A method for testing the permeability of a rock mass

Abstract

The application discloses a kind of rock-soil body permeability test methods, specifically related to rock-soil body material permeability detection technical field, for solving the problem that the sensitivity of traditional method is insufficient when testing high-performance concrete and bedrock interface due to low permeability and micro-pore structure of material, and the applicability deviation problem of macroscopic model;It is by constructing sealed test environment under keeping in situ stress state, applying multistage incremental fluid pressure step to stimulate interface micro-pore flow response, monitoring transient seepage signal and analyzing the nonlinear growth law of the decay rate required by initial signal amplitude to reach stability, combined with waveform form and the matching identification of preset typical seepage mode and form deviation quantification, finally based on multi-parameter fusion generates the equivalent permeability of interface to realize accurate and reliable permeability evaluation.

Description

Technical Field

[0001] This invention relates to the field of permeability testing technology for soil and rock materials, and more specifically, to a method for testing the permeability of soil and rock materials. Background Technology

[0002] In geotechnical and hydraulic engineering, accurate testing of the permeability of soil and rock masses and their interfaces with structures is crucial for assessing structural stability, leakage risk, and long-term durability. Traditional methods for testing soil and rock permeability, such as constant head or variable head permeability tests based on Darcy's law, are primarily designed for materials like soil, conventional rock, or ordinary concrete. These methods, which apply fluid pressure and measure steady-state flow rates, have formed a relatively mature testing system and are widely used for permeability evaluation in homogeneous or macroscopically fractured media.

[0003] However, when the aforementioned traditional testing methods are applied to novel composite interfaces composed of high-performance concrete and bedrock, the following problems arise: due to the extremely low permeability of high-performance concrete and the presence of microscopic pore and fissure structures at the interface with bedrock, it is difficult for the fluid to form a stable and detectable seepage signal under conventional test pressure gradients, resulting in significantly insufficient test sensitivity. At the same time, at this microscopic scale, the flow behavior of the fluid at the solid-wall boundary may deviate from the assumptions of Darcy's law, producing microscopic effects such as boundary slip, which leads to systematic deviations in the calculation results based on the macroscopic continuous medium model, resulting in an inaccurate and unreliable evaluation of the true permeability performance of the interface. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, the present invention provides a method for testing the permeability of soil and rock to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: A method for testing the permeability of rock and soil includes the following steps: S1. Construct a sealed test environment containing the interface between high-performance concrete and bedrock while maintaining in-situ stress. S2. Apply progressively increasing fluid pressure steps to the sealing test environment; S3. Monitor the transient seepage signal through the interface under each fluid pressure step; S4. Analyze the decay rate of the transient seepage signal required to reach stability under adjacent pressure steps, and the nonlinear growth law of the initial signal amplitude. S5. Identify the matching degree between the waveform shape of the transient seepage signal and a variety of preset typical interface seepage modes, and quantify the deviation of its shape from the preset reference seepage mode. S6. Based on the decay rate, nonlinear growth law, and morphological deviation, the equivalent penetration rate of the interface is generated.

[0006] Furthermore, a sealed test environment containing a high-performance concrete-bedrock interface was constructed while maintaining in-situ stress, including: A three-dimensional stress field is applied to the interface between high-performance concrete and bedrock using a confining pressure loading device to simulate the in-situ stress state. An elasto-plastic sealing layer is installed around the interface and interfacial sealing pressure is applied; The pressure inside the sealed test environment is kept in dynamic equilibrium with the three-dimensional stress field applied by the confining pressure loading device through the pressure balancing pipeline.

[0007] Furthermore, the elastoplastic sealing layer is made of a composite material that matches the deformation characteristics of high-performance concrete and bedrock, and the interface sealing pressure is directionally adjusted according to the principal stress direction of the three-dimensional stress field.

[0008] Furthermore, a series of progressively increasing fluid pressure steps are applied to the sealing test environment, including: Initial fluid pressure is applied to the sealing test environment using a pressure control device; Maintain the initial fluid pressure for a preset duration; The fluid pressure is gradually increased in stages according to the preset pressure increment to form a multi-stage progressive fluid pressure step. During each fluid pressure step transition, the rate of change of fluid pressure is kept constant.

[0009] Furthermore, the transient seepage signal across the interface is monitored for each fluid pressure step, including: A differential pressure sensor array is symmetrically arranged on both sides of the interface to capture local pressure gradient changes in the interface region; Set the sampling frequency of the differential pressure sensor array to be one order of magnitude higher than the frequency of fluid pressure step change; The voltage signal output by the differential pressure sensor array is recorded synchronously during the duration of each fluid pressure step. The voltage signal is converted into a fluid pressure data sequence as a transient seepage signal through a calibration curve.

