Specifically, the basic weighting is as follows: Based on the core requirements of the direct shear apparatus's seepage shear coupling function (supporting the simultaneous application of seepage and static / dynamic loads), the following weightings are set: particle motion influence weight 0.3 (related to the variable rate shear capacity of the tangential loading system, 0.00025-5mm / min), seepage path influence weight 0.4 (related to the seepage system's 1MPa stable seepage pressure function, fluctuation ≤±0.05MPa), and synergistic effect influence weight 0.3 (related to the "synchronous application of seepage-stress" function, such as the synchronous activation of normal dynamic load and seepage).

[0099] Dynamic weight adjustment: For the special working condition of "high osmotic pressure (1MPa) plus dynamic load (2Hz)" in typhoon simulation, based on the experimental conclusion that "the contribution rate of seepage to the attenuation of contact surface strength is >45%", the weight of seepage path influence is increased to 0.45, the weight of synergistic effect correlation is increased to 0.35, and the weight of particle movement influence is reduced to 0.2.

[0100] Weight Verification: The rationality of the weights is verified by using "historical test data under the same working conditions". The test data of the last three "coastal sand-1MPa seepage pressure-2Hz dynamic load" tests are selected, and the collaborative evaluation results (such as the collaborative matching index) are calculated according to the weights. The results are compared with the actual failure state of the contact surface (shear band position deviation, strength value deviation). If the deviation is <5% (such as a strength value deviation of 3.2%), the weight rules are determined. Otherwise, the weight coefficients are iteratively optimized again using the "analytic hierarchy process (AHP)" until the deviation is <5%.

[0101] S32. Input the soil contact surface particle foundation parameters, seepage medium property data and multi-dimensional synergy evaluation weighting rules into the DEM-LBM coupled architecture model. Simulate the interaction process between particle movement and seepage field through the DEM-LBM coupled architecture model, and quantify the particle matching degree of particle displacement and seepage flow rate, the porosity and seepage path porosity correlation and the interface mechanical properties on the interface influence value of the synergy state.

[0102] Specifically, the model boundary conditions are calibrated as follows: The physical structure of the PWS-Y200 / 150J direct shear apparatus is strictly matched. The particle motion boundary is set according to "horizontal sliding of the lower shear box (guided by a slide rail, deviation ≤0.1mm)," with the lower shear box and the base plate sealed by fluororubber O-rings. The seepage boundary is set according to "water inlet on the right side wall of the upper shear box (built-in sponge to disperse water flow, pore size ≤1mm, to avoid local seepage paths), water outlet on the left side wall plus drainage from the bottom of the lower shear box," with the outer frame height 0.2mm lower than the lower shear box to ensure the lower box can move freely when there is no pressure. The dynamic load boundary is calibrated according to the normal dynamic load parameters of the direct shear apparatus (0.01-5Hz, sine wave simulating seismic waves, square wave simulating impact loads). In this embodiment, a 2Hz sine wave with an amplitude of 50kPa and a maximum normal stress of 2.2MPa is used.

[0103] Model parameter calibration: Input the basic particle parameters (particle size 2-5mm, friction angle 32°) and seepage medium property data (seepage pressure 1MPa, porosity 0.35) into the model, and adjust the DEM-LBM parameters through "comparison of direct shear tester measured data and model simulation data": calibrate the particle contact stiffness to 80N / m according to the "measured value of normal force sensor (e.g., contact stiffness corresponding to 120kPa)", and calibrate the fluid viscosity coefficient to 3×10 according to the "measured value of seepage flow (0.45cm / s)". -6 Pa・s, ensuring that the deviation between the model simulation results and the measured data is ≤±5% (e.g., simulated seepage flow rate is 0.47 cm / s, measured is 0.45 cm / s, deviation is 4.4%).

[0104] Cooperative state simulation and quantification: The interaction between particle motion (displacement, overturning) and seepage field (seepage path distribution, pore water pressure change) is simulated by the model, and three types of indicators are quantified and output: particle matching degree (the temporal correlation between particle displacement and seepage flow rate, calculated by Pearson correlation analysis, which is 0.85 in this embodiment, and the threshold ≥0.8 is qualified); porosity correlation (the correlation coefficient between porosity and seepage path conductivity, which is 0.78 in this embodiment, and the threshold ≥0.7 is qualified); and interface influence value (the resistance coefficient of the mechanical properties of the contact surface to seepage, calculated based on the cohesion and internal friction angle of the contact surface, which is 0.25 in this embodiment, and the threshold ≤0.3 is qualified).

[0105] S33. Verify the consistency of particle matching degree, pore correlation and interface influence value, and integrate the verified particle matching degree, pore correlation and interface influence value to form the model simulation results;

[0106] As a preferred embodiment, the process of verifying the consistency of particle matching degree, porosity correlation, and interface influence values, and integrating the verified particle matching degree, porosity correlation, and interface influence values ​​to form the model simulation results includes the following steps:

[0107] S331, preset particle matching degree qualification threshold (≥0.8), pore correlation qualification threshold (≥0.7) and interface influence value qualification threshold (≤0.3), and the threshold setting is based on the direct shear test acceptance standard;

[0108] S332. Compare the particle matching degree (0.85) with the qualified threshold of particle matching degree (0.8), the porosity correlation (0.78) with the qualified threshold of porosity correlation (0.7), and the interface influence value (0.25) with the qualified threshold of interface influence value (0.3). Determine whether the particle matching degree, porosity correlation and interface influence value meet the qualified thresholds of particle matching degree, porosity correlation and interface influence value. If any one or more do not meet the threshold requirements (e.g., particle matching degree 0.75 < 0.8), return to S32 to adjust the simulation boundary conditions of the DEM-LBM coupled architecture model (e.g., correct particle contact stiffness, adjust seepage boundary pressure), and re-simulate until all indicators meet the thresholds. If all indicators meet the threshold requirements, the consistency verification is deemed successful.

[0109] S333. Following the dimensional order of particle matching degree, porosity correlation, and interface influence value, the verified particle matching degree (0.85), porosity correlation (0.78), and interface influence value (0.25) are integrated into multidimensional structural data (format: [0.85, 0.78, 0.25]), and the model simulation timestamp (e.g., 20251015-144500) and direct shear tester condition number (e.g., coastal sand-C30 concrete-1MPa seepage pressure-2Hz normal dynamic load) are labeled to form the model simulation results.

