A method and system for constructing an expansive soil expansibility evaluation model

Abstract

The present application relates to the technical field of data prediction, in particular to a swelling soil swelling evaluation model construction method and system, comprising the following steps: collecting wetting and drying cycle data to construct closed state evolution coordinate sequence, using trapezoidal numerical integration algorithm to calculate single cycle hysteresis energy dissipation density, calculating structure damage increment to update structure integrity coefficient, constructing suction stress contribution factor and correcting real-time matrix suction, superimposing to obtain equivalent effective stress, using Lagrange interpolation combined with pore ratio relationship table to calculate pore ratio change and volume expansion rate. In the present application, the structure damage and integrity degradation are quantified by calculating the hysteresis energy dissipation value, the stress state is corrected by constructing the suction stress contribution factor, and the pore ratio change is deduced, thereby establishing a mechanical property prediction system based on energy dissipation and structure evolution, overcoming the limitations of static physical index evaluation, and significantly improving the prediction accuracy and applicability of swelling soil volume deformation.

Description

Technical Field

[0001] This invention relates to the field of data prediction technology, and in particular to a method and system for constructing an assessment model for the expansibility of expansive soil. Background Technology

[0002] The field of data prediction technology refers to the technical scope of using computer hardware and software systems to perform statistical analysis, trend extrapolation, and model building on various types of collected numerical or textual data to provide decision support. Among these methods, the traditional approach to constructing an assessment model for the expansiveness of expansive soil involves researchers first collecting undisturbed soil samples in the field using a ring cutter and a soil cutter. Then, they measure the natural moisture content, void ratio, and liquid and plastic limits of the soil samples using an electronic balance, an electric heating drying oven, and a combined liquid and plastic limit analyzer. The data is then manually entered into a computer spreadsheet software. The software's built-in linear regression statistical tool is used to calculate the correlation coefficient, and the correlation curve between the expansive potential and the physical indicators is manually plotted based on the calculation results to establish an assessment reference model.

[0003] Existing technologies rely solely on basic physical indicators such as moisture content and liquid limit for simple linear regression statistics. They fail to deeply consider the dynamic evolution of the internal microstructure of expansive soil under complex conditions of moisture absorption and desaturation, and neglect the cumulative impact of hydraulic hysteresis on soil energy dissipation and structural damage. As a result, the assessment model cannot accurately reflect the true expansion mechanism of soil under variable stress. Furthermore, static correlation based solely on physical indicators is insufficient to capture the nonlinear mechanical response caused by the degradation of soil structural integrity. This leads to a significant deviation between the expansion potential prediction results and actual engineering performance, failing to meet the decision-making requirements for high-precision settlement control in geotechnical engineering and severely reducing the scientific rigor and reliability of engineering construction safety assessments. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by proposing a method and system for constructing an assessment model for the expansibility of expansive soil.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a method for constructing an assessment model for the expansibility of expansive soil, comprising the following steps:

[0006] S1: Collect the matrix suction and volumetric water content values ​​of expansive soil during the moisture absorption and desiccation cycle, and construct a closed-state evolution coordinate sequence based on the matrix suction and volumetric water content values;

[0007] S2: Call the trapezoidal numerical integration algorithm to perform polygon area calculation on the closed state evolution coordinate sequence, calculate the area value of the closed region, and calculate the single-cycle hysteresis energy dissipation density based on the area value of the closed region.

[0008] S3: Obtain the mineral sensitivity constant, calculate the product of the single-cycle hysteresis energy dissipation density and the mineral sensitivity constant to obtain the structural damage increment, update the initial integrity variable using the structural damage increment to obtain the current structural integrity coefficient, and construct the suction stress contribution factor based on the current structural integrity coefficient through a piecewise linear mapping function;

[0009] S4: Collect real-time matrix suction and total overlying stress, use the suction stress contribution factor to weight and correct the real-time matrix suction to obtain corrected suction stress, superimpose the total overlying stress and the corrected suction stress to obtain equivalent effective stress, input the equivalent effective stress into the Lagrange interpolation algorithm and combine it with the effective stress-vomage ratio relationship table to obtain the target porosity change, and calculate the volume expansion rate percentage based on the target porosity change.

[0010] As a further aspect of the present invention, the process of constructing the closed-state evolution coordinate sequence in S1 includes:

[0011] S11: In a preset constant temperature and humidity controlled environment, the changes in matrix suction and volumetric water content of the expansive soil sample during a single absorption and dehumidification cycle are monitored in real time by a tensiometer sensor and a time domain reflectometer sensor, respectively.

[0012] S12: Use a timestamp synchronization algorithm to register and align the matrix suction change data and the volumetric water content change data, remove outlier noise points caused by sensor response delay, and generate a state dataset including multiple sets of suction-water content paired data.

[0013] S13: Map each pair of data in the state dataset to a vector node in a two-dimensional Cartesian coordinate system according to the time evolution order, and force a closed connection between the first vector node and the last vector node to generate a closed state evolution coordinate sequence.

[0014] As a further aspect of the present invention, the calculation process of the single-cycle hysteresis energy dissipation density in S2 includes:

[0015] S21: Extract all vector node coordinates in the closed state evolution coordinate sequence, divide the sequence into a moisture absorption path point set according to the ascending order of matrix suction values, and divide the sequence into a desiccation path point set according to the descending order of matrix suction values;

[0016] S22: Call the trapezoidal numerical integration algorithm to calculate the first integral area below the dehumidification path point set and the second integral area below the moisture absorption path point set respectively, and subtract the second integral area from the first integral area to obtain the closed region area value.

[0017] S23: Obtain the dimensionless effective energy conversion ratio coefficient, multiply the area value of the closed region with the energy conversion coefficient, and generate the single-cycle hysteresis energy dissipation density.

[0018] As a further aspect of the present invention, the process of constructing the suction stress contribution factor in S3 includes:

[0019] S31: The mineral composition of expansive soil is analyzed by X-ray diffraction test, and the mineral sensitivity constant is determined according to the percentage weight of hydrophilic mineral content;

[0020] S32: Perform the multiplication operation between the single-cycle hysteresis energy dissipation density and the mineral sensitivity constant to generate structural damage increment;

[0021] S33: Obtain the initial integrity variable from the previous moment, subtract the structural damage increment from it, and obtain the current structural integrity coefficient;

[0022] S34: Substitute the current structural integrity coefficient into the preset damage-contribution mapping model, determine the corresponding weight coefficient through piecewise linear discrimination logic, and generate the suction stress contribution factor.

[0023] As a further aspect of the present invention, the calculation process of the volume expansion rate percentage in S4 includes:

[0024] S41: Use a monitoring probe embedded in the soil to obtain the current real-time matrix suction, and calculate the total overburden stress by combining the soil depth parameters and soil unit weight parameters.

[0025] S42: The real-time matrix suction is reduced and corrected by the suction stress contribution factor, and the reduced value is superimposed with the total overlying stress to generate an equivalent effective stress.

[0026] S43: Find the reference stress node adjacent to the equivalent effective stress in the preset effective stress-void ratio standard relationship table, estimate the corresponding current void ratio using the Lagrange interpolation polynomial, and calculate the difference between the current void ratio and the initial void ratio to obtain the target void ratio change.

[0027] S44: Calculate and output the percentage of volume expansion rate based on the ratio of the target porosity change to the initial porosity.

[0028] As a further aspect of the present invention, the calculation process for the area of ​​the closed region in step S22 is specifically as follows:

[0029] Obtain adjacent coordinate point pairs in the dehumidification path point set and adjacent coordinate point pairs in the moisture absorption path point set;

[0030] For each pair of adjacent coordinate points, the product of the absolute value of the difference in matrix suction and the average value of volumetric water content is calculated to obtain the area of ​​the infinitesimal trapezoid.

[0031] The areas of all the aforementioned trapezoidal micro-elements are summed, and the absolute value of the difference between the total area of ​​the dehumidification path and the total area of ​​the moisture absorption path is calculated to generate the area value of the closed region.

[0032] As a further aspect of the present invention, the process of constructing the suction stress contribution factor in S34 adopts the following piecewise linear mapping function formula:

[0033] ;

[0034] in, This represents the suction stress contribution factor. Represents the current structural integrity coefficient. This represents the preset structural elastic limit threshold.

[0035] As a further aspect of the present invention, the process of obtaining the target porosity change in S43 is specifically as follows:

[0036] Retrieve the preset effective stress-porosity ratio standard relationship table, and filter out three known stress nodes and their corresponding known porosity ratio data that are closest to the equivalent effective stress value;

[0037] A second-order Lagrange interpolation basis function is constructed, with the three known stress nodes as interpolation nodes and the equivalent effective stress as the independent variable input into the basis function. The current predicted porosity is obtained by weighted calculation.

[0038] The numerical difference between the current predicted void ratio and the initial void ratio under the initial state of the soil is calculated to generate the target void ratio change.

[0039] As a further aspect of the present invention, the calculation process of the volume expansion rate percentage in S44 adopts the following formula:

[0040] ;

[0041] in, This represents the percentage of volume expansion rate. This represents the change in the target porosity. This represents the initial void ratio of the soil at the initial moment.

[0042] A system for constructing an assessment model for the expansiveness of expansive soil, the system being used to implement the aforementioned method for constructing an assessment model for the expansiveness of expansive soil, the system comprising:

[0043] The state sequence construction module is used to collect the matrix suction and volumetric water content values ​​of expansive soil during the absorption and desiccation cycle, perform time synchronization and data cleaning operations, and construct a closed state evolution coordinate sequence with the beginning and end connected based on the cleaned data.

[0044] The energy dissipation calculation module is used to call the trapezoidal numerical integration algorithm to perform area calculation on the polygonal region surrounded by the closed state evolution coordinate sequence, calculate the area value of the closed region, and convert the area value to obtain the single-cycle hysteresis energy dissipation density.

[0045] The factor mapping generation module is used to obtain the mineral sensitivity constant and calculate the structural damage increment by combining the single-cycle hysteresis energy dissipation density. The initial integrity variable is updated using the structural damage increment to obtain the current structural integrity coefficient. The suction stress contribution factor is generated by a piecewise linear mapping function.

[0046] The expansion assessment calculation module is used to collect real-time matrix suction and total overlying stress, calculate the equivalent effective stress using the suction stress contribution factor, obtain the target porosity change by combining the effective stress porosity ratio relationship table with the Lagrange interpolation algorithm, and calculate the output volume expansion rate percentage.

[0047] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0048] In this invention, a closed-state evolution sequence is constructed by collecting data from the absorption and desiccation cycles, and the hysteretic energy dissipation value is calculated using the trapezoidal numerical integration algorithm. This quantifies the incremental structural damage and integrity degradation of the soil under hydraulic action. A piecewise linear mapping is used to construct a suction stress contribution factor to achieve accurate correction of the stress state. The equivalent effective stress is obtained by combining the overlying stress, and the change in void ratio is deduced using the Lagrange interpolation algorithm. This effectively establishes a mechanical performance prediction system based on energy dissipation and structural evolution, overcomes the limitations of traditional static physical index evaluation, and significantly improves the accuracy and applicability of volume deformation prediction for expansive soil under complex working conditions. This provides solid theoretical support and data guarantee for the stability analysis of geotechnical engineering. Attached Figure Description

[0049] Figure 1 This is the main flowchart of the method for constructing an assessment model for the expansibility of expansive soil according to the present invention;

[0050] Figure 2 This is a flowchart illustrating the construction of the closed-state evolution coordinate sequence of the present invention.

[0051] Figure 3 This is a flowchart of the calculation of single-cycle hysteresis energy dissipation density in this invention;

[0052] Figure 4 This is a flowchart illustrating the construction process of the suction stress contribution factor in this invention.

[0053] Figure 5 This is a flowchart illustrating the calculation of the percentage volume expansion rate in this invention. Detailed Implementation

[0054] To make the objectives, technical solutions, and advantages of this invention clearer, the software-based technical solution is described in detail below with reference to system architecture diagrams and embodiments. It should be understood that the specific embodiments described herein are only for explaining the technical solutions of this invention and do not constitute a limitation on the scope of protection.

[0055] In the description of this invention, the system architecture relationships or data processing flows indicated by terms such as "layer," "module," "interface," "data flow," "client," and "server" are all defined based on the architecture diagram or flowchart corresponding to the embodiments. This way of describing is only used to clearly illustrate the logical relationships between the elements in the technical solution, and not to limit the physical deployment form. The term "multiple" includes two or more technical units, including but not limited to multiple data nodes, processing threads, service instances, or functional components and other scalable elements. The specific number is determined according to the actual business scenario and needs to be specifically specified.

[0056] Please see Figure 1 and Figure 2 This invention provides a technical solution: a method for constructing an assessment model for the expansibility of expansive soil, comprising the following steps:

[0057] S1: Collect the matrix suction and volumetric water content values ​​of expansive soil during the moisture absorption and desiccation cycle, and construct a closed-state evolution coordinate sequence based on the matrix suction and volumetric water content values.

[0058] The process of constructing the closed-state evolution coordinate sequence in S1 includes:

[0059] S11: In a preset constant temperature and humidity controlled environment, the changes in matrix suction and volumetric water content of the expansive soil sample during a single absorption and dehumidification cycle are monitored in real time by a tensiometer sensor and a time domain reflectometer sensor, respectively.

[0060] S12: Use a timestamp synchronization algorithm to register and align matrix suction change data and volumetric water content change data, remove outlier noise points caused by sensor response delay, and generate a state dataset including multiple sets of suction-water content paired data.

[0061] S13: Map each pair of data in the state dataset to a vector node in a two-dimensional Cartesian coordinate system according to the time evolution order, and force a closed connection between the first vector node and the last vector node to generate a closed state evolution coordinate sequence.

[0062] In the specific implementation of step S1, in order to accurately capture the hydraulic response characteristics of expansive soil under changes in environmental humidity and construct a high-precision closed-state evolution coordinate sequence, this step is performed in a strictly controlled geotechnical engineering laboratory environment. For S11, the experimental environment is set in a temperature and humidity coupled control chamber, and the ambient temperature is kept constant at [temperature value missing]. To eliminate the interference of temperature fluctuations on the pore water pressure of the soil, the expansive soil samples selected were remolded samples taken from the engineering site, prepared into standard ring specimens with a diameter of 50 mm and a height of 20 mm, and the initial dry density was controlled at [value missing]. Regarding monitoring equipment, a miniature high-intake tensiometer was used to monitor matrix suction in real time. The sensor probe was pre-saturated and vertically inserted into the center of the sample. Simultaneously, a miniature time-domain reflectometer probe was inserted horizontally into the side of the sample to monitor volumetric moisture content. The experimentally defined single-cycle moisture absorption and dehumidification was as follows: first, the relative humidity was gradually reduced from 95% to 30% (dehumidification stage); after the sample mass change rate was less than 0.01 g / h, the relative humidity was gradually restored from 30% to 95% (moisture absorption stage). Throughout this process, the sensor data sampling frequency was uniformly set to 1 Hz.

[0063] For S12, although the data transmission of the tension meter and time-domain reflectometer sensors is both 1Hz, there is a slight deviation in the millisecond-level timestamps. Therefore, strict timestamp synchronization and cleaning must be performed. First, a linear interpolation alignment algorithm is used to process the data. Using the tension meter's timestamp... Using the reference axis, for any time If the timestamp of the time domain reflectometer Not precisely located Then search Mid-range The most recent moment before and the next moment The data, through formula Calculate the aligned volumetric water content to generate a strictly paired original dataset. Here, Representative moment Volumetric moisture content after alignment and These represent the measured volumetric water content at the previous and next time points, respectively. Next, outlier removal is performed. The sliding window size is set to 20 data points, and the mean of the matrix suction data within the window is calculated. with standard deviation If the current data point satisfy If the noise is found to be outlier noise caused by poor sensor contact or electromagnetic interference, it is removed and replaced with a window mean. After the above processing, a state dataset containing 1200 sets of high-confidence paired suction-moisture content data is generated.

[0064] For S13, the construction process is performed in a two-dimensional Cartesian coordinate system, defining the horizontal axis as matrix suction (unit: kPa) and the vertical axis as volumetric water content (unit: %). Each set of data in the state dataset... The nodes are mapped sequentially to vector nodes in a coordinate system. Due to plastic adjustment of the soil's microstructure at the end of the experiment, the final nodes... Often unable to connect with the first node Complete overlap. To quantify the energy characteristics of the hysteresis loop, a forced closure connection must be performed. Specifically, this involves constructing a connection from... point to A virtual linear path is constructed, with the number of interpolation points on the path set to 1% of the total number of points. Virtual data points are generated through linear interpolation and appended to the end of the sequence, thus forming a geometrically closed annular polygon sequence.

[0065] The aforementioned closed-state evolution coordinate sequence refers to the geometric trajectory sequence formed by connecting the data points of matrix suction and volumetric water content corresponding to the hygroscopic process of expansive soil in a two-dimensional coordinate system in chronological order and forcing the beginning and end to close. This sequence intuitively reflects the hysteretic characteristics of the soil under the action of hydraulic circulation.

[0066] The aforementioned linear interpolation alignment algorithm is a mathematical method for estimating the unknown value between two data points using known discrete data points. In this embodiment, it is used to solve the problem that the sampling time points of two different sensors do not completely overlap, ensuring strict correspondence of data on the time axis.

[0067] Please see Figure 1 and Figure 3 S2: Call the trapezoidal numerical integration algorithm to perform polygon area calculation on the closed state evolution coordinate sequence, calculate the area value of the closed region, and calculate the single-cycle hysteresis energy dissipation density based on the area value of the closed region.

[0068] The calculation process for the single-cycle hysteresis energy dissipation density in S2 includes:

[0069] S21: Extract all vector node coordinates in the closed state evolution coordinate sequence, divide the sequence into a moisture absorption path point set according to the ascending order of matrix suction values, and divide the sequence into a desiccation path point set according to the descending order of matrix suction values.

[0070] S22: Call the trapezoidal numerical integration algorithm to calculate the first integral area below the dehumidification path point set and the second integral area below the moisture absorption path point set respectively, and subtract the second integral area from the first integral area to obtain the closed region area value.

[0071] S23: Obtain the dimensionless effective energy conversion ratio coefficient, multiply the closed region area value with the energy conversion coefficient to generate the single-cycle hysteresis energy dissipation density.

[0072] The specific calculation process for the area of ​​the closed region in S22 is as follows:

[0073] Obtain adjacent coordinate point pairs in the desiccation path point set and adjacent coordinate point pairs in the moisture absorption path point set;

[0074] For each pair of adjacent coordinate points, the product of the absolute value of the difference in matrix suction and the average value of volumetric water content is calculated to obtain the area of ​​the infinitesimal trapezoid.

[0075] The areas of all infinitesimal trapezoidal elements are summed, and the absolute value of the difference between the total area of ​​the dehumidification path and the total area of ​​the moisture absorption path is calculated to generate the area value of the closed region.

[0076] In the specific implementation of step S2, the core task is to transform the geometrically closed sequence constructed in S1 into a physical energy dissipation index. This step is implemented through a rigorous numerical integration algorithm, and its physical essence is to calculate the area enclosed by the desiccation curve and the hygroscopic curve in the soil-water characteristic curve. This area represents the energy consumed by the soil in one wet-dry cycle due to irreversible microstructural adjustments. For S21, the program first traverses the closed-state evolution coordinate sequence generated in S1. This is achieved by calculating the difference in matric suction between adjacent nodes. To determine the path direction. When If the duration exceeds 5% of the total sequence length, it is determined to be a dehumidification path, and the corresponding node is assigned to the dehumidification path point set. ;when When a node is identified as a moisture absorption path, it is assigned to the moisture absorption path point set. Because forced closure was performed in S1, the beginning and end of the sequence are continuous, ensuring complete coverage of the point set.

[0077] For S22, the complex trapezoidal numerical integration algorithm is used to calculate the area of ​​the closed region. First, the area of ​​the first integration below the dehumidification path is calculated. .for Each pair of adjacent points in and Calculate the area of ​​its infinitesimal trapezoidal element: .right Summing the areas of all infinitesimal elements yields Similarly, for Perform the same operation to calculate the second integral area below the moisture absorption path. Finally, the area of ​​the closed region is calculated. It is calculated as the absolute value of the difference between the two, i.e. .

[0078] For S23, in order to convert the geometric area into standard energy density units, a unit conversion factor needs to be introduced. Considering Dimensionally equivalent to The area of ​​a closed region is calculated based on the volumetric water content as a percentage. Its value is 100 times the actual physical work calculated based on decimals, hence the setting. The formula for calculating the energy dissipation density of a single-cycle hysteresis loop is as follows: ,in, Represents the energy dissipation density of a single-week hysteresis loop, in units of ; This represents the area of ​​a closed region, in units of... ; The unit conversion factor is 0.01, used to correct the area value in percentage form to the standard volumetric energy density value.

[0079] To verify the above calculation process, some key node data extracted from the experiment were selected for calculation. As shown in Table 1, the key points of the simplified dehumidification and moisture absorption paths are illustrated.

[0080] Table 1 Key Node Data of Expansive Soil Absorption and Dehydration Cycle

[0081] ;

[0082] The calculation process is as follows: First, calculate the dehumidification path integral; the area of ​​interval 1-2 is... The area of ​​interval 2-3 is Therefore Next, calculate the moisture absorption path integral; the area of ​​interval 4-5 is... The area of ​​interval 5-6 is Therefore Area of ​​the closed region One-week hysteresis energy dissipation density The results indicate that during a complete moisture absorption and desiccation cycle, 28.0 kJ of energy is dissipated per unit volume of soil due to moisture migration and structural rearrangement. This value reflects the cumulative damage to the soil structure.

[0083] The aforementioned complex trapezoidal numerical integration algorithm refers to a method that approximates the value of a definite integral by dividing the integration interval into several small trapezoids and calculating the sum of the areas of these small trapezoids. In this embodiment, it is used to accurately calculate the irregular geometric area enclosed by the hysteresis loop.

[0084] Please see Figure 1 and Figure 4S3: Obtain the mineral sensitivity constant, calculate the product of the single-cycle hysteresis energy dissipation density and the mineral sensitivity constant to obtain the structural damage increment, update the initial integrity variable using the structural damage increment to obtain the current structural integrity coefficient, and construct the suction stress contribution factor based on the current structural integrity coefficient through a piecewise linear mapping function.

[0085] The process of constructing the suction stress contribution factor in S3 includes:

[0086] S31: The mineral composition of expansive soil is analyzed by X-ray diffraction test, and the mineral sensitivity constant is determined according to the percentage weight of hydrophilic mineral content;

[0087] S32: Perform a multiplication operation between the single-cycle hysteresis energy dissipation density and the mineral sensitivity constant to generate structural damage increment;

[0088] S33: Obtain the initial integrity variable from the previous moment, subtract the structural damage increment from it, and obtain the current structural integrity coefficient;

[0089] S34: Substitute the current structural integrity coefficient into the preset damage-contribution mapping model, determine the corresponding weight coefficient through piecewise linear discrimination logic, and generate the suction stress contribution factor;

[0090] The process of constructing the suction stress contribution factor in S34 uses the following piecewise linear mapping function formula:

[0091] ;

[0092] in, Represents the suction stress contribution factor. Represents the current structural integrity coefficient. This represents the preset structural elastic limit threshold.

[0093] In the specific implementation of step S3, the aim is to establish a quantitative relationship between energy dissipation and the reduction of macroscopic mechanical parameters of the soil. This step introduces mineralogical composition analysis and damage mechanics theory, mapping microscopic energy loss to macroscopic stress transfer efficiency coefficients. For S31, the mineral composition of the expansive soil is first analyzed using X-ray diffraction tests. A small amount of soil from the same batch used in S1 is taken, ground to below 200 mesh, and tested on an X-ray diffractometer. Hydrophilic and non-hydrophilic minerals are identified using the Rietveld full-spectrum fitting quantitative analysis method. Mineral sensitivity constants are also discussed. The calculation formula is: .

[0094] For S32 and S33, perform cumulative updates of the damage variables. Define initial integrity variables. The value is 1.0 when the soil is undisturbed. Calculate the structural damage increment: Then, update the current structural integrity coefficient: Here, The results are directly derived from the calculation of S2.

[0095] For S34, a suction stress contribution factor is constructed. This factor reflects the efficiency with which matrix suction is converted into effective stress. The piecewise linear mapping function used has a preset structural elastic limit threshold. This is a key parameter. This threshold is determined through unconfined compressive strength tests; the integrity factor corresponding to a decrease in soil strength to 60% of its initial strength is defined as... In this embodiment, based on previous experimental calibration, the following settings are established. The suction stress contribution factor is constructed using the following piecewise linear mapping function formula:

[0096] ;

[0097] in, This represents the suction stress contribution factor, which is a dimensionless parameter. This represents the current structural integrity coefficient, which is obtained by subtracting the structural damage increment from the initial integrity variable. This represents the preset structural elastic limit threshold, which is set to 0.60 in this embodiment.

[0098] The actual calculation is as follows: Based on the S2 calculation results, the single-cycle hysteresis energy dissipation density is... X-ray diffraction analysis determined the montmorillonite content. (i.e., 20%) illite content (i.e., 15%). Substitute into the formula to calculate the mineral sensitivity constant. Calculate the structural damage increment. Assuming initial integrity variables Then the current structural integrity coefficient .Will With threshold Comparison, because Calculate the suction stress contribution factor according to the second line of the formula. The results indicate that the significant water absorption and swelling of hydrophilic minerals within the soil leads to the accumulation of structural damage, causing the current contribution of matrix suction to effective stress to decrease to 75.7%.

[0099] The Rietveld full-spectrum fitting quantitative analysis method mentioned above refers to a method that uses the least squares method to adjust the calculated diffraction pattern to best match the measured pattern, thereby accurately resolving the crystal structure and content of each phase in a multiphase mixture.

[0100] Please see Figure 1 and Figure 5 S4: Collect real-time matrix suction and total overlying stress, use the suction stress contribution factor to weight and correct the real-time matrix suction to obtain the corrected suction stress, superimpose the total overlying stress and the corrected suction stress to obtain the equivalent effective stress, input the equivalent effective stress into the Lagrange interpolation algorithm and combine it with the effective stress-vomage ratio relationship table to obtain the target porosity change, and calculate the volume expansion rate percentage based on the target porosity change.

[0101] The calculation process for the percentage of volume expansion rate in S4 includes:

[0102] S41: Use a monitoring probe embedded in the soil to obtain the current real-time matrix suction, and combine it with soil depth parameters and soil unit weight parameters to calculate the total overburden stress.

[0103] S42: The real-time matrix suction is reduced and corrected by the suction stress contribution factor. The reduced value is then superimposed on the total overlying stress to generate an equivalent effective stress.

[0104] S43: Find the reference stress node adjacent to the equivalent effective stress in the preset effective stress-void ratio standard relationship table, estimate the corresponding current void ratio using the Lagrange interpolation polynomial, and calculate the difference between the current void ratio and the initial void ratio to obtain the target void ratio change.

[0105] S44: Calculate and output the percentage of volume expansion rate based on the ratio of the target void ratio change to the initial void ratio;

[0106] The specific process for obtaining the target porosity change in S43 is as follows:

[0107] Retrieve the preset effective stress-void ratio standard relationship table, and filter out three known stress nodes and their corresponding known void ratio data that are closest to the equivalent effective stress values;

[0108] A second-order Lagrange interpolation basis function is constructed, with the three known stress nodes as interpolation nodes and the equivalent effective stress as the independent variable input basis function. The current predicted porosity is obtained by weighted calculation.

[0109] Calculate the numerical difference between the current predicted void ratio and the initial void ratio value under the initial state of the soil, and generate the target void ratio change.

[0110] The percentage of volume expansion in S44 is calculated using the following formula:

[0111] ;

[0112] in, Represents the percentage of volume expansion. This represents the change in the target porosity. This represents the initial void ratio of the soil at the initial moment.

[0113] In the specific implementation of step S4, the corrected parameters obtained in the previous steps are applied to the on-site working conditions, and the final volumetric expansion of the soil is predicted by combining the effective stress principle with changes in void ratio. For S41, automated monitoring stations are deployed at the expansive soil engineering site. At soil depth... A matrix suction sensor is installed at the location to collect real-time matrix suction data. Simultaneously, the natural unit weight of the soil layer at this depth was obtained. Total stress of the overlying layer Through formula Calculation. For S42, calculate the equivalent effective stress. Utilizing the suction stress contribution factor output in S3. The real-time matrix suction is weighted and reduced, and the total stress is then added. The formula is as follows: .

[0114] For S43, the change in target void ratio is obtained. First, a standard effective stress-void ratio relationship table needs to be pre-set. This table contains a set of data points obtained from indoor consolidation tests. To accurately calculate the void ratio under the current stress, a second-order Lagrange interpolation algorithm is used. The relationship table is then searched to find the equivalent effective stress that corresponds to the calculated value. The three closest known nodes. Construct a second-order Lagrange interpolation polynomial to estimate the current predicted porosity. And calculate its ratio to the initial void ratio of the soil. The difference For S44, the percentage of volume expansion is calculated based on the ratio of the change in target void ratio to the initial void ratio. The formula is as follows: in, Represents the percentage of volume expansion rate; The change in target porosity is obtained by subtracting the initial porosity from the predicted porosity. This represents the initial void ratio of the soil at the initial moment.

[0115] The actual calculation example is as follows: assuming the actual measured data is depth. average bulk density Real-time matrix suction And the initial porosity First, calculate the total stress of the overlying layer. The suction stress contribution factor calculated using S3. Calculate the equivalent effective stress To obtain the porosity, refer to the preset effective stress-porosity standard relationship table, as shown in Table 2.

[0116] Table 2. Standard Relationship between Effective Stress and Porosity Ratio

[0117] ;

[0118] Select Node , , Used as the interpolation base point. Substituting into the second-order Lagrange interpolation formula for calculation: basis functions Basis functions Basis functions Predicting porosity We take an approximate value of 0.886. Calculate the change in the target void ratio. Finally, calculate the percentage of volume expansion. The results indicate that, after accounting for the reduction in suction stress due to structural damage, the predicted volumetric expansion rate of expansive soil at this depth under the current stress state is 14.30%.

[0119] The aforementioned second-order Lagrange interpolation algorithm is a numerical analysis method that uses three known data points to construct a quadratic polynomial, thereby enabling high-precision estimation of the function value at any point within the interval.

[0120] A system for constructing an assessment model for expansive soil expansibility, the system being used to execute the aforementioned method for constructing an assessment model for expansive soil expansibility, the system comprising:

[0121] The state sequence construction module is used to collect the matrix suction and volumetric water content values ​​of expansive soil during the absorption and desiccation cycle, perform time synchronization and data cleaning operations, and construct a closed state evolution coordinate sequence with the beginning and end connected based on the cleaned data.

[0122] The energy dissipation calculation module is used to call the trapezoidal numerical integration algorithm to perform area calculation on the polygonal region enclosed by the closed state evolution coordinate sequence, calculate the area value of the closed region, and convert the area value to obtain the single-cycle hysteresis energy dissipation density.

[0123] The factor mapping generation module is used to obtain the mineral sensitivity constant and calculate the structural damage increment by combining it with the single-cycle hysteresis energy dissipation density. The initial integrity variable is updated using the structural damage increment to obtain the current structural integrity coefficient, and the suction stress contribution factor is generated through a piecewise linear mapping function.

[0124] The expansion assessment and calculation module is used to collect real-time matrix suction and total overlying stress, calculate the equivalent effective stress using the suction stress contribution factor, obtain the target porosity change through the Lagrange interpolation algorithm combined with the effective stress-porosity relationship table, and calculate the output volume expansion rate percentage.

[0125] The above embodiments illustrate preferred embodiments of the present invention. Any equivalent adjustments to the technical solution based on software engineering methods are within the scope of protection, including but not limited to: implementing algorithm logic using different programming languages, refactoring functional modules into services, adjusting data interaction protocols, and optimizing resource scheduling strategies. Any implementation scheme derived from reasonable modifications to the data processing flow, service call chain, or system architecture layer without departing from the core technology of the present invention should be considered within the protection scope defined by the technical solution of the present invention.

Claims

1. A method for constructing an assessment model for the expansibility of expansive soil, characterized in that, Includes the following steps: S1: Collect the matrix suction and volumetric water content values ​​of expansive soil during the moisture absorption and desiccation cycle, and construct a closed-state evolution coordinate sequence based on the matrix suction and volumetric water content values; The process of constructing the closed-state evolution coordinate sequence includes: S11: In a preset constant temperature and humidity controlled environment, the changes in matrix suction and volumetric water content of the expansive soil sample during a single absorption and dehumidification cycle are monitored in real time by a tensiometer sensor and a time domain reflectometer sensor, respectively. S12: Use a timestamp synchronization algorithm to register and align the matrix suction change data and the volumetric water content change data, remove outlier noise points caused by sensor response delay, and generate a state dataset including multiple sets of suction-water content paired data. S13: Map each pair of paired data in the state dataset to a vector node in a two-dimensional Cartesian coordinate system according to the time evolution order, and force a closed connection between the first vector node and the last vector node to generate a closed state evolution coordinate sequence. S2: Call the trapezoidal numerical integration algorithm to perform polygon area calculation on the closed state evolution coordinate sequence, calculate the area value of the closed region, and calculate the single-cycle hysteresis energy dissipation density based on the area value of the closed region. S3: Obtain the mineral sensitivity constant, calculate the product of the single-cycle hysteresis energy dissipation density and the mineral sensitivity constant to obtain the structural damage increment, update the initial integrity variable using the structural damage increment to obtain the current structural integrity coefficient, and construct the suction stress contribution factor based on the current structural integrity coefficient through a piecewise linear mapping function; S4: Collect real-time matrix suction and total overlying stress, use the suction stress contribution factor to weight and correct the real-time matrix suction to obtain corrected suction stress, superimpose the total overlying stress and the corrected suction stress to obtain equivalent effective stress, input the equivalent effective stress into the Lagrange interpolation algorithm and combine it with the effective stress-vomage ratio relationship table to obtain the target porosity change, and calculate the volume expansion rate percentage based on the target porosity change.

2. The method for constructing an assessment model for the expansiveness of expansive soil according to claim 1, characterized in that, The calculation process for the single-cycle hysteresis energy dissipation density in S2 includes: S21: Extract all vector node coordinates in the closed state evolution coordinate sequence, divide the sequence into a moisture absorption path point set according to the ascending order of matrix suction values, and divide the sequence into a desiccation path point set according to the descending order of matrix suction values. S22: Call the trapezoidal numerical integration algorithm to calculate the first integral area below the dehumidification path point set and the second integral area below the moisture absorption path point set respectively, and subtract the second integral area from the first integral area to obtain the closed region area value. S23: Obtain the dimensionless effective energy conversion ratio coefficient, multiply the area value of the closed region with the energy conversion coefficient, and generate the single-cycle hysteresis energy dissipation density.

3. The method for constructing an assessment model for the expansiveness of expansive soil according to claim 1, characterized in that, The process of constructing the suction stress contribution factor in S3 includes: S31: The mineral composition of expansive soil is analyzed by X-ray diffraction test, and the mineral sensitivity constant is determined according to the percentage weight of hydrophilic mineral content; S32: Perform the multiplication operation between the single-cycle hysteresis energy dissipation density and the mineral sensitivity constant to generate structural damage increment; S33: Obtain the initial integrity variable from the previous moment, subtract the structural damage increment from it, and obtain the current structural integrity coefficient; S34: Substitute the current structural integrity coefficient into the preset damage-contribution mapping model, determine the corresponding weight coefficient through piecewise linear discrimination logic, and generate the suction stress contribution factor.

4. The method for constructing an assessment model for the expansiveness of expansive soil according to claim 1, characterized in that, The calculation process for the percentage of volume expansion rate in S4 includes: S41: Use a monitoring probe embedded in the soil to obtain the current real-time matrix suction, and calculate the total overburden stress by combining the soil depth parameters and soil unit weight parameters. S42: The real-time matrix suction is reduced and corrected by the suction stress contribution factor, and the reduced value is superimposed with the total overlying stress to generate an equivalent effective stress. S43: Find the reference stress node adjacent to the equivalent effective stress in the preset effective stress-porosity standard relationship table, estimate the corresponding current porosity using the Lagrange interpolation polynomial, and calculate the difference between the current porosity and the initial porosity to obtain the target porosity change. S44: Calculate and output the percentage of volume expansion rate based on the ratio of the target porosity change to the initial porosity.

5. The method for constructing an assessment model for the expansiveness of expansive soil according to claim 2, characterized in that, The calculation process for the area of ​​the closed region in S22 is as follows: Obtain adjacent coordinate point pairs in the dehumidification path point set and adjacent coordinate point pairs in the moisture absorption path point set; For each pair of adjacent coordinate points, the product of the absolute value of the difference in matrix suction and the average value of volumetric water content is calculated to obtain the area of ​​the infinitesimal trapezoid. The areas of all the aforementioned trapezoidal micro-elements are summed, and the absolute value of the difference between the total area of ​​the dehumidification path and the total area of ​​the moisture absorption path is calculated to generate the area value of the closed region.

6. The method for constructing an assessment model for the expansiveness of expansive soil according to claim 3, characterized in that, The process of constructing the suction stress contribution factor in S34 adopts the following piecewise linear mapping function formula: ； in, This represents the suction stress contribution factor. Represents the current structural integrity coefficient. This represents the preset structural elastic limit threshold.

7. The method for constructing an assessment model for the expansiveness of expansive soil according to claim 4, characterized in that, The specific process for obtaining the target porosity change in S43 is as follows: Retrieve the preset effective stress-porosity ratio standard relationship table, and filter out three known stress nodes and their corresponding known porosity ratio data that are closest to the equivalent effective stress value; A second-order Lagrange interpolation basis function is constructed, and the three known stress nodes are used as interpolation nodes. The equivalent effective stress is used as the independent variable and input into the basis function. The current predicted porosity is obtained by weighted calculation. The numerical difference between the current predicted void ratio and the initial void ratio of the soil under initial conditions is calculated to generate the target void ratio change.

8. The method for constructing an assessment model for the expansiveness of expansive soil according to claim 4, characterized in that, The calculation process for the percentage of volume expansion rate in S44 uses the following formula: ； in, This represents the percentage of volume expansion rate. This represents the change in the target porosity. This represents the initial void ratio of the soil at the initial moment.

9. A system for constructing an assessment model for the expansiveness of expansive soil, characterized in that, The system is used to implement the method for constructing an expansive soil expansibility assessment model according to any one of claims 1-8, and the system includes: The state sequence construction module is used to collect the matrix suction and volumetric water content values ​​of expansive soil during the absorption and desiccation cycle, perform time synchronization and data cleaning operations, and construct a closed state evolution coordinate sequence with the beginning and end connected based on the cleaned data. The energy dissipation calculation module is used to call the trapezoidal numerical integration algorithm to perform area calculation on the polygonal region surrounded by the closed state evolution coordinate sequence, calculate the area value of the closed region, and convert the area value to obtain the single-cycle hysteresis energy dissipation density. The factor mapping generation module is used to obtain the mineral sensitivity constant and calculate the structural damage increment by combining the single-cycle hysteresis energy dissipation density. The initial integrity variable is updated using the structural damage increment to obtain the current structural integrity coefficient. The suction stress contribution factor is generated by a piecewise linear mapping function. The expansion assessment calculation module is used to collect real-time matrix suction and total overlying stress, calculate the equivalent effective stress using the suction stress contribution factor, obtain the target porosity change by combining the effective stress porosity ratio relationship table with the Lagrange interpolation algorithm, and calculate the output volume expansion rate percentage.

Images