[0010] Furthermore, the decay rate of the transient seepage signal required to reach stability under adjacent pressure steps, and the nonlinear growth law of the initial signal amplitude are analyzed, including: For each fluid pressure step, the transient seepage signal corresponding to the signal is identified by the moving window variance method to determine the time point when the signal reaches stability. The ratio of the time required for the signal to reach stability under adjacent pressure steps is calculated as the attenuation rate. Extract the initial amplitude of the transient seepage signal at the start of each fluid pressure step; Establish a nonlinear functional relationship between the initial amplitude and the step change of fluid pressure; The characteristic parameters in the nonlinear function relationship are fitted using the least squares method to obtain the nonlinear growth law.

[0011] Furthermore, establishing the nonlinear functional relationship between the initial amplitude and the fluid pressure step change includes: constructing a data sequence with the fluid pressure step value as the independent variable and the initial signal amplitude as the dependent variable; fitting the data sequence into an exponential or power-law function using the nonlinear least squares method; and determining the characteristic parameters in the function through the fitting process to complete the establishment of the nonlinear functional relationship.

[0012] Furthermore, the waveform morphology of the transient seepage signal is matched with several preset typical interface seepage modes to identify the degree of matching, and its deviation from the preset benchmark seepage mode is quantified, including: Extract the waveform morphology feature vector of the transient seepage signal; Calculate the similarity between the waveform morphology feature vector and the feature vectors corresponding to each preset typical interface seepage mode; Based on the similarity results, the optimal matching mode between the transient seepage signal and a variety of preset typical interface seepage modes is determined. The Euclidean distance between the waveform morphology feature vector and the feature vector corresponding to the preset benchmark seepage pattern is calculated as the morphological deviation.

[0013] Furthermore, based on the decay rate, nonlinear growth law, and morphological deviation, the equivalent permeability of the interface is generated, including: The decay rate, characteristic parameters of the nonlinear growth law, and morphological deviation are normalized to form the input parameter set. Substitute the input parameter set into the penetration rate calculation formula obtained through sample training; The equivalent permeability value of the interface is output by solving the permeability calculation formula.

[0014] Furthermore, substituting the input parameter set into the penetration rate calculation formula obtained through sample training includes: obtaining a sample dataset containing attenuation rate, nonlinear growth law characteristic parameters, morphological deviation degree and true penetration rate through standard sample testing in advance; using multivariate regression analysis to train the sample dataset to establish the penetration rate calculation formula; and substituting the normalized input parameter set into the penetration rate calculation formula to calculate the equivalent penetration rate value.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. By constructing a sealed test environment that maintains in-situ stress, the actual working conditions of the high-performance concrete-bedrock interface were effectively simulated, making the test results closer to engineering practice. Employing a multi-stage, incremental fluid pressure step loading method, the seepage response of the interfacial micropores can be excited at different pressure levels, overcoming the difficulty of capturing weak signals in low-permeability interface tests using traditional methods. By monitoring transient seepage signals and analyzing their attenuation time to reach stability, the influence of the interfacial microstructure on the seepage process can be sensitively reflected, providing a dynamic response basis for evaluating interfacial permeability characteristics.

[0016] 2. By analyzing the nonlinear growth law of the initial signal amplitude, the non-Darcy flow characteristics existing in the interfacial seepage process were revealed, effectively avoiding the systematic bias of traditional linear models in microscale seepage analysis. Combining waveform morphology with typical seepage patterns and quantitative evaluation of morphological deviation, a comprehensive characterization of interfacial seepage features was achieved. Finally, the equivalent permeability generated based on multi-parameter fusion accurately reflects the comprehensive permeability performance of the interface at the microscale, providing reliable technical support for the quality evaluation of new material interfaces. Attached Figure Description

[0017] Figure 1 This is a flowchart of a method for testing the permeability of soil and rock according to the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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.

[0019] Example: Figure 1 This invention provides a method for testing the permeability of soil and rock masses, which includes the following steps: S1. Construct a sealed test environment containing the interface between high-performance concrete and bedrock while maintaining in-situ stress. S2. Apply progressively increasing fluid pressure steps to the sealing test environment; S3. Monitor the transient seepage signal through the interface under each fluid pressure step; S4. Analyze the decay rate of the transient seepage signal required to reach stability under adjacent pressure steps, and the nonlinear growth law of the initial signal amplitude. S5. Identify the matching degree between the waveform shape of the transient seepage signal and a variety of preset typical interface seepage modes, and quantify the deviation of its shape from the preset reference seepage mode. S6. Based on the decay rate, nonlinear growth law, and morphological deviation, the equivalent penetration rate of the interface is generated.