[0110] S34. Extract the collaborative state characteristic indicators from the model simulation results, and generate a collaborative evaluation result that includes the collaborative matching index, the collaborative anomaly risk level, and the parameter correlation matrix.

[0111] As a preferred embodiment, the step of extracting the collaborative state characteristic indicators from the model simulation results and generating a collaborative evaluation result including a collaborative matching index, a collaborative anomaly risk level, and a parameter correlation matrix includes the following steps:

[0112] S341. Preset simulation result weights and matrix correlation thresholds, extract particle matching degree and porosity correlation from model simulation results, calculate the weighted average of particle matching degree and porosity correlation according to simulation result weights, and define it as the cooperative matching index;

[0113] Specifically, the following indicators are extracted to represent the synergistic state characteristics: particle matching degree and pore correlation are extracted from the model simulation results (with weights of 0.5 and 0.5, respectively), and the weighted average is calculated as the "synergistic matching index" (≥0.75 is excellent, 0.6-0.75 is good, and <0.6 is poor); a 3×3 parameter correlation matrix is ​​constructed based on particle displacement, seepage flow, and interface pressure data, and the proportion of elements with an absolute value <0.3 (matrix correlation threshold) in the matrix is ​​statistically analyzed (proportion >60% is low risk, 30%-60% is medium risk, and <30% is high risk), which is used as the "synergistic anomaly risk level".

[0114] S342. Based on the particle displacement, seepage flow rate and interface pressure data in the model simulation results, construct a parameter correlation matrix, count the proportion of elements in the parameter correlation matrix whose absolute value is less than the matrix correlation threshold, and generate a collaborative anomaly risk level based on the proportion of elements.

[0115] Specifically, the particle displacement (0.12-0.13 mm / min), seepage flow rate (0.45-0.47 cm / s), and interface pressure (118-122 kPa) data from the model simulation results are used to construct a 3×3 parameter correlation matrix (rows / columns for particle displacement, seepage flow rate, and interface pressure, respectively). The matrix elements are the Pearson correlation coefficients of the pairs of parameters (e.g., particle displacement-seepage flow rate correlation coefficient 0.82, seepage flow rate-interface pressure correlation coefficient 0.75, particle displacement-interface pressure correlation coefficient 0.68). The percentage of elements with absolute values ​​less than the matrix correlation threshold (0.3) in the parameter correlation matrix is ​​statistically analyzed: in this embodiment, there are no elements with absolute values ​​< 0.3, accounting for 0%. According to the risk level classification standard (percentage > 60% is low risk, 30%-60% is medium risk, and < 30% is high risk), the collaborative anomaly risk level is generated as "low risk".

[0116] S343. Verify and correct the parameter correlation matrix with particle displacement, seepage flow rate and interface pressure data, and integrate the synergy matching index, synergy anomaly risk level and the verified and corrected parameter correlation matrix to generate synergy assessment results and associate them with the corresponding direct shear apparatus operating condition number.

[0117] Specifically, the parameter correlation matrix is ​​verified and corrected with the interface pressure data (118-122 kPa) collected in real time by the direct shear apparatus: the correlation coefficient of "interface pressure-seepage flow" (0.75) in the matrix is ​​compared with the measured correlation coefficient (0.73). If the deviation is 2.7% ≤ ±10%, the verification is deemed successful; if the deviation is > ±10%, the model boundary conditions are readjusted (such as correcting the dynamic load amplitude), and the parameter correlation matrix is ​​recalculated; the synergy matching index (0.815), the synergy anomaly risk level (low risk), and the verified and corrected parameter correlation matrix are integrated to generate the synergy assessment result, which is associated with the corresponding direct shear apparatus working condition number (coastal sand-C30 concrete-1MPa seepage pressure-2Hz normal dynamic load) and stored in the "Assessment Result Subdirectory" of the soil particle seepage synergy tracking data repository for subsequent synchronous tracking scheme generation and calling.

[0118] S4. Preset multi-parameter synchronous tracking rules adapted to the hardware capabilities of the direct shear apparatus, adopt the particle seepage multi-parameter collaborative matching algorithm, combine the standardized particle seepage collaborative feature vector and collaborative evaluation results to obtain a synchronous visualization tracking scheme, and verify and optimize the accuracy of the synchronous visualization tracking scheme with historical synchronous tracking data;

[0119] As a preferred embodiment, the preset multi-parameter synchronous tracking rule adapted to the hardware capabilities of the direct shear apparatus adopts a particle seepage multi-parameter collaborative matching algorithm, combined with standardized particle seepage collaborative feature vectors and collaborative evaluation results, to obtain a synchronous visual tracking scheme. The synchronous visual tracking scheme is then verified and optimized against historical synchronous tracking data, including the following steps:

[0120] S41. Preset tracking parameter priority rules, data acquisition frequency adaptation rules, and visualization observation accuracy adaptation rules, and combine them to form multi-parameter synchronous tracking rules;

[0121] Specifically, the priority rules for tracking parameters are as follows: Based on the hardware acquisition capabilities of the PWS-Y200 / 150J direct shear meter, parameters are sorted in order of "core parameters > auxiliary parameters". Core parameters include particle displacement rate (tangential displacement sensor, wire type, range 0-150mm, resolution 0.001mm), seepage flow rate (flow meter, accuracy ±5%), and interface pressure (normal force sensor, spoke type, range 0-200kN, accuracy ±1%FS). Auxiliary parameters include visualization image frames (CCD high-speed camera, 79 frames / s) and pore water pressure (pore water pressure gauge, accuracy ±2%FS). Priority is given to ensuring the acquisition frequency and accuracy of core parameters (e.g., the acquisition frequency of core parameters is not lower than the system's maximum frequency of 5kHz), and auxiliary parameters are adapted to the timing of core parameters (e.g., image frame acquisition and particle displacement acquisition are triggered synchronously) to avoid data acquisition conflicts.

[0122] Data acquisition frequency adaptation rules: Strictly match the 5kHz acquisition frequency of the direct shear tester's computer control system. The core parameter acquisition frequency is set to 5kHz (synchronized with the system to ensure no data delay). Among the auxiliary parameters, the visualization image frame acquisition frequency is set to 79 frames / s (CCD hardware limit, 12.6ms exposure time per frame), and the pore water pressure acquisition frequency is set to 1kHz (balancing accuracy and storage cost, avoiding data redundancy). At the same time, a "dynamic frequency adjustment mechanism" is set: when the collaborative anomaly risk level is "high", the core parameter acquisition frequency remains at 5kHz (the highest in the system), and the pore water pressure acquisition frequency in the auxiliary parameters is increased to 2kHz to ensure the capture of details of abnormal states (such as sudden changes in seepage pressure).

[0123] Visualization observation accuracy adaptation rules: Based on the 0.01mm / pixel resolution of the CCD camera, a threshold for particle motion observation accuracy is set (the minimum displacement recognition is 0.01mm, achieved by analyzing image frames using DIC software); seepage path observation is indirectly identified through "image grayscale changes," with the grayscale value of the seepage area being 15-20 grayscale levels lower than that of the non-seepage area (18 grayscale levels lower in this embodiment); the observation window uses double-layer tempered glass (explosion-proof design, light transmittance ≥90%), coupled with a sealing gasket (2.5mm thick, fluororubber material) to prevent seepage leakage from affecting image clarity; the light source uses a uniform surface light source (intensity 800 lux) to avoid shadows on the soil sample contact surface, ensuring that the visualization data meets the requirements of "particle-seepage synchronous tracking."

[0124] S42. Input the standardized particle seepage collaborative feature vector, collaborative evaluation results and multi-parameter synchronous tracking rules into the particle seepage multi-parameter collaborative matching algorithm. Quantify the fit between the feature vector and the evaluation results through the particle seepage multi-parameter collaborative matching algorithm to generate an initial visual tracking scheme that includes data acquisition parameters, visual observation optimization parameters and anomaly warning thresholds.

[0125] Specifically, the application of the multi-parameter collaborative matching algorithm for particle seepage involves the following steps in the algorithm input layer: importing a standardized particle seepage collaborative feature vector (9-dimensional, labeled with the working conditions of "coastal sand - high permeability pressure - 2Hz dynamic load"), collaborative assessment results (collaborative matching index 0.815, low risk), and multi-parameter synchronous tracking rules; the feature vector needs to be labeled with the parameter deviation range (e.g., permeability deviation ±0.02MPa), and the assessment results need to be associated with the model calibration deviation (±4.4%).

[0126] Algorithm calculation layer: Initial parameters are generated through "feature vector-evaluation result fit calculation": Based on the cooperative matching index "excellent" (0.815≥0.75), the particle displacement acquisition accuracy is set to 0.01mm and the seepage water pressure control accuracy is set to ±0.05MPa; Based on the cooperative anomaly risk level "low risk", the anomaly warning threshold is set to "seepage flow fluctuation > 5%, interface pressure deviation > 1%FS, particle displacement rate deviation > 10%"; Based on the multi-parameter synchronous tracking rules, the core parameter acquisition frequency is set to 5kHz, the image frame frequency is 79 frames / s, and the pore water pressure acquisition frequency is set to 1kHz.

[0127] Initial scheme generation: Integrate the calculation results to form an initial visualization tracking scheme, which includes three parts: data acquisition parameters (particle displacement 5kHz / 0.01mm, seepage flow rate 5kHz / ±5%, interface pressure 5kHz / ±1%FS, pore water pressure 1kHz / ±2%FS); visualization observation optimization parameters (CCD exposure 12.6ms, light source 800lux, grayscale recognition threshold 18 levels); abnormal warning thresholds (seepage pressure fluctuation > ±0.05MPa, flow rate difference > 5%, data interruption > 10s); scheme label "adapted working condition number (coastal sand - C30 concrete - 1MPa seepage pressure - 2Hz normal dynamic load)" and "hardware dependent modules (normal actuator, CCD camera, seepage control system)".

[0128] S43. Retrieve historical synchronous tracking data from the soil particle seepage collaborative tracking data repository, compare the initial visualization tracking scheme with the historical data, and calculate the scheme deviation value of the initial visualization tracking scheme in terms of tracking accuracy and collaborative matching degree.

[0129] As a preferred embodiment, the step of retrieving historical synchronous tracking data from the soil particle seepage collaborative tracking data repository, comparing the initial visual tracking scheme with the historical data, and calculating the scheme deviation value of the initial visual tracking scheme in terms of tracking accuracy and collaborative matching degree includes the following steps:

[0130] S431. Filter the soil particle seepage collaborative tracking data in the soil particle seepage collaborative tracking data repository according to the working condition number and timestamp index rules to select historical synchronous tracking data, and integrate the historical synchronous tracking data to form a benchmark dataset.

[0131] Specifically, from the soil particle seepage collaborative tracking data repository, historical synchronous tracking data are filtered according to "complete match of working condition number plus 3-5 recent valid tests plus no abnormality label", core indicators such as particle displacement accuracy and seepage flow control deviation of historical schemes are extracted, and the average value is calculated as the "benchmark dataset".

[0132] S432. Construct a deviation calculation model based on the benchmark dataset in different dimensions, and input the initial visual tracking scheme and historical data into the deviation calculation model to calculate the tracking accuracy deviation value and the collaborative matching degree deviation value respectively.

[0133] Specifically, the initial scheme parameters are compared with the benchmark dataset item by item, and the tracking accuracy deviation (|initial parameter - benchmark mean| / benchmark mean×100%, such as particle displacement accuracy deviation ≤±5% is acceptable) and the coordination matching degree deviation (|initial scheme coupling coefficient - historical coupling coefficient| / historical coupling coefficient×100%, ≤10% is acceptable).

[0134] S433. Verify and adjust the tracking accuracy deviation value and the collaborative matching degree deviation value, and output the verified and adjusted tracking accuracy deviation value and collaborative matching degree deviation value as the initial visual tracking scheme's scheme deviation value in terms of tracking accuracy and collaborative matching degree.

[0135] Specifically, if the deviation exceeds the standard (e.g., seepage flow deviation > 5%), combined with the optimization of the collaborative abnormal risk level, the seepage parameters are adjusted first for high-risk working conditions (e.g., the seepage pressure control accuracy is increased from ±0.1MPa to ±0.05MPa); after adjustment, it is compared with historical data again until the advanced deviation value is ≤ the scheme deviation threshold, and finally the synchronous visualization tracking scheme is determined.

[0136] S44. Based on the deviation value of the scheme and the risk level of collaborative anomaly in the collaborative assessment results, adjust the collection frequency, visualization observation optimization parameters and early warning threshold in the initial scheme to form a synchronous visualization tracking scheme.

[0137] As a preferred option, the process of adjusting the initial sampling frequency, visualization observation optimization parameters, and early warning threshold based on the scheme deviation value and the risk level of collaboration anomaly in the collaboration assessment results to form a synchronous visualization tracking scheme includes the following steps:

[0138] S441. Preset the scheme deviation thresholds for the accuracy of the direct shear apparatus (tracking accuracy deviation ≤ 5%, coordination matching degree deviation ≤ 10%), compare the scheme deviation values ​​(tracking accuracy 4.5%, coordination matching degree 1.875%) with the scheme deviation thresholds, and determine whether the deviation exceeds the standard based on the scheme deviation comparison results, forming a preliminary adjustment scheme (only adjust the seepage flow warning threshold from ±5% to ±2.3%, and keep the other parameters unchanged);

[0139] S442. Re-compare the preliminary adjustment plan with historical synchronous tracking data and calculate the advanced deviation value.

[0140] If the advanced deviation value is less than or equal to the scheme deviation threshold, the initial adjustment scheme is defined as the synchronous visual tracking scheme.

[0141] If the advanced deviation value is still greater than the scheme deviation threshold, return to S441 to re-execute the adjustment strategy until the advanced deviation value is less than or equal to the scheme deviation threshold.

[0142] Specifically, the preliminary adjustment plan is recalculated by comparing it with historical synchronous tracking data: seepage flow deviation = |2.3-2.2| / 2.2×100% = 4.5%≤5%, coordination matching deviation = 1.875%≤10%. The advanced deviation value meets the plan deviation threshold. The preliminary adjustment plan is then defined as a synchronous visual tracking plan, marked with "Optimization Time (20251015-153000)" and "Adjustment Record (seepage flow warning threshold from ±5%-±2.3%)", and stored in the "Tracking Plan Subdirectory" of the soil particle seepage coordination tracking data repository. If the advanced deviation value is still greater than the plan deviation threshold (e.g., seepage flow deviation 6%>5%), the process returns to S441 to re-execute the adjustment strategy (e.g., continue to lower the warning threshold to ±2.2%) until the advanced deviation value is ≤ the plan deviation threshold.

[0143] S5. The verified and adjusted synchronous visualization tracking scheme is sent to the monitoring equipment of the direct shear machine to track the coordinated state of particle movement and seepage path at the soil contact surface in real time, and collect particle displacement rate, seepage flow rate, visualization image frame data and interface pressure change data.

[0144] Specifically, the scheme distribution path is as follows: Through the "Scheme Management Module" of the direct shear machine computer measurement and control system (ASUS industrial PC, equipped with two-channel measurement and control cards), the synchronous visualization tracking scheme (including data acquisition parameters, visualization parameters, and early warning thresholds) is distributed to each monitoring device via Ethernet, including the normal / tangential actuator (servo oil source system pressure 21MPa, German Rexroth constant pressure variable piston pump, flow rate 80L / min), water pressure seepage control system (water pump, pressure control valve, seepage pressure control accuracy ±0.05MPa), and CCD high-speed camera system; after distribution, through the "parameter feedback verification" mechanism, the parameter reception results fed back by each device are received (such as actuator feedback acquisition frequency 5kHz, CCD feedback frame rate 79 frames / s), confirming that the device received parameters are consistent with the scheme, with a deviation ≤ ±0.1%; if the deviation exceeds the tolerance (such as CCD feedback frame rate 75 frames / s), the scheme is re-distributed until the deviation is acceptable.

[0145] Equipment Readiness Check: Before the test, perform a full-process equipment check to ensure that the hardware status meets the tracking requirements: Shear box sealing status: Check that the fluororubber O-rings at the connection between the upper / lower shear box and the observation window are undamaged, and that the bolt tightening torque is 25 N·m (meeting the equipment acceptance standard). Pass the 1MPa seepage pressure holding test for 10 minutes to confirm that there is no leakage; CCD camera: Adjust the focus to the soil sample contact surface, with a resolution of 0.01 mm / pixel and a light source intensity of 800 lux, and take test frames (no shadows, clear particles); Normal / tangential actuators: Check the stroke (normal 200 mm, tangential 150 mm), manually test for no mechanical jamming, and ensure that the servo valve response delay is ≤10 ms; Seepage system pipeline: Check that the inlet sponge is not blocked (pore diameter ≤1 mm), the drainage channel is unobstructed, and flush the pipeline with clean water to ensure that there is no sediment residue.

[0146] Particle motion tracking: Particle displacement (mm, e.g., 0.125mm) is acquired in real time using a tangential displacement sensor (wire type, range 0-150mm, accuracy ±0.5%FS). The displacement rate (mm / min, e.g., 0.125mm / 1min=0.125mm / min) is calculated according to "displacement rate = displacement / time". Combined with image frames captured by a CCD camera (79 frames / s), the particle flipping and offset trajectory (e.g., maximum particle offset of 0.15mm) is analyzed using digital image processing technology (DIC software) to generate a displacement field cloud map (color rendering, the larger the displacement, the darker the color), ensuring that the macroscopic displacement (sensor acquisition) and microscopic motion (image analysis) data are synchronized, with a timestamp deviation ≤1ms.

[0147] Seepage path tracking: The seepage pressure (0-1MPa, e.g., 1.0±0.02MPa) and seepage flow rate (cm / s, e.g., 0.46cm / s) are collected in real time by the water pressure seepage control system. The difference in flow rate between the inlet and outlet is recorded (e.g., inlet 0.46cm / s, outlet 0.45cm / s, difference 2.2%≤5%, indicating stable seepage). The pore water pressure change inside the soil sample is collected in real time by the pore water pressure gauge (e.g., 85kPa→88kPa), and the seepage path distribution is indirectly inverted (the seepage conduction path corresponds to the abrupt change in pore water pressure, such as the area from 85kPa to 90kPa). At the same time, the seepage trajectory between particles is intuitively identified by the grayscale change of the image in the observation window (the grayscale value of the seepage area is 18 levels lower).

[0148] Interface pressure and image acquisition: Interface pressure (kPa, e.g., 120±1.2kPa) is acquired via a normal force sensor (spoke type, range 0-200kN, overload 150%) at a sampling frequency of 5kHz to ensure capture of pressure fluctuations under dynamic loads (e.g., a 2Hz sine wave causing pressure to fluctuate between 118-122kPa); A CCD camera acquires visual image frames at 79 frames / s, with each frame associated with a timestamp (20251015-160000-001) and a working condition number, stored in "TIFF" format to preserve details (1920×1080 pixels).

[0149] The normal force control employs a fully digital PID closed-loop control algorithm, with the core formula as follows:

[0150]

[0151] in, This is the output value of the controller at the k-th sampling time; It is a proportionality coefficient, which is proportional to the current error and determines the strength of the system's response to the current error; These are sampled values ​​of normal force or shear force. This is the sampled value of the previous normal force or shear force; The integral coefficient is obtained by accumulating historical errors. Eliminates static error, but may cause overshoot; These are the differential coefficients, based on the rate of change of error. Predict trends, suppress oscillations and overshoot; This represents the current normal force or the target value of the normal force. is the sampled value of the current normal force or shear force; P is the scaling factor, the scaling coefficient between the output and input difference signals; I is the integral factor, used for steady-state differences; D is the differential factor, used to improve the sensitivity of the system response or increase damping.

[0152] The system employs a dual-mode approach of wired transmission and local temporary storage. Data is transmitted in real-time to the computer monitoring and control system via Ethernet (100Mbps transmission rate), with a transmission delay of ≤1ms. Simultaneously, data is temporarily stored locally on each monitoring device: a 32GB memory card built into the CCD camera (stores image frames), a seepage system data buffer (stores seepage pressure / flow data), and an actuator controller buffer (stores force / displacement data), preventing data loss due to transmission interruptions. The temporarily stored data is named using "timestamp (accurate to milliseconds, e.g., 20251015160000001) + device number (e.g., CCD-01)," and a data packet (e.g., "20251015160000001-CCD-01.zip") is generated every 5 minutes. This packet contains particle displacement, seepage flow rate, interface pressure, and image frame index (recording the image storage path) for that time period, supporting subsequent data integration and anomaly tracing.

[0153] Temporary data is named with "timestamp (accurate to milliseconds) plus device number". A data packet is generated every 5 minutes. The packet contains particle displacement, seepage flow rate, interface pressure and image frame index for that period, supporting subsequent data integration and anomaly tracing.

[0154] S6. Preset tracking accuracy thresholds and anomaly judgment criteria that match the accuracy of the direct shear apparatus. Based on the tracking accuracy thresholds and anomaly judgment criteria, compare and analyze historical synchronous tracking data with real-time acquired data, add tags according to the comparison results, and store them in the soil particle seepage collaborative tracking data repository.

[0155] As a preferred embodiment, the preset tracking accuracy threshold and anomaly judgment criteria that match the accuracy of the direct shear apparatus, and the comparison and analysis of historical synchronous tracking data with real-time acquired data based on the tracking accuracy threshold and anomaly judgment criteria, and the addition of tags according to the comparison results, and the storage in the soil particle seepage collaborative tracking data repository, include the following steps:

[0156] S61. Based on the parameter measurement accuracy of the direct shear apparatus, preset multi-dimensional tracking accuracy thresholds, and based on the seepage shear coupling test scenario of the direct shear apparatus, preset anomaly judgment criteria.

[0157] Specifically, the tracking accuracy threshold settings are as follows: Fully matching the hardware accuracy and testing acceptance standards of the PWS-Y200 / 150J direct shear apparatus, the thresholds for each dimension are set at “equipment accuracy × 1.2” (leaving a reasonable error margin): Particle displacement accuracy threshold ±0.012mm (CCD resolution 0.01mm × 1.2); seepage water pressure accuracy threshold ±0.06MPa (seepage pressure control accuracy ±0.05MPa × 1.2); interface pressure accuracy threshold ±1.2%FS (force sensor accuracy ±1%FS × 1.2); visualization image accuracy threshold ±0.012mm / pixel (telecentric lens accuracy 0.01mm / pixel × 1.2); seepage flow rate accuracy threshold ±6% (flowmeter accuracy ±5% × 1.2).

[0158] Anomaly detection criteria are set in three scenarios, linked to the direct shear tester's fault log (such as servo oil supply system overload records and seepage system leakage alarms) to ensure that the judgment logic matches the actual operating status of the equipment: Seepage anomaly: seepage water pressure fluctuation > ±0.06MPa, inlet / outlet flow difference > 6%, interface dripping > 1 drop / min (monitored by the humidity sensor on the outside of the observation window); Load anomaly: normal pressure deviation > ±1.2%FS, tangential rate deviation > 6% (deviation > 0.006mm / min at 0.1mm / min), actuator temperature rise > 40℃ (monitored by the actuator's built-in temperature sensor); Data anomaly: particle displacement data interruption > 10s, image frame loss rate > 3% (number of frames lost per hour / total number of frames > 3%), no change in pore water pressure (no fluctuation for 5 minutes, excluding soil sample saturation).

[0159] S62. Compare the real-time collected data with the historical synchronous tracking data under the same working conditions point by point, calculate the relative error value of each dimension of data, and compare and judge the relative error value with the tracking accuracy threshold and the anomaly judgment standard.

[0160] Specifically, the comparison dimensions and methods are as follows: Comparison is performed by "parameter type" to ensure coverage of particles, seepage, and interfaces across all dimensions: Particle displacement: Calculated using "time-by-time relative error," the formula is "Relative error = |real-time displacement value - historical average value at the same time| / historical average value at the same time × 100%" (e.g., real-time displacement 0.125mm, historical average 0.12mm, relative error 4.17%); Seepage flow rate: Compared using "maximum value deviation within a time period," the formula is "Deviation = |real-time maximum value - historical maximum value| / historical maximum value × 100%" (e.g., real-time maximum value 0.125mm, historical average 0.125mm, relative error 4.17%). 47cm / s, historical maximum 0.46cm / s, deviation 2.17%); Interface pressure: compared using "fluctuation amplitude deviation", the formula is "deviation = |real-time fluctuation amplitude - historical fluctuation amplitude| / historical fluctuation amplitude × 100%" (e.g., real-time fluctuation amplitude 4kPa, historical fluctuation amplitude 3.8kPa, deviation 5.26%); Visualized images: compared using "particle displacement recognition deviation", the displacement difference of the same particle in real-time image and historical image is analyzed by DIC software (e.g., real-time recognition displacement 0.15mm, historical recognition 0.145mm, deviation 3.45%).

[0161] Comparison Result Classification: Based on the comparison results between the relative error value and the tracking accuracy threshold, the real-time data is divided into three levels: Normal data: relative error ≤ accuracy threshold (e.g., particle displacement error 4.17% ≤ the error range corresponding to the particle displacement accuracy threshold, seepage flow deviation 2.17% ≤ 6%); Warning data: error > accuracy threshold but ≤ 1.5 times the accuracy threshold (e.g., seepage flow deviation 7%, between 6% and 9%); Abnormal data: error > 1.5 times the accuracy threshold (e.g., seepage flow deviation 10% > 9%). In this embodiment, the real-time collected particle displacement error of 4.17%, seepage flow deviation of 2.17%, interface pressure deviation of 5.26%, and image recognition deviation of 3.45% are all ≤ the corresponding accuracy thresholds and are judged as "normal data".

[0162] S63. Add tags to the real-time collected data based on the comparison and judgment results, and record the real-time collected data when an anomaly occurs;

[0163] Specifically, the label classification and rules are as follows: A two-layer labeling system of "basic labels + extended labels" is adopted to ensure that the data attributes are clear and traceable: Basic labels: "Normal", "Warning", and "Abnormal" are added according to the comparison results ("Normal" is added in this embodiment); Abnormal type labels: Only abnormal / warning data are added, such as "Seepage flow deviation exceeds the standard" for seepage abnormality, "Actuator temperature rise is too high" for load abnormality, and "Image frame loss" for data abnormality; Working condition association labels: All data are added, in the format of "soil sample type-permeability level-load type" (such as "coastal sand-1MPa-2Hz dynamic load"); Warning treatment labels: Only warning data are added, with "warning parameters" and "suggested treatment measures" marked (such as "seepage flow deviation 7%, it is recommended to check whether the inlet sponge is blocked").

[0164] Abnormal Data Recording: If abnormal data exists, detailed information about the occurrence of the abnormality must be recorded: abnormal time (accurate to the second, e.g., 20251015162030); abnormal parameter value (e.g., real-time flow rate of 0.51 cm / s when the seepage flow rate deviation is 10%); equipment status (e.g., actuator temperature of 38℃ and servo oil source pressure of 21 MPa during the abnormality); on-site handling record (e.g., whether the test was suspended or whether the sponge was replaced); In this embodiment, there is no abnormal data, only the collection time and parameter range of normal data are recorded.

[0165] S64. Store the tagged real-time collected data into the soil particle seepage collaborative tracking data repository according to the indexing rules of timestamp and working condition number.

[0166] Specifically, the storage architecture and paths are as follows: A distributed storage architecture is adopted, with storage paths divided according to "operating condition number - test timestamp - data type", such as "Coastal Sand - C30 Concrete - 1MPa Seepage Pressure - 2Hz Normal Dynamic Load / 20251015-160000 / Particle Displacement Data", and "Coastal Sand - C30 Concrete - 1MPa Seepage Pressure - 2Hz Normal Dynamic Load / 20251015-160000 / Visualization Images"; numerical data (particle displacement, seepage flow rate, interface pressure) are stored in "CSV format". The format includes fields such as "timestamp (ms), parameter value, label, error range, and acquisition device number" (e.g., "20251015160000001, 0.125mm, normal, ±0.001mm, displacement sensor-01"); the visualized image is stored in "TIFF format", and the image file name includes "operating condition number-timestamp-label-frame number" (e.g., "Coastal Sand-C30-1MPa-2Hz-20251015160000001-normal-001.tif").

[0167] Data Indexing and Backup: Establish a "tag-data" related index to support multi-dimensional fast queries (such as querying "October 15, 2025 - coastal sand - normal data" and "abnormal seepage - 2Hz dynamic load data"); at the same time, perform dual backups: local backup: back up the data of the day to the industrial machine's spare hard drive (2TB) at 24:00 every day; cloud backup: incremental backup to Alibaba Cloud server every hour, and perform a full backup every Sunday. Backup data is retained for 6 months to ensure data security and traceability.

[0168] Data reuse support: The data storage interface can be directly used for "scheme deviation comparison" (such as retrieving the current data as a historical benchmark in the next test) and "model parameter calibration" (such as correcting the particle contact stiffness of DEM-LBM) in subsequent tests under the same working conditions. It also supports the automatic generation of test reports (extracting CSV data to generate trend charts and image data to generate seepage path diagrams), providing data support for the study of the mechanical properties of soil contact surfaces.

[0169] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for synchronously visually tracking particle movement and seepage path at soil contact surfaces, characterized in that, Includes the following steps: S1. Obtain the soil contact surface particle foundation parameters, seepage medium property data and historical synchronous tracking data under the same working conditions that can be collected by the direct shear apparatus, and set up a soil particle seepage collaborative tracking data repository. S2. Extract key feature parameters from the particle foundation parameters of the soil contact surface, and pre-set contact surface feature mapping rules that match the direct shear test scenario. Based on the contact surface feature mapping rules, transform the key feature parameters into standardized particle seepage collaborative feature vectors. The process of extracting key feature parameters from the particle foundation parameters of the soil contact surface, and pre-setting contact surface feature mapping rules that match the direct shear test scenario, and converting the key feature parameters into standardized particle seepage collaborative feature vectors according to the contact surface feature mapping rules, includes the following steps: S21. Extract the particle geometric and mechanical characteristic parameters, seepage medium physical characteristic parameters, and contact surface mechanical and structural characteristic parameters from the particle foundation parameters of the soil contact surface as three types of key characteristic parameters. S22. Preset classification mapping rules, dimension normalization mapping rules and synergistic correlation mapping rules. The classification mapping rules correspond the three key feature parameters to the particle motion influence dimension, seepage path influence dimension and synergistic correlation dimension respectively. The dimension normalization mapping rules unify the parameter dimensions. At the same time, the synergistic correlation mapping rules quantify the synergistic weights between parameters. S23. Based on the three types of key feature parameters after mapping, construct an N-dimensional feature vector, perform consistency verification, add working condition labels and collaborative association weight labels to form a standardized particle seepage collaborative feature vector, and store it in the soil particle seepage collaborative tracking data repository. S3. Based on the soil contact surface particle foundation parameters and seepage medium property data, and combined with the seepage shear coupling function of the direct shear apparatus, a multi-dimensional synergy evaluation weighting rule is set, and a DEM-LBM coupling architecture model calibrated by the direct shear apparatus is used to evaluate the synergy status of particle movement and seepage path, and generate synergy evaluation results. S4. Preset multi-parameter synchronous tracking rules adapted to the hardware capabilities of the direct shear apparatus, adopt the particle seepage multi-parameter collaborative matching algorithm, combine the standardized particle seepage collaborative feature vector and collaborative evaluation results to obtain a synchronous visualization tracking scheme, and verify and optimize the accuracy of the synchronous visualization tracking scheme with historical synchronous tracking data; S5. The verified and adjusted synchronous visualization tracking scheme is sent to the monitoring equipment of the direct shear machine to track the coordinated state of particle movement and seepage path at the soil contact surface in real time, and collect particle displacement rate, seepage flow rate, visualization image frame data and interface pressure change data. S6. Preset tracking accuracy thresholds and anomaly judgment criteria that match the accuracy of the direct shear apparatus. Based on the tracking accuracy thresholds and anomaly judgment criteria, compare and analyze historical synchronous tracking data with real-time acquired data, add tags according to the comparison results, and store them in the soil particle seepage collaborative tracking data repository.

2. The method for synchronously visually tracking the particle movement and seepage path at the soil contact surface according to claim 1, characterized in that, The method involves setting multi-dimensional synergistic evaluation weighting rules based on soil contact surface particle foundation parameters and seepage medium property data, combined with the seepage-shear coupling function of the direct shear apparatus, and using a DEM-LBM coupled architecture model calibrated under direct shear apparatus operating conditions to evaluate the synergistic state of particle movement and seepage path, generating synergistic evaluation results, including the following steps: S31. Based on the particle foundation parameters of the soil contact surface and the property data of the seepage medium, set the influence weight of particle movement, the influence weight of seepage path, and the synergistic correlation weight of the influence weight of particle movement and the influence weight of seepage path, and combine them to form a multi-dimensional synergistic evaluation weighting rule. S32. Input the soil contact surface particle foundation parameters, seepage medium property data and multi-dimensional synergy evaluation weighting rules into the DEM-LBM coupled architecture model. Simulate the interaction process between particle movement and seepage field through the DEM-LBM coupled architecture model, and quantify the particle matching degree of particle displacement and seepage flow rate, the porosity and seepage path porosity correlation and the interface mechanical properties on the interface influence value of the synergy state. S33. Verify the consistency of particle matching degree, pore correlation and interface influence value, and integrate the verified particle matching degree, pore correlation and interface influence value to form the model simulation results; S34. Extract the collaborative state characteristic indicators from the model simulation results, and generate a collaborative evaluation result that includes the collaborative matching index, the collaborative anomaly risk level, and the parameter correlation matrix.

3. The method for synchronously visually tracking the particle movement and seepage path at the soil contact surface according to claim 1, characterized in that, The preset multi-parameter synchronous tracking rules adapted to the hardware capabilities of the direct shear apparatus employ a particle seepage multi-parameter collaborative matching algorithm, combined with standardized particle seepage collaborative feature vectors and collaborative evaluation results, to obtain a synchronous visual tracking scheme. The synchronous visual tracking scheme is then verified and optimized against historical synchronous tracking data, including the following steps: S41. Preset tracking parameter priority rules, data acquisition frequency adaptation rules, and visualization observation accuracy adaptation rules, and combine them to form multi-parameter synchronous tracking rules; S42. Input the standardized particle seepage collaborative feature vector, collaborative evaluation results and multi-parameter synchronous tracking rules into the particle seepage multi-parameter collaborative matching algorithm. Quantify the fit between the feature vector and the evaluation results through the particle seepage multi-parameter collaborative matching algorithm to generate an initial visual tracking scheme that includes data acquisition parameters, visual observation optimization parameters and anomaly warning thresholds. S43. Retrieve historical synchronous tracking data from the soil particle seepage collaborative tracking data repository, compare the initial visualization tracking scheme with the historical data, and calculate the scheme deviation value of the initial visualization tracking scheme in terms of tracking accuracy and collaborative matching degree. S44. Based on the deviation value of the scheme and the risk level of collaborative anomaly in the collaborative assessment results, adjust the collection frequency, visualization observation optimization parameters and early warning threshold in the initial scheme to form a synchronous visualization tracking scheme.

4. The method for synchronously visually tracking the particle movement and seepage path at the soil contact surface according to claim 1, characterized in that, The preset tracking accuracy threshold and anomaly judgment criteria, which match the accuracy of the direct shear apparatus, are used to compare and analyze historical synchronous tracking data with real-time acquired data. Tags are added based on the comparison results, and the data is stored in the soil particle seepage collaborative tracking data repository. This process includes the following steps: S61. Based on the parameter measurement accuracy of the direct shear apparatus, preset multi-dimensional tracking accuracy thresholds, and based on the seepage shear coupling test scenario of the direct shear apparatus, preset anomaly judgment criteria. S62. Compare the real-time collected data with the historical synchronous tracking data under the same working conditions point by point, calculate the relative error value of each dimension of data, and compare and judge the relative error value with the tracking accuracy threshold and the anomaly judgment standard. S63. Add tags to the real-time collected data based on the comparison and judgment results, and record the real-time collected data when an anomaly occurs; S64. Store the tagged real-time collected data into the soil particle seepage collaborative tracking data repository according to the indexing rules of timestamp and working condition number.

5. The method for synchronously visually tracking the particle movement and seepage path at the soil contact surface according to claim 1, characterized in that, The process of constructing an N-dimensional feature vector based on the three types of key feature parameters after mapping, performing consistency verification, adding working condition labels and collaborative association weight labels, forming a standardized particle seepage collaborative feature vector, and storing it in the soil particle seepage collaborative tracking data repository includes the following steps: S231. Arrange the particle motion influence dimension parameter, seepage path influence dimension parameter, and synergistic effect correlation dimension parameter in the three types of key feature parameters after mapping processing in the order of particles, seepage and synergy to construct an N-dimensional feature vector. S232. Preset the measurement accuracy range of the direct shear apparatus parameters, compare the deviation values ​​of each parameter in the N-dimensional feature vector with the particle foundation parameters of the soil contact surface, and compare the deviation values ​​with the measurement accuracy range of the direct shear apparatus parameters. If the deviation value exceeds the measurement accuracy range of the direct shear apparatus parameters, return to S22 to readjust the feature mapping rules. If the deviation value is within the measurement accuracy range of the direct shear apparatus parameters, retain the N-dimensional feature vector unchanged. S233. Generate working condition labels based on three types of key feature parameters, generate collaborative association weight labels by combining collaborative association weights, and associate the working condition labels and collaborative association weight labels with N-dimensional feature vectors. S234. Define the labeled N-dimensional feature vector as a standardized particle seepage collaborative feature vector, and store it in the soil particle seepage collaborative tracking data repository according to the indexing rules of the direct shear instrument condition number and generation time in the historical synchronous tracking data of the same working condition.

6. The method for synchronously visually tracking the particle movement and seepage path at the soil contact surface according to claim 2, characterized in that, The process of verifying the consistency of particle matching degree, porosity correlation, and interface influence values, and integrating the verified particle matching degree, porosity correlation, and interface influence values ​​to form the model simulation results includes the following steps: S331, preset particle matching degree qualification threshold, pore association qualification threshold and interface influence value qualification threshold; S332. Compare the particle matching degree with the qualified threshold of particle matching degree, the porosity correlation with the qualified threshold of porosity correlation, and the interface influence value with the qualified threshold of interface influence value. Determine whether the particle matching degree, porosity correlation, and interface influence value meet the qualified thresholds of particle matching degree, porosity correlation, and interface influence value. If any one or more of them do not meet the threshold requirements, return to S32 to adjust the simulation boundary conditions of the DEM-LBM coupled architecture model. If all of them meet the threshold requirements, the consistency verification is deemed to have passed. S333. Following the dimensional order of particle matching degree, porosity correlation, and interface influence value, the verified particle matching degree, porosity correlation, and interface influence value are integrated into multidimensional structural data, and the model simulation timestamp and direct shear apparatus condition number are labeled to form the model simulation results.

7. The method for synchronously visually tracking the particle movement and seepage path at the soil contact surface according to claim 2, characterized in that, The process of extracting collaborative state characteristic indicators from the model simulation results and generating collaborative evaluation results including a collaborative matching index, collaborative anomaly risk level, and parameter correlation matrix includes the following steps: S341. Preset simulation result weights and matrix correlation thresholds, extract particle matching degree and porosity correlation from model simulation results, calculate the weighted average of particle matching degree and porosity correlation according to simulation result weights, and define it as the cooperative matching index; S342. Based on the particle displacement, seepage flow rate and interface pressure data in the model simulation results, construct a parameter correlation matrix, count the proportion of elements in the parameter correlation matrix whose absolute value is less than the matrix correlation threshold, and generate a collaborative anomaly risk level based on the proportion of elements. S343. Verify and correct the parameter correlation matrix with particle displacement, seepage flow rate and interface pressure data, and integrate the synergy matching index, synergy anomaly risk level and the verified and corrected parameter correlation matrix to generate synergy assessment results and associate them with the corresponding direct shear apparatus operating condition number.

8. The method for synchronously visually tracking the particle movement and seepage path at the soil contact surface according to claim 3, characterized in that, The steps involved in retrieving historical synchronous tracking data from the soil particle seepage collaborative tracking data repository, comparing the initial visual tracking scheme with the historical data, and calculating the scheme deviation value of the initial visual tracking scheme in terms of tracking accuracy and collaborative matching degree. S431. Filter the soil particle seepage collaborative tracking data in the soil particle seepage collaborative tracking data repository according to the working condition number and timestamp index rules to select historical synchronous tracking data, and integrate the historical synchronous tracking data to form a benchmark dataset. S432. Construct a deviation calculation model based on the benchmark dataset in different dimensions, and input the initial visual tracking scheme and historical data into the deviation calculation model to calculate the tracking accuracy deviation value and the collaborative matching degree deviation value respectively. S433. Verify and adjust the tracking accuracy deviation value and the collaborative matching degree deviation value, and output the verified and adjusted tracking accuracy deviation value and collaborative matching degree deviation value as the initial visual tracking scheme's scheme deviation value in terms of tracking accuracy and collaborative matching degree.

9. The method for synchronously visually tracking the particle movement and seepage path at the soil contact surface according to claim 3, characterized in that, The process of adjusting the initial sampling frequency, visualization observation optimization parameters, and early warning threshold based on the scheme deviation value and the coordination anomaly risk level in the coordination assessment results to form a synchronous visualization tracking scheme includes the following steps: S441. Preset the scheme deviation threshold to match the accuracy of the direct shearing instrument, compare the scheme deviation value with the scheme deviation threshold, and determine whether the deviation exceeds the standard based on the scheme deviation comparison result, and form a preliminary adjustment scheme. S442. Re-compare the preliminary adjustment plan with historical synchronous tracking data and calculate the advanced deviation value. If the advanced deviation value is less than or equal to the scheme deviation threshold, the initial adjustment scheme is defined as the synchronous visual tracking scheme. If the advanced deviation value is still greater than the scheme deviation threshold, return to S441 to re-execute the adjustment strategy until the advanced deviation value is less than or equal to the scheme deviation threshold.

Images