An indoor test visualization method for pile foundation mud skin mixing ratio test

Abstract

This invention relates to the field of materials testing technology, specifically to an indoor visualization method for testing the mix proportion of mud cake for pile foundations, comprising the following steps: S1: Based on the geological report of the Weiyi Expressway project, bentonite and PHP dispersant are selected as slurry materials, and a simulation device with adjustable confining pressure and groundwater infiltration rate is configured. The mud slurry ratio, formation pressure, and groundwater conditions are set as variables. This invention achieves dynamic optimization of the mud slurry ratio through response surface modeling combined with sensing and image recognition, improving the accuracy and real-time performance of the ratio. Furthermore, by leveraging multi-technology fusion monitoring and early warning, it ensures the comprehensiveness of borehole wall stability monitoring and the timeliness of risk prediction. This achieves efficient linkage between model construction, dynamic adjustment, risk management, and intelligent optimization in pile foundation mud cake mix proportion testing, significantly enhancing the scientific rigor and engineering application value of the experiment.

Description

Technical Field

[0001] This invention relates to the field of materials testing technology, specifically to an indoor test visualization method for testing the mix proportion of mud cake in pile foundations. Background Technology

[0002] In the construction of expressways in mountainous areas, pile foundations are a crucial foundation for bridge engineering, and their construction quality directly determines structural safety. Taking the Weixin-Yiliang Expressway in Yunnan Province as an example, this section traverses complex geological areas such as karst development zones and sand-clay cross-sections. The geological conditions are extremely heterogeneous, with prominent issues such as well-developed karst fissures and loose, easily collapsing sand layers. During the pile foundation drilling process, accidents such as borehole collapse and diameter reduction are highly likely due to borehole wall instability. The mud cake formed by the mud slurry acts as a "protective layer" for borehole wall stability. Its properties (such as viscosity, filtration loss, and density) are closely related to the mix ratio and directly affect the wall protection effect. High-quality mud cake can effectively block groundwater seepage and balance formation pressure, while low-quality mud cake can lead to borehole wall erosion and mud cake peeling, thereby causing construction risks.

[0003] However, in the existing technology, the mud cake mix ratio test for pile foundation has problems such as insufficient accuracy due to reliance on experience, one-sided monitoring of borehole wall stability and delayed risk warning, disconnect between testing and engineering and lack of intelligent optimization. It is difficult to meet the stringent requirements for mud cake performance under complex geological conditions, which restricts the safety and efficiency of pile foundation construction in mountainous highways.

[0004] Based on this, the present invention provides an indoor test visualization method for testing the mud cake mix ratio of pile foundations, in order to solve the above-mentioned technical problems. Summary of the Invention

[0005] The purpose of this invention is to provide an indoor visualization method for testing the mud cake mix proportion of pile foundations. This invention uses a response surface model combined with sensing and image recognition to achieve dynamic optimization of mud cake mix proportions, improving the accuracy and real-time performance of the mix proportions. Furthermore, by leveraging multi-technology fusion for monitoring and early warning, it ensures the comprehensiveness of borehole wall stability monitoring and the timeliness of risk prediction. By utilizing the model and digital twin platform to recommend and provide feedback on the optimal mix proportions, it enhances the intelligence and engineering adaptability of the mix proportion optimization. Thus, it achieves efficient linkage from model construction and dynamic adjustment to risk management and intelligent optimization in pile foundation mud cake mix proportion testing, significantly improving the scientific rigor and engineering application value of the test.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] This invention provides an indoor experimental visualization method for testing the mix proportion of mud cake in pile foundations, comprising the following steps:

[0008] S1: Based on the geological report of the Weiyi Expressway project, bentonite and PHP dispersant were selected as slurry materials, and a simulation device with adjustable confining pressure and groundwater infiltration rate was configured. The mud ratio, formation pressure and groundwater conditions were set as variables.

[0009] S2: A mud performance ratio model was established using the response surface methodology, and the ratio was dynamically adjusted by combining fiber optic sensors and microscopic image recognition technology.

[0010] S3: Real-time monitoring of aperture changes using ultrasonic pulse echo method, combined with CT scanning and acoustic emission technology to reconstruct mud skin structure in three dimensions and provide early warning of hole collapse risk;

[0011] S4: Train a random forest model based on historical data, develop a digital twin platform to recommend the optimal ratio, and feed it back to the S2 stage in real time;

[0012] S5: Verify the mix proportions in typical geological sections of the Weiyi Expressway, compare the filling coefficient and the borehole qualification rate, and generate a standardized visual report.

[0013] In step S1, based on the geological report of the Weiyi Expressway project, bentonite and PHP dispersant were selected as slurry materials. A simulation device with adjustable confining pressure and groundwater infiltration rate was configured, and the mud ratio, formation pressure, and groundwater conditions were set as variables. The specific steps are as follows:

[0014] S1.1 Based on the geological survey report of the Weiyi Expressway project, bentonite was selected as the main slurry material, and PHP dispersant was selected as an additive.

[0015] S1.2 Build an indoor test simulation device that can precisely adjust the confining pressure and groundwater seepage rate;

[0016] S1.3. Clearly define mud mix ratio, formation pressure, and groundwater conditions as key variables in the experiment.

[0017] In step S2, a mud performance ratio model is established using the response surface methodology, and the ratio is dynamically adjusted by combining fiber optic sensors and microscopic image recognition technology. The specific steps are as follows:

[0018] S2.1 Response surface methodology was used to design experiments and establish a nonlinear regression model between the mud mix ratio and the performance parameters of viscosity, density, and filtration loss.

[0019] S2.2. Collect mud density and rheological parameters through fiber optic sensors, and analyze the microscopic dispersion state of mud using microscopic image recognition technology.

[0020] S2.3. Based on monitoring data and model output results, optimize the ratio of bentonite to PHP dispersant in the mud in real time.

[0021] The specific steps of S2.3 are as follows:

[0022] S2.3.1 Data Comparison and Deviation Grading: The real-time density, rheological parameters, and dispersion coefficient of the microscopic image recognition collected by the fiber optic sensor in S2.2 are compared with the predicted values ​​of the regression model in S2.1. The deviation rate of each parameter is calculated and divided into three levels: deviation rate ≤ 10% is qualified, 10% < deviation rate ≤ 20% is slightly exceeding the standard, and deviation rate > 20% is seriously exceeding the standard.

[0023] S2.3.2, Grading Adjustment Rules:

[0024] ① If density > 1.25 g / cm³ 3 The level is slightly excessive; for every 1% reduction in bentonite usage, the density can be reduced by 0.02 g / cm³. 3 The proportion was slightly adjusted while keeping the proportion of PHP dispersant unchanged.

[0025] ② If the filtration loss is >15mL / 30min, it is considered a serious overdose. Under the premise that the amount of bentonite remains unchanged, the filtration loss can be reduced by 2mL / 30min for every 0.3% increase in PHP dispersant.

[0026] ③ If the microscopic image shows a dispersion coefficient < 0.85, increase the amount of PHP dispersant by 0.1%-0.2%. If the dispersion coefficient does not improve after 3 minutes, reduce the amount of bentonite by 0.5%.

[0027] S2.3.3 Closed-loop verification and iteration: After adjustment, recollect data after 5 minutes. If the deviation rate of all parameters is ≤10%, lock the current ratio. If there are still parameters exceeding the standard, repeat steps S2.3.1-S2.3.2 until two consecutive monitoring results meet the standard.

[0028] In step S3, the pore size change is monitored in real time using the ultrasonic pulse echo method. Combined with CT scanning and acoustic emission technology, the mud cake structure is reconstructed in three dimensions, and the risk of pore collapse is warned. The specific steps are as follows:

[0029] S3.1. The ultrasonic pulse echo method is used to capture the aperture change data in real time during the hole forming process;

[0030] S3.2. Obtain the microstructure information of mud skin through CT scanning, and superimpose the stress release signal monitored by acoustic emission technology to reconstruct the three-dimensional structure of mud skin.

[0031] S3.3 Based on the pore size variation trend and mud cake structure characteristics, comprehensively analyze and warn of the risk of pore wall collapse.

[0032] The specific steps of S3.2 are as follows:

[0033] S3.2.1 CT Scan Parameters:

[0034] ① A microfocus CT scanner was used to acquire cross-sectional and longitudinal images of the mud skin at a frequency of 1 minute / time;

[0035] ② The porosity and thickness distribution of the mud cake were extracted using a threshold segmentation algorithm;

[0036] S3.2.2 Acoustic emission signal processing:

[0037] ① Deploy broadband acoustic emission sensors on the side wall of the simulation device to capture mud skin rupture signals in real time;

[0038] ② Identify precursor signals of hole collapse by analyzing the frequency domain characteristics of the signal based on wavelet transform;

[0039] S3.2.3, Three-dimensional fusion reconstruction:

[0040] ① Register the CT structural data with the acoustic emission localization results;

[0041] ② Construct a three-dimensional model of mud crust that includes pore distribution and stress field changes. When the porosity is >25% and accompanied by high-frequency acoustic emission signals, trigger a first-level pore collapse warning.

[0042] The specific steps of S3.3 are as follows:

[0043] S3.3.1 Multi-dimensional feature extraction:

[0044] The pore size shrinkage rate and pore size fluctuation amplitude are extracted from the pore size variation data in S3.1; the average thickness of the mud skin, the proportion of local damaged area and the energy rate of the acoustic emission signal are extracted from the three-dimensional mud skin structure in S3.2, and a three-dimensional feature matrix of "pore size dynamic characteristics - mud skin structure parameters - acoustic emission energy" is established.

[0045] S3.3.2 Risk Level Classification and Threshold Setting:

[0046] Low risk: pore size shrinkage rate ≤0.5mm / min, fluctuation range ≤5%; mud cake thickness ≥3mm and damaged area ≤3%; acoustic emission energy rate ≤10μJ / s;

[0047] Medium risk: pore size shrinkage rate 0.5-1 mm / min, fluctuation range 5%-10%; mud cake thickness 2-3 mm or damaged area percentage 3%-10%; acoustic emission energy rate 10-50 μJ / s;

[0048] High risk: pore shrinkage rate > 1 mm / min, fluctuation range > 10%; mud cake thickness < 2 mm or damaged area percentage > 10%; acoustic emission energy rate > 50 μJ / s;

[0049] S3.3.3 Integrated Early Warning and Response Mechanism:

[0050] When any two indicators in the three-dimensional feature matrix reach the medium-risk threshold, an early warning is triggered, and it is recommended to suspend the test and check the mud performance.

[0051] If any indicator reaches the high-risk threshold, an emergency warning will be triggered immediately, and the current mud skin structure image and borehole diameter data will be automatically recorded. At the same time, the borehole mud pressure compensation program will be started until the risk level drops to the low-risk range.

[0052] The S4 stage trains a random forest model based on historical data, develops a digital twin platform to recommend the optimal ratio, and feeds back to the S2 stage in real time. The specific steps are as follows:

[0053] S4.1 Collect historical data on mud ratio, performance parameters, and filling coefficient, and train a random forest model to establish the correlation between mud ratio and hole formation quality;

[0054] S4.2 Build a digital twin platform to map the physical test process with virtual simulation and make a visual prediction of the mud mixing effect;

[0055] S4.3 Output the optimal mixing ratio through the digital twin platform and feed it back to the S2 stage in real time to support the dynamic adjustment of the mud mixing ratio.

[0056] The specific steps in S4.2 are as follows:

[0057] S4.2.1 Data Interface Layer Construction:

[0058] The OPC UA protocol allows real-time access to physical test data on mud ratio and performance parameters, as well as sensor data on pore size and mud cake thickness.

[0059] Establish data cleaning rules to remove noisy data;

[0060] S4.2.2 Virtual Model Layer Development:

[0061] A 3D geological-pile borehole model was built using the Unity3D engine to reproduce the complex geological conditions in mountainous areas.

[0062] Integrate the random forest prediction model and provide allocation optimization services via API interface;

[0063] S4.2.3 Virtual-Real Mapping and Verification:

[0064] Set the synchronization period between physical experiments and virtual simulations;

[0065] When the deviation of the virtual prediction filling coefficient is greater than 15%, a physical test parameter calibration command is triggered.

[0066] In S5, the mix design was verified in a typical geological section of the Weiyi Expressway. The filling coefficient and the borehole qualification rate were compared to generate a standardized visual report. The specific steps are as follows:

[0067] S5.1. Select a representative typical stratum section in the Weiyi Expressway and verify the pile foundation construction using the optimized mud mix ratio.

[0068] S5.2 Compare and analyze the deviation between the actual filling coefficient and the design value during on-site construction, calculate the hole-forming qualification rate, and evaluate the applicability and reliability of the mix proportion.

[0069] S5.3 Integrate experimental data, verification results, and analysis conclusions to form a standardized report that includes visual charts.

[0070] Compared with the prior art, the beneficial effects of the present invention are:

[0071] This invention achieves dynamic optimization of mud mix proportions by combining response surface methodology with sensing and image recognition, improving the accuracy and real-time performance of the mix proportions. Furthermore, by leveraging multi-technology fusion for monitoring and early warning, it ensures comprehensive monitoring of borehole wall stability and timely risk prediction. By utilizing models and digital twin platforms to recommend and provide feedback on the optimal mix proportions, it enhances the intelligence and engineering adaptability of the mix proportion optimization. This enables efficient linkage between model construction, dynamic adjustment, risk management, and intelligent optimization in pile foundation mud cake mix proportion testing, significantly improving the scientific rigor and engineering application value of the experiment. Attached Figure Description

[0072] Figure 1 This is a flowchart of an indoor test visualization method for testing the mud cake mix ratio of pile foundations according to the present invention.

[0073] Figure 2 This is a flowchart of the multimodal detection process in an indoor test visualization method for testing the mud cake mix ratio of pile foundations, as described in this invention.

[0074] Figure 3 This is a flowchart of the digital twin platform in an indoor test visualization method for testing the mud cake mix ratio of pile foundations according to the present invention. Detailed Implementation

[0075] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0076] Example:

[0077] like Figures 1-3As shown in the figure, this embodiment provides an indoor test visualization method for testing the mud cake mix ratio of pile foundations, including the following steps: S1: Based on the geological report of the Weiyi Expressway project, bentonite and PHP dispersant are selected as slurry materials, and a simulation device with adjustable confining pressure and groundwater permeability rate is configured. The mud cake mix ratio, formation pressure, and groundwater conditions are set as variables; S2: A mud cake performance mix ratio model is established using the response surface methodology, and the mix ratio is dynamically adjusted by combining fiber optic sensors and microscopic image recognition technology; S3: The pore size change is monitored in real time by ultrasonic pulse echo method, and the mud cake structure is reconstructed in three dimensions by combining CT scanning and acoustic emission technology to provide early warning of the risk of borehole collapse; S4: A random forest model is trained based on historical data, a digital twin platform is developed to recommend the optimal mix ratio, and the results are fed back to stage S2 in real time; S5: The mix ratio is verified in a typical stratum section of the Weiyi Expressway, the filling coefficient and borehole qualification rate are compared, and a standardized visualization report is generated.

[0078] In this embodiment, it should also be noted that, in S1, based on the geological report of the Weiyi Expressway project, bentonite and PHP dispersant are selected as slurry materials, and a simulation device that can adjust the confining pressure and groundwater infiltration rate is configured. The mud ratio, formation pressure, and groundwater conditions are set as variables. The specific steps are as follows: S1.1, Based on the geological survey report of the Weiyi Expressway project, bentonite is determined to be the main slurry material, and PHP dispersant is selected as an additive; S1.2, An indoor test simulation device that can accurately adjust the confining pressure and groundwater infiltration rate is built; S1.3, The mud ratio, formation pressure, and groundwater conditions are clearly set as key variables in the test.

[0079] Furthermore, it should be noted that bentonite was selected as the basic mud-making material (colloidal content ≥95%, montmorillonite content ≥70%), combined with PHP dispersant (molecular weight 2-3 million, degree of hydrolysis 30%-40%) to improve mud dispersibility and filtration control. The simulation device uses a steel pressure chamber, with confining pressure adjusted via a hydraulic servo system (range 0.1-1.0 MPa, accuracy ±0.01 MPa), and groundwater infiltration rate controlled by a constant pressure water supply device (0-50 m / d continuously adjustable) to reproduce the stress state and hydrological conditions of different strata. Key variables were set as follows: mud mix ratio (bentonite 5%-10%, PHP dispersant 0.1%-0.5%), formation pressure (0.3-0.8 MPa), and groundwater infiltration rate (10-40 m / d).

[0080] In this embodiment, it should also be noted that in S2, a response surface methodology is used to establish a mud performance ratio model, and the ratio is dynamically adjusted by combining fiber optic sensors and microscopic image recognition technology. The specific steps are as follows: S2.1, Design an experiment using the response surface methodology to establish a nonlinear regression model between the mud ratio and the performance parameters of viscosity, density, and filtration loss; S2.2, Collect mud density and rheological parameters through fiber optic sensors, and analyze the microscopic dispersion state of the mud using microscopic image recognition technology; S2.3, Optimize the ratio of bentonite and PHP dispersant in the mud in real time based on the monitoring data and model output results. The specific steps are as follows: S2.3.1, Data Comparison and Deviation Grading: Compare the real-time density, rheological parameters, and dispersion coefficient of the microscopic image recognition collected by the fiber optic sensor in S2.2 with the predicted values ​​of the regression model in S2.1, calculate the deviation rate of each parameter, and divide them into three levels: deviation rate ≤ 10% is qualified, 10% < deviation rate ≤ 20% is slightly exceeding the standard, and deviation rate > 20% is seriously exceeding the standard; S2.3.2, Grading Adjustment Rules: ① If density > 1.25 g / cm³ 3 The level is slightly excessive; for every 1% reduction in bentonite usage, the density can be reduced by 0.02 g / cm³. 3 ① Fine-tune the ratio while keeping the PHP dispersant ratio constant; ② If the filtration loss is >15mL / 30min, it is considered a serious overshoot. Under the premise of keeping the bentonite dosage unchanged, add more PHP dispersant at a rate that reduces the filtration loss by 2mL / 30min for every 0.3% increase; ③ If the microscopic image shows a dispersion coefficient <0.85, prioritize increasing the PHP dispersant by 0.1%-0.2%. If the dispersion coefficient still does not improve after 3 minutes, reduce the bentonite dosage by 0.5%; S2.3.3, Closed-loop verification and iteration: After adjustment, re-collect data after 5 minutes. If the deviation rate of all parameters is ≤10%, lock the current ratio; if there are still parameters exceeding the standard, repeat steps S2.3.1-S2.3.2 until two consecutive monitoring results meet the standard.

[0081] Furthermore, it should be noted that the response surface methodology formula is as follows: Let the amount of bentonite be A (%) and the amount of PHP dispersant be B (%), then the following second-order regression model is established:

[0082] Viscosity Y1(s): Y1=a0+a1A+a2B+a3A 2 +a4B 2 +a5AB;

[0083] Density Y2 (g / cm³) 3 Y2 = b0 + b1A + b2B + b3A 2 +b4B 2 +b5AB;

[0084] Filtration loss Y3 (mL / 30min): Y3=c0+c1A+c2B+c3A2 +c4B 2 +c5AB;

[0085] Where a0-a5, b0-b5, and c0-c5 are regression coefficients, obtained by least squares fitting. The model needs to be tested by F, and P < 0.05. The fiber optic sensor uses a distributed fiber optic strain gauge (spatial resolution 1m, accuracy ±2με), and the density is calculated using the Brillouin scattering effect (density = K × strain value + K0, where K and K0 are calibration coefficients). The microscopic image recognition uses a 40x objective lens (resolution 0.5μm), and the dispersion coefficient is calculated using ImageJ software (formula: dispersion coefficient = 1 - (area of ​​aggregated particles / total particle area)). The deviation rate is calculated as |measured value - model predicted value| / 1.20 × 100%.

[0086] In this embodiment, it should also be noted that in S3, the pore size change is monitored in real time using the ultrasonic pulse echo method, and the mud skin structure is reconstructed in three dimensions using CT scanning and acoustic emission technology to provide early warning of pore collapse risk. The specific steps are as follows: S3.1, The pore size change data during the pore formation process is captured in real time using the ultrasonic pulse echo method; S3.2, The microstructure information of the mud skin is obtained through CT scanning, and the stress release signal monitored by acoustic emission technology is superimposed to perform three-dimensional reconstruction of the mud skin structure; The specific steps are as follows: S3.2.1, CT scanning parameters: ① A microfocus CT scanner is used, with a scanning speed of 1 minute / ① Obtain cross-sectional images of the mud crust at a frequency of 1000 Hz; ② Extract the porosity and thickness distribution of the mud crust using a threshold segmentation algorithm; S3.2.2 Acoustic emission signal processing: ① Deploy broadband acoustic emission sensors on the sidewall of the simulation device to capture mud crust rupture signals in real time; ② Analyze the frequency domain characteristics of the signals based on wavelet transform to identify precursor signals of borehole collapse; S3.2.3 Three-dimensional fusion reconstruction: ① Register CT structural data with acoustic emission positioning results; ② Construct a three-dimensional model of the mud crust that includes pore distribution and stress field changes. When the porosity is >25% and accompanied by high-frequency acoustic emission signals, a first-level borehole collapse warning is triggered. S3.3 Based on the pore size change trend and mud crust structural characteristics, comprehensively analyze and warn of borehole wall collapse risk. The specific steps are as follows: S3.3.1 Multi-dimensional feature extraction: Extract the pore shrinkage rate and pore fluctuation amplitude from the pore size change data in S3.1; extract the average thickness of the mud crust, the proportion of local damaged area, and the energy rate of the acoustic emission signal from the three-dimensional mud crust structure in S3.2, and establish a three-dimensional feature matrix of "pore size dynamic characteristics - mud crust structure parameters - acoustic emission energy"; S3.3.2 Risk level classification and threshold setting: Low risk: pore shrinkage rate ≤ 0.5 mm / min, fluctuation amplitude ≤ 5%; mud crust thickness ≥ 3 mm and damaged area proportion ≤ 3%; acoustic emission energy rate ≤ 10 μJ / s; Medium risk: pore shrinkage rate 0.5-1 mm / min, fluctuation amplitude 5%-10 %; Mud cake thickness 2-3mm or damaged area percentage 3%-10%; Acoustic emission energy rate 10-50μJ / s; High risk: pore size shrinkage rate >1mm / min, fluctuation range >10%; mud cake thickness <2mm or damaged area percentage >10%; acoustic emission energy rate >50μJ / s; S3.3.3 Comprehensive early warning and response mechanism: When any two indicators in the three-dimensional feature matrix reach the medium risk threshold, an early warning is triggered, and it is recommended to suspend the test and check the mud performance; If any indicator reaches the high risk threshold, an emergency early warning is triggered immediately, the current mud cake structure image and pore size data are automatically recorded, and the in-hole mud pressure compensation program is started until the risk level drops to the low risk range.

[0087] Furthermore, it should be noted that the formula for calculating the aperture in S3.1 is: Where v is the propagation speed of the ultrasonic wave in the mud, and t is the round-trip time of the echo. The threshold setting in S3.3.2 is based on: "Based on the statistical data of the early construction of the Weiyi Expressway, when the borehole diameter shrinkage rate is >1mm / min, the probability of borehole collapse within 30 minutes reaches 70%; when the mud cake thickness is <2mm, the borehole wall stability decreases by 60%."

[0088] In this embodiment, it should also be noted that S4 trains a random forest model based on historical data, develops a digital twin platform to recommend the optimal mud ratio, and feeds it back to stage S2 in real time. The specific steps are as follows: S4.1, Collect historical data on mud ratio, performance parameters, and filling coefficient, and train a random forest model to establish the correlation between mud ratio and hole formation quality; S4.2, Build a digital twin platform to map the physical test process with virtual simulation and visualize and predict the mud ratio effect; The specific steps are as follows: S4.2.1, Data interface layer construction: through OPC The UA protocol provides real-time access to physical test data on mud proportions and performance parameters, as well as sensor data on borehole diameter and mud cake thickness. Data cleaning rules are established to remove noisy data. S4.2.2 Virtual Model Layer Development: A 3D geological-pile borehole model is constructed based on the Unity3D engine to reproduce complex geological conditions in mountainous areas. A random forest prediction model is integrated, providing proportion optimization services via API. S4.2.3 Virtual-Real Mapping and Verification: A synchronization cycle between physical tests and virtual simulations is set. When the filling coefficient deviation of the virtual prediction exceeds 15%, a physical test parameter calibration command is triggered. S4.3 Outputting the optimal proportion scheme through the digital twin platform and feeding it back to stage S2 in real time supports dynamic adjustment of the mud proportions.

[0089] Furthermore, it should be noted that in model training in S4.1, historical data is standardized using Z-score, and the feature variables include bentonite content, PHP content, viscosity, density, filtration loss, formation pressure, and permeability rate. The target variable is the filling coefficient deviation rate.

[0090] In this embodiment, it should also be noted that in step S5, the mix proportion is verified in a typical geological section of the Weiyi Expressway, and the filling coefficient and borehole qualification rate are compared to form a standardized visualization report. The specific steps are as follows: S5.1, Select a representative typical geological section of the Weiyi Expressway and apply the optimized mud mix proportion to verify the pile foundation construction; S5.2, Compare and analyze the deviation between the actual filling coefficient and the design value during on-site construction, calculate the borehole qualification rate, and evaluate the applicability and reliability of the mix proportion; S5.3, Integrate the test data, verification results, and analysis conclusions to form a standardized report containing visualization charts.

[0091] Furthermore, it should be noted that the filling coefficient is: K = V 实际 / V 理论 , where V 理论 =π×(d / 2)2 ×L, where d is the designed pile diameter and L is the pile length. Hole formation qualification rate: Qualification rate = Number of qualified piles / Total number of piles × 100%, qualification standard is "hole diameter deviation ≤ 5% (|actual diameter - design diameter| / design diameter × 100% ≤ 5%), verticality ≤ 1% (deviation value / pile length × 100% ≤ 1%)".

[0092] In this embodiment, an indoor test visualization method for pile foundation mud cake mix ratio testing is described as follows: First, suitable materials are selected and a simulation device is built based on the engineering geological report. The mud slurry ratio is dynamically adjusted by combining response surface methodology modeling with sensing and image recognition. Simultaneously, ultrasonic, CT scanning, and acoustic emission technologies are used to monitor pore size and mud cake structure and provide early warning of risks. Then, a digital twin platform is developed based on historical data to train the model, recommend the optimal mix ratio, and provide feedback for optimization. Finally, after verification in typical strata, a standardized visualization report is generated, realizing the full-process visualization testing and optimization of pile foundation mud cake mix ratio from indoor testing to engineering application.

[0093] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0094] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A laboratory test visualization method for pile-soil interaction ratio test, characterized in that, Includes the following steps: S1: Based on the geological report of the Weiyi Expressway project, bentonite and PHP dispersant were selected as slurry materials, and a simulation device with adjustable confining pressure and groundwater infiltration rate was configured. The mud ratio, formation pressure and groundwater conditions were set as variables. S2: A mud performance ratio model was established using the response surface methodology, and the ratio was dynamically adjusted by combining fiber optic sensors and microscopic image recognition technology. S3: Real-time monitoring of aperture changes using ultrasonic pulse echo method, combined with CT scanning and acoustic emission technology to reconstruct mud skin structure in three dimensions and provide early warning of hole collapse risk; S4: Train a random forest model based on historical data, develop a digital twin platform to recommend the optimal ratio, and feed it back to the S2 stage in real time; S5: Verify the mix proportions in typical geological sections of the Weiyi Expressway, compare the filling coefficient with the borehole qualification rate, and generate a standardized visual report; In S2, a response surface methodology is used to establish a mud performance ratio model, and the ratio is dynamically adjusted by combining fiber optic sensors and microscopic image recognition technology. The specific steps are as follows: S2.1 Response surface methodology was used to design experiments and establish a nonlinear regression model between the mud mix ratio and the performance parameters of viscosity, density, and filtration loss. S2.

2. Collect mud density and rheological parameters through fiber optic sensors, and analyze the microscopic dispersion state of mud using microscopic image recognition technology. S2.

3. Based on monitoring data and model output results, optimize the ratio of bentonite to PHP dispersant in the mud in real time; S4 trains a random forest model based on historical data, develops a digital twin platform to recommend the optimal allocation, and feeds back to the S2 stage in real time. The specific steps are as follows: S4.1 Collect historical data on mud ratio, performance parameters, and filling coefficient, and train a random forest model to establish the correlation between mud ratio and hole formation quality; S4.2 Build a digital twin platform to map the physical test process with virtual simulation and make a visual prediction of the mud mixing effect; S4.3 Output the optimal mixing ratio through the digital twin platform and feed it back to the S2 stage in real time to support the dynamic adjustment of the mud mixing ratio.

2. The indoor test visualization method for testing the mud cake mix ratio of pile foundations according to claim 1, characterized in that, In step S1, based on the geological report of the Weiyi Expressway project, bentonite and PHP dispersant were selected as slurry materials. A simulation device with adjustable confining pressure and groundwater infiltration rate was configured, and the mud ratio, formation pressure, and groundwater conditions were set as variables. The specific steps are as follows: S1.1 Based on the geological survey report of the Weiyi Expressway project, bentonite was selected as the main slurry material, and PHP dispersant was selected as an additive. S1.2 Build an indoor test simulation device that can precisely adjust the confining pressure and groundwater seepage rate; S1.

3. Clearly define mud mix ratio, formation pressure, and groundwater conditions as key variables in the experiment.

3. The indoor test visualization method for testing the mud cake mix ratio of pile foundations according to claim 1, characterized in that, The specific steps of S2.3 are as follows: S2.3.1 Data Comparison and Deviation Grading: The real-time density, rheological parameters, and dispersion coefficient of microscopic image recognition collected by the fiber optic sensor in S2.2 are compared with the predicted values ​​of the regression model in S2.

1. The deviation rate of each parameter is calculated and divided into three levels: deviation rate ≤ 10% is qualified, 10% < deviation rate ≤ 20% is slightly exceeding the standard, and deviation rate > 20% is seriously exceeding the standard. S2.3.2, Grading Adjustment Rules: ① If the density is >1.25g / cm³, it is considered a slight exceedance. The density can be slightly adjusted by reducing the amount of bentonite by 0.02g / cm³ for every 1% reduction in the amount of bentonite used, while keeping the proportion of PHP dispersant unchanged. ② If the filtration loss is >15mL / 30min, it is considered a serious overdose. Under the premise that the amount of bentonite remains unchanged, the filtration loss can be reduced by 2mL / 30min for every 0.3% increase in PHP dispersant. ③ If the microscopic image shows a dispersion coefficient < 0.85, increase the amount of PHP dispersant by 0.1%-0.2%. If the dispersion coefficient still does not improve after 3 minutes, reduce the amount of bentonite by 0.5%. S2.3.3 Closed-loop verification and iteration: After adjustment, recollect data after 5 minutes. If the deviation rate of all parameters is ≤10%, lock the current ratio. If there are still parameters exceeding the standard, repeat steps S2.3.1-S2.3.2 until two consecutive monitoring results meet the standard.

4. The indoor test visualization method for testing the mud cake mix ratio of pile foundations according to claim 1, characterized in that, In step S3, the pore size change is monitored in real time using the ultrasonic pulse echo method. Combined with CT scanning and acoustic emission technology, the mud cake structure is reconstructed in three dimensions, and the risk of pore collapse is warned. The specific steps are as follows: S3.

1. The ultrasonic pulse echo method is used to capture the aperture change data in real time during the hole forming process; S3.

2. Obtain the microstructure information of mud skin through CT scanning, and superimpose the stress release signal monitored by acoustic emission technology to reconstruct the three-dimensional structure of mud skin. S3.3 Based on the pore size variation trend and mud cake structure characteristics, comprehensively analyze and warn of the risk of pore wall collapse.

5. The indoor test visualization method for testing the mud cake mix proportion of pile foundations according to claim 4, characterized in that, The specific steps of S3.2 are as follows: S3.2.1 CT Scan Parameters: ① A microfocus CT scanner was used to acquire cross-sectional and longitudinal images of the mud skin at a frequency of 1 minute / time; ② The porosity and thickness distribution of the mud cake were extracted using a threshold segmentation algorithm; S3.2.2 Acoustic emission signal processing: ① Deploy broadband acoustic emission sensors on the side wall of the simulation device to capture mud skin rupture signals in real time; ② Identify precursor signals of hole collapse by analyzing the frequency domain characteristics of the signal based on wavelet transform; S3.2.3, Three-dimensional fusion reconstruction: ① Register the CT structural data with the acoustic emission localization results; ② Construct a three-dimensional model of mud crust that includes pore distribution and stress field changes. When the porosity is >25% and accompanied by high-frequency acoustic emission signals, trigger a first-level pore collapse warning.

6. The indoor test visualization method for testing the mud cake mix ratio of pile foundations according to claim 4, characterized in that, The specific steps of S3.3 are as follows: S3.3.1 Multi-dimensional feature extraction: The pore size shrinkage rate and pore size fluctuation amplitude are extracted from the pore size variation data in S3.1; the average thickness of the mud skin, the proportion of local damaged area and the energy rate of the acoustic emission signal are extracted from the three-dimensional mud skin structure in S3.2, and a three-dimensional feature matrix of "pore size dynamic characteristics - mud skin structure parameters - acoustic emission energy" is established. S3.3.2 Risk Level Classification and Threshold Setting: Low risk: pore shrinkage rate ≤0.5mm / min, fluctuation range ≤5%; mud cake thickness ≥3mm and damaged area ≤3%; acoustic emission energy rate ≤10μJ / s; Medium risk: pore size shrinkage rate 0.5-1mm / min, fluctuation range 5%-10%; mud cake thickness 2-3mm or damaged area 3%-10%; acoustic emission energy rate 10-50μJ / s; High risk: Aperture shrinkage rate > 1 mm / min, fluctuation range > 10%; mud cake thickness < 2 mm or damaged area ratio > 10%; acoustic emission energy rate > 50 μJ / s; S3.3.3 Integrated Early Warning and Response Mechanism: When any two indicators in the three-dimensional feature matrix reach the medium-risk threshold, an early warning is triggered, and it is recommended to suspend the test and check the mud performance. If any indicator reaches the high-risk threshold, an emergency warning will be triggered immediately, and the current mud skin structure image and borehole diameter data will be automatically recorded. At the same time, the borehole mud pressure compensation program will be started until the risk level drops to the low-risk range.

7. The indoor test visualization method for testing the mud cake mix proportion of pile foundations according to claim 1, characterized in that, The specific steps in S4.2 are as follows: S4.2.1 Data Interface Layer Construction: The OPC UA protocol allows real-time access to physical test data on mud ratio and performance parameters, as well as sensor data on pore size and mud cake thickness. Establish data cleaning rules to remove noisy data; S4.2.2 Virtual Model Layer Development: A 3D geological-pile borehole model was built using the Unity3D engine to reproduce the complex geological conditions in mountainous areas. Integrate the random forest prediction model and provide allocation optimization services via API interface; S4.2.3 Virtual-Real Mapping and Verification: Set the synchronization period between physical experiments and virtual simulations; When the deviation of the virtual prediction filling coefficient is greater than 15%, a physical test parameter calibration command is triggered.

8. The indoor test visualization method for testing the mud cake mix ratio of pile foundations according to claim 1, characterized in that, In S5, the mix design was verified in a typical geological section of the Weiyi Expressway. The filling coefficient and the borehole qualification rate were compared to generate a standardized visual report. The specific steps are as follows: S5.

1. Select a representative typical stratum section in the Weiyi Expressway and verify the pile foundation construction using the optimized mud mix ratio. S5.2 Compare and analyze the deviation between the actual filling coefficient and the design value during on-site construction, calculate the hole-forming qualification rate, and evaluate the applicability and reliability of the mix proportion. S5.3 Integrate experimental data, verification results, and analysis conclusions to form a standardized report that includes visual charts.

Images