[0020] When constructing a sealed test environment containing the interface between high-performance concrete and bedrock, a three-dimensional stress field is first applied to the interface using a confining pressure loading device to simulate the in-situ stress state. This confining pressure loading device employs three independently controlled hydraulic actuators, applying normal pressure to the test components along the vertical and two horizontal directions. The magnitude of the three-dimensional stress field is determined based on in-situ stress test data from the engineering geological survey report. The vertical stress is calculated based on the self-weight of the overlying rock strata, and the stresses in the two horizontal directions are obtained by multiplying the lateral pressure coefficient by the vertical stress. The lateral pressure coefficient is obtained through in-situ hydraulic fracturing tests or laboratory rock mechanics tests, typically ranging from 0.3 to 0.8. The output pressure of each hydraulic actuator is controlled in a closed loop using a precision pressure sensor, achieving a control accuracy of ±1 kPa to ensure the stability of the three-dimensional stress field during testing. The pressure output range of the hydraulic actuators is set according to test requirements, for example, a minimum output pressure of 0.1 MPa and a maximum output pressure of 10 MPa. The sampling frequency of the pressure sensor is set to 100 Hz to ensure real-time monitoring of stress changes. The application of the three-dimensional stress field includes an initialization phase and a stabilization phase. During the initialization phase, the stress in each direction is gradually increased to the target value. During the stabilization phase, the stress is kept constant, and the duration is set according to the test requirements, for example, no less than 30 minutes. The uniformity of the stress field is verified by an array of strain gauges arranged around the interface. The spacing between the strain gauge measurement points is 50 mm, and the deviation between the measured data and the theoretical value is controlled within 5%.

[0021] An elasto-plastic sealing layer is installed around the interface, and interfacial sealing pressure is applied. The elasto-plastic sealing layer is made of a composite material matched to the deformation characteristics of high-performance concrete and bedrock. This composite material is formed by hot pressing with EPDM rubber as the matrix, incorporating carbon fiber reinforcement and silicone additives. For example, the carbon fiber content is controlled between 15% and 25%, and the silicone additive content is controlled between 5% and 10%. The thickness of the sealing layer is determined based on the expected maximum deformation of the interface, for example, a thickness of 5 mm to 15 mm. The interfacial sealing pressure is applied through a dedicated sealing pressure control system, which includes a pressure source, a regulating valve, and an annular pressure bladder. The annular pressure bladder is positioned close to the outer side of the elasto-plastic sealing layer, and the hydraulic oil pressure injected into the pressure bladder is controlled by the regulating valve to bring the interfacial sealing pressure to the set value. The set value of the interface sealing pressure is determined according to the principal stress direction of the three-dimensional stress field. For example, the interface sealing pressure in the direction of maximum principal stress is 80% to 120% of the stress in that direction, the interface sealing pressure in the direction of intermediate principal stress is 60% to 90% of the stress in that direction, and the interface sealing pressure in the direction of minimum principal stress is 40% to 70% of the stress in that direction. The interface sealing pressure in each direction is adjusted separately through an independent pressure control loop. The response time of the pressure control loop is less than 1 second to ensure the matching of the sealing effect with the stress state. The installation process of the sealing layer includes pretreatment, positioning, and curing steps. The pretreatment stage cleans the interface surface, the positioning stage ensures that the sealing layer is in close contact with the interface, and the curing stage keeps it at room temperature for 24 hours to achieve optimal sealing performance.

[0022] A pressure balancing pipeline maintains a dynamic balance between the internal pressure of the sealed test environment and the three-dimensional stress field applied by the confining pressure loading device. The pressure balancing pipeline is made of stainless steel with an inner diameter of 6 mm, and its two ends connect to the pressure compensator of the sealed test environment cavity and the confining pressure loading device, respectively. A differential pressure sensor and an automatic regulating valve are installed on the pressure balancing pipeline. The differential pressure sensor monitors the pressure difference between the internal pressure of the sealed test environment and the three-dimensional stress field applied by the confining pressure loading device in real time. When the pressure difference exceeds a set threshold, the automatic regulating valve adjusts its opening based on the feedback signal from the differential pressure sensor, ensuring that the internal pressure of the sealed test environment remains dynamically balanced with the three-dimensional stress field applied by the confining pressure loading device. This set threshold is determined according to the required test accuracy, for example, controlled within ±2 kPa. The differential pressure sensor has a measurement range of -10 kPa to +10 kPa and an accuracy of ±0.1 kPa. The opening control of the automatic regulating valve is based on a proportional-integral-derivative algorithm, with a proportional coefficient set to 0.5, an integral time set to 10 seconds, and a derivative time set to 1 second to ensure rapid response and stability. The layout of the pressure balancing pipeline avoids sharp-angle bends, with a bend radius greater than 50 mm to reduce flow resistance. The dynamic balance maintenance process includes initial calibration and operational adjustment. The initial calibration stage involves adjusting the zero point under no-fluid pressure, while the operational adjustment stage involves fine-tuning the parameters based on real-time data.

[0023] The elastoplastic sealing layer is made of a composite material that matches the deformation characteristics of high-performance concrete and bedrock. Its elastic modulus is determined through material tests, for example, the value ranges from 1.5 GPa to 3.5 GPa. The ratio of the elastic modulus of the sealing layer to that of high-performance concrete is controlled between 0.1 and 0.3, and the ratio of the elastic modulus of the sealing layer to that of bedrock is controlled between 0.05 and 0.2. The interface sealing pressure is directionally adjusted according to the principal stress direction of the three-dimensional stress field. In the adjustment process based on the principle of stress tensor decomposition, the three-dimensional stress field is decomposed into three principal stress directions, and the corresponding interface sealing pressure is calculated according to the stress magnitude in each direction. The principal stress direction is determined by an eigenvalue analysis method, where eigenvectors represent the principal stress directions and eigenvalues represent the magnitudes of principal stresses. The interface sealing pressure is calculated by a linear interpolation method, and the pressure value is determined according to the stress proportional relationship. For example, when the maximum principal stress is 5 MPa, the interface sealing pressure ranges from 4 MPa to 6 MPa. The step size of pressure adjustment is 0.1 MPa, and the adjustment frequency is 1 Hz to ensure a smooth transition. The deformation characteristics of the sealing layer are verified by a stress-strain curve, and the curve data are obtained through a universal testing machine, with the strain rate controlled at 0.001 s -1 or less. The sealing performance of the sealed test environment is verified by a leak detection test. For example, the pressure is maintained for 1 hour under an applied pressure of 1 MPa, and it is qualified if the pressure drop does not exceed 0.01 MPa. All parameter adjustments are based on experimental data and theoretical models to ensure the feasibility and repeatability of the scheme.

[0024] When applying progressively increasing fluid pressure steps to the sealing test environment, an initial fluid pressure is first applied via a pressure control device. This device consists of a high-pressure gas source, a precision pressure regulating valve, and a digital pressure controller. The high-pressure gas source provides compressed gas as the pressure medium, the precision pressure regulating valve precisely adjusts the output pressure, and the digital pressure controller, based on a microprocessor, implements pressure setting and feedback control. The initial fluid pressure is determined based on the permeability characteristics of the test object. For example, for low-permeability interfaces, the initial fluid pressure is between 0.1 MPa and 0.5 MPa. The specific value is calculated using prior permeability test data or empirical formulas. The calculation process considers the porosity and fracture development of the interface material. Porosity is obtained through core laboratory measurements, and fracture development is assessed through ground-penetrating radar scanning or microscopic observation. The digital pressure controller uses a proportional-integral-derivative (PI-DE) control algorithm, with a proportional gain set to 0.8, an integral time set to 15 seconds, and a derivative time set to 2 seconds to ensure stable pressure output. The connecting pipeline between the pressure control device and the sealing test environment is made of stainless steel, with an inner diameter of 8 mm and a length not exceeding 2 meters, to minimize pressure loss and response delay. The initial fluid pressure application process includes a pre-pressurization phase and a stabilization phase. During the pre-pressurization phase, the pressure is gradually increased to 80% of the target value. During the stabilization phase, the pressure is slowly adjusted to the final set value, with the duration set according to the system response characteristics, for example, no less than 10 minutes. The output accuracy of the pressure control device is confirmed by a calibration certificate, for example, an accuracy of ±0.5% of full scale, ensuring accurate application of the initial fluid pressure.

[0025] The initial fluid pressure is maintained for a preset duration. This preset duration is determined based on the response characteristics of the transient seepage signal, for example, ranging from 5 to 30 minutes. Specific values ​​are obtained through pre-tests or numerical simulation analysis. Pre-tests observe signal stability at different durations, while numerical simulations use finite element software to establish a fluid flow model and analyze pressure propagation time. During the maintenance of the initial fluid pressure, the digital pressure controller monitors the pressure value in real time and compares it with the set value. When the deviation exceeds the allowable range, the output is automatically adjusted. The allowable range is set based on test accuracy requirements, such as pressure fluctuations controlled within ±0.5%. Pressure stability during the maintenance phase is verified through sampling data from the pressure sensor, with a sampling frequency set to 10 Hz and a data recording interval of 1 second. If pressure fluctuations continue to exceed the allowable range, the system triggers an alarm and automatically adjusts the pressure regulating valve opening. The adjustment magnitude is calculated based on the deviation magnitude and rate of change; for example, for every 0.1% increase in deviation, the pressure regulating valve opening increases by 0.5%. At the end of the maintenance duration, the system records a pressure stabilization flag and prepares for the next step. The preset duration is adjusted based on changes in ambient temperature and medium properties. For example, the duration is adjusted by 1 minute for every 5 degrees Celsius change in temperature. The viscosity of the medium is measured by a viscometer and used to correct the duration.

[0026] The fluid pressure is gradually increased in preset increments to form a multi-stage, progressively increasing fluid pressure step. The pressure increment is set based on test sensitivity and equipment capability; for example, the increment value is 0.2 MPa to 0.5 MPa. The specific value is determined through step pressure testing, which evaluates signal detectability at different increments. The number of stages in the multi-stage progressive fluid pressure step is set according to the test range and resolution requirements; for example, 5 to 10 stages. The total pressure range covers from the initial pressure to the maximum allowable pressure, which is determined through material strength testing. The pressure increase process is controlled by a digital pressure controller program, which is programmed as a cyclic step. Each step includes a pressure increase and a stabilization confirmation. The pressure increment is calculated using linear or nonlinear rules. For example, under linear rules, a fixed increment is added per stage; under nonlinear rules, the increment size is dynamically adjusted based on the response of the preceding signal. The dynamic adjustment is based on the attenuation rate or the amplitude change rate, with a change rate threshold set at 10%. The formation of multi-stage progressive fluid pressure steps ensures that the fluid permeation process at the interface transitions from low-speed to high-speed flow, thereby covering the test requirements under different flow regimes. The step sequence is validated through pressure-time curve analysis. Curve smoothness index is used to evaluate the quality of the step, for example, the smoothness index requires that the pressure change is not abrupt.

[0027] During each fluid pressure step transition, the rate of change of fluid pressure is kept constant. The rate of change is set based on interface response characteristics and equipment performance, for example, ranging from 0.05 MPa per minute to 0.2 MPa per minute. Specific values ​​are optimized through rate testing, which observes signal distortion at different rates. The rate of change is controlled by the slope limiting function of the digital pressure controller. The slope limit value is calculated based on the set rate; for example, a rate of 0.1 MPa per minute corresponds to a slope limit of 0.00167 MPa per second. During the transition, the constant pressure change rate is maintained through real-time feedback adjustment. The feedback signal comes from a high-frequency pressure sensor with a sampling frequency of 50 Hz, and the adjustment algorithm uses feedforward control to compensate for system delay. If the rate deviation exceeds an allowable threshold, for example, a threshold set at ±5%, the system automatically adjusts the output flow rate. Flow rate adjustment is achieved through a proportional valve, the opening of which is proportional to the rate deviation. Maintaining the rate of change ensures a smooth transition of fluid pressure steps, avoids damage to the interface structure from pressure shocks, and ensures the analyzability of transient seepage signals. Rate constancy was verified through differential pressure-time data calculations, with the differential value fluctuation controlled within ±0.01 MPa per second. Parameter records for the entire conversion process, including rate, time, and pressure values, were used for subsequent data analysis and quality control.

[0028] When monitoring transient seepage signals across an interface under each fluid pressure step, a differential pressure sensor array is first symmetrically arranged on both sides of the interface to capture local pressure gradient changes in the interface region. The differential pressure sensor array consists of multiple differential pressure sensor units, each including two pressure ports and a differential measurement circuit. The pressure ports are connected to the interface region via flexible conduits, and the differential measurement circuit operates based on the Wheatstone bridge principle. The arrangement of the differential pressure sensor array is determined according to the interface geometry and the expected seepage path. For example, in a rectangular interface, the sensor units are arranged in a grid pattern with a row and column spacing of 50 mm; in a circular interface, the sensor units are uniformly distributed radially and circumferentially, with a radial spacing of 30 mm and a circumferential spacing of 45 degrees. The capture of local pressure gradient changes is achieved by measuring the pressure difference at symmetrical points on both sides of the interface. The measurement range of the pressure difference is set according to the test pressure, for example, -10 kPa to +10 kPa, with an accuracy of ±0.1% of full scale. The installation process of the differential pressure sensor array includes positioning, fixing, and sealing steps. Positioning uses a laser rangefinder to ensure symmetry; fixing employs a magnetic base or adhesive; and sealing uses silicone gaskets to prevent leakage. Sensor unit calibration is performed before installation, including zero-point calibration and full-scale calibration. Zero-point calibration adjusts the output to zero under no-pressure conditions, while full-scale calibration adjusts the output to match the standard value under a known pressure. Data on local pressure gradient changes are acquired through the real-time output of the sensor array for subsequent analysis of interfacial seepage characteristics.

[0029] The sampling frequency of the differential pressure sensor array is set to be one order of magnitude higher than the frequency of the fluid pressure step change. The sampling frequency setting is based on the Nyquist sampling theorem to ensure distortion-free signal acquisition. The fluid pressure step change frequency is calculated using the operating parameters of the pressure control device. For example, if the duration of the pressure step change is 5 minutes, the change frequency is 0.0033 Hz, therefore the sampling frequency is set to 0.033 Hz. The specific value of the sampling frequency is implemented through the settings of the data acquisition card. The data acquisition card is based on FPGA technology and supports programmable sampling rates. For example, the sampling frequency can be set to 1 Hz to meet the requirement of being one order of magnitude higher than the change frequency. The adjustment of the sampling frequency is based on signal frequency components and system response. For example, the main frequency components of the signal are determined through spectrum analysis, and the sampling frequency is at least twice that of the highest frequency component. The clock synchronization function of the data acquisition card ensures that all sensor units sample synchronously. The clock source uses a crystal oscillator with a frequency stability of ±0.001%. The sampling frequency is verified by recording a standard signal, such as inputting a sine wave signal and checking the waveform integrity of the acquired data. If the sampling frequency is insufficient, the system automatically prompts for adjustment in 0.1 Hz steps until the requirements are met. During the sampling process, the data is buffered to memory in real time. The buffer size is set according to the sampling frequency and test duration. For example, if the sampling frequency is 1 Hz and the test duration is 1 hour, the buffer size is at least 3600 data points.

[0030] The voltage signal output by the differential pressure sensor array is recorded synchronously for each fluid pressure step duration. Synchronization is achieved through a timestamp mechanism provided by the global clock of the data acquisition system, with a clock resolution of 1 millisecond. Voltage signal recording is based on analog-to-digital conversion (ADC) technology, using a 16-bit ADC with an input range of -10V to +10V and a conversion rate of 100 kilovolts per second. The duration of each fluid pressure step is determined based on the output signals of the pressure control device, such as the start and end times of the pressure step, triggered by recording via a digital input channel. The voltage signal recording process includes initialization, acquisition, and storage steps. Initialization sets the acquisition parameters; acquisition continuously reads the sensor output; and storage writes the data to non-volatile memory. The recorded data format includes a timestamp, voltage value, and sensor identifier, stored in binary format to reduce storage space. Synchronization accuracy is ensured by a hardware trigger signal issued by the pressure control device during step transitions, with a trigger delay of less than 1 millisecond. The integrity of the voltage signal is verified through checksums, such as attaching a CRC checksum to each data packet. During recording, the voltage range is monitored in real time. If it exceeds the expected range, an alarm is triggered. The expected range is set according to the sensor specifications; for example, the voltage value should be between -5 volts and +5 volts. The recorded data is subsequently used for signal analysis and conversion.

[0031] The voltage signal is converted into a fluid pressure data sequence as a transient seepage signal using a calibration curve. The calibration curve is established through a laboratory calibration process, which involves applying a series of known pressures and recording the corresponding voltage outputs. These known pressures are provided by a standard pressure gauge with an accuracy of ±0.05% of the reading. The calibration curve is established using a polynomial fitting method, such as quadratic polynomial fitting, with the polynomial coefficients calculated using the least squares method. The conversion from voltage signal to fluid pressure data sequence is achieved by querying the calibration curve. The conversion algorithm is based on interpolation calculations, such as linear interpolation or spline interpolation. The output of the fluid pressure data sequence includes the pressure value and the corresponding timestamp; the length of the data sequence is the same as the recorded length of the voltage signal. The accuracy of the conversion process is evaluated using calibration residuals, which are required to be less than 0.1% of full scale. Periodic validation of the calibration curve is performed through repeated calibration, with a validation cycle of 6 months or after every 100 tests. The transient seepage signal data sequence is used in subsequent analysis steps, such as calculating attenuation rates and morphological characteristics. During the conversion process, data preprocessing includes filtering and noise reduction. Low-pass filters are used, with the cutoff frequency set according to signal characteristics, for example, 0.5 Hz. Data post-processing includes unit conversion and format standardization, such as standardizing pressure units to kilopascals and time units to seconds. The entire conversion process ensures data accuracy and consistency, providing reliable input for permeability calculations.

[0032] When analyzing the decay rate of transient seepage signals requiring time to stabilize under adjacent pressure steps and the nonlinear growth law of the initial signal amplitude, the moving window variance method is first used to identify the time point when the signal reaches stability for each fluid pressure step. The moving window variance method is based on the calculation of data variance within a sliding time window. The window size is determined according to the characteristics of the transient seepage signal; for example, the window size ranges from 10 to 60 seconds, and the specific value is optimized through pre-experiments. The pre-experiments evaluate the smoothness of the variance curve under different window sizes. The window sliding step is set to one-tenth of the window size; for example, when the window size is 30 seconds, the sliding step is 3 seconds. Variance is calculated using the sample variance formula, taking the variance value for each data point within the window. The time point when the signal reaches stability is determined when the variance value falls below a set threshold. This threshold is calculated based on the background noise level; for example, the threshold is 1.5 times the background noise variance, obtained from signal recordings under no-pressure conditions. The application of the moving window variance method ensures that the identification of the stable time point is based on statistical characteristics, avoiding subjective judgment. During the identification process, data preprocessing includes outlier removal, defined as data points exceeding three standard deviations from the mean. The standard deviation is calculated based on the entire signal sequence. Validation at stable time points is performed through repeated analysis, for example, three independent analyses of the same signal, with the time point differences required to be less than 5%.

[0033] The attenuation rate is calculated as the ratio of the time required for the signal to stabilize under adjacent pressure steps. The selection of adjacent pressure steps is based on the pressure increase order; for example, the first and second steps form one pair, and the second and third steps form another. The time required for the signal to stabilize is obtained from the moving window variance method output, and the ratio is calculated by dividing the stabilization time of the subsequent step by the stabilization time of the previous step. The attenuation rate typically ranges from 0.1 to 10, with the specific value reflecting the changing trend of the interface penetration characteristics. The ratio calculation process includes data verification; for example, the stabilization time value must be positive and not exceed the total step duration. The application of the attenuation rate is based on a physical model; for example, a ratio less than 1 indicates a shortened stabilization time, which may be related to the propagation of interface cracks. If an outlier is encountered during the calculation, such as a ratio exceeding a reasonable range, the system automatically checks the original data and recalculates. The attenuation rate record includes the corresponding step pair identifier and calculation timestamp for subsequent analysis. Matching adjacent pressure steps ensures coverage of all consecutive steps; for example, for n steps, n-1 attenuation rate values ​​are generated. The statistical characteristics of the decay rate are obtained by averaging multiple tests, for example, by performing three repeated tests and taking the average value.

[0034] The initial amplitude of the transient seepage signal at the onset of each fluid pressure step is extracted. The initial amplitude is defined based on the characteristics of the signal's initial phase, such as the maximum amplitude within the first complete cycle after the fluid pressure step or the first peak value before stabilization. The extraction process is implemented using a digital signal processing algorithm. The algorithm first locates the step onset point, which is synchronously determined by a trigger signal from a pressure control device. The initial amplitude is calculated using a peak detection method, such as finding the global maximum value within a 0.1-second time window after the onset point. The amplitude unit is consistent with the transient seepage signal unit, such as kilopascals for pressure. The extracted initial amplitudes are stored as a sequence, with one amplitude corresponding to each step. Data quality control includes amplitude range checks, such as ensuring the amplitude is within the sensor's range; values ​​outside this range are marked as invalid. The repeatability of the initial amplitude is verified through multiple extractions, such as extracting the same signal three times with a difference of less than 2%. During extraction, signal preprocessing includes baseline correction, calculated by averaging shortly recorded data before the step. The initial amplitudes are then used to provide input data for subsequent nonlinear analysis.

[0035] A nonlinear functional relationship is established for the initial amplitude as a function of fluid pressure step changes. This step constructs a data sequence with the fluid pressure step value as the independent variable and the initial signal amplitude as the dependent variable. The data sequence construction involves collecting all fluid pressure step values ​​and their corresponding initial amplitudes. The fluid pressure step values ​​are obtained from the pressure control device records, and the initial amplitudes are obtained from the extraction results. The nonlinear functional relationship is established by fitting the data sequence to an exponential or power-law function using the nonlinear least squares method. The function type is chosen based on the data distribution characteristics; for example, an exponential function describes the exponential growth of amplitude with pressure, while a power-law function describes a power-law relationship. The nonlinear least squares method is implemented using an iterative optimization algorithm, such as the Levenberg-Marquardt algorithm, with the iteration stopping condition set to a parameter change of less than 0.001%. The characteristic parameters in the function are determined through the fitting process; for example, an exponential function includes a scaling parameter and an exponential parameter, while a power-law function includes a coefficient parameter and a power-law parameter. The nonlinear functional relationship is completed by evaluating the goodness of fit, for example, requiring a coefficient of determination greater than 0.9. Data sequence preprocessing includes normalization using a min-max scaling method, mapping the data to a range of 0 to 1. The establishment of nonlinear functional relationships provides a mathematical model for permeability analysis.

[0036] The least squares method is used to fit characteristic parameters in a nonlinear function relationship to represent the nonlinear growth law. The application of the least squares method is based on the principle of minimizing the sum of squared errors, where the error is defined as the difference between the measured amplitude and the predicted function value. The fitting process for the characteristic parameters includes initializing parameter values ​​using linear regression results or empirical values. In iterative optimization, the parameter adjustment step size is calculated based on the gradient direction, and the gradient is approximated by numerical differentiation. The evaluation of the fitting results includes residual analysis, requiring the residuals to follow a normal distribution. The nonlinear growth law is characterized by the characteristic parameter values; for example, the exponential parameter value reflects the growth rate. Post-processing during the fitting process includes outlier removal, defined as data points with residuals exceeding three standard deviations. The physical meaning of the characteristic parameters is related to the interfacial seepage mechanism; for example, changes in parameter values ​​may indicate changes in the interfacial pore structure. The software implementation of the least squares fitting uses scientific computing libraries, such as Python's SciPy library or MATLAB's curve fitting tools. The fitting results are stored as a parameter vector for subsequent permeability calculations. The entire fitting process ensures the accurate quantification of the nonlinear growth law, providing a basis for evaluating interfacial characteristics.

[0037] When identifying and quantifying the morphological deviation between the waveform morphology of transient seepage signals and preset typical interface seepage modes, the waveform morphology feature vector of the transient seepage signals is first extracted. The extraction of the waveform morphology feature vector is based on multiple feature dimensions, including time-domain and frequency-domain features. Time-domain features include signal rise time, peak time, and fall time, while frequency-domain features are obtained through Fast Fourier Transform (FFT) to obtain the main frequency components and power spectral density. The dimensions of the feature vector are determined according to the analysis requirements; for example, eight key features may be selected to form an eight-dimensional feature vector, including rise time, peak time, fall time, overshoot, steady-state value, main frequency, bandwidth, and harmonic distortion. Each feature is calculated using a specific method. For example, rise time is defined as the time required for the signal to rise from 10% to 90% of its final value; peak time is defined as the time from the start of a step jump to reaching the peak value; fall time is defined as the time from the peak value to the steady-state value; overshoot is defined as the percentage by which the peak value exceeds the steady-state value; the steady-state value is calculated using the average value of the signal's final segment; the dominant frequency is determined by the frequency corresponding to the largest amplitude in the spectral analysis; the bandwidth is calculated by the frequency range where the power spectrum drops by 3 dB; and the harmonic distortion is calculated using the total harmonic distortion rate. During extraction, data preprocessing includes signal alignment and normalization. Signal alignment is based on the step jump start point, and normalization scales the amplitude to the range of 0 to 1. The feature vector is stored as an array, with each feature corresponding to one array element for subsequent matching calculations.

[0038] The similarity between the waveform morphology feature vector and the feature vectors corresponding to various preset typical interface seepage modes is calculated. These preset typical interface seepage modes include uniform seepage mode, fracture seepage mode, pore seepage mode, and layered seepage mode. The feature vectors for each mode are obtained through historical test data or numerical simulation. For example, the feature vectors of the uniform seepage mode are characterized by short rise time, small overshoot, and high dominant frequency; the feature vectors of the fracture seepage mode are characterized by short rise time, large overshoot, and medium dominant frequency; the feature vectors of the pore seepage mode are characterized by long rise time, small overshoot, and low dominant frequency; and the feature vectors of the layered seepage mode are characterized by medium rise time, medium overshoot, and medium dominant frequency. The similarity calculation uses the cosine similarity method, calculating the cosine value of the angle between two feature vectors, ranging from -1 to 1. The closer the value is to 1, the higher the similarity. The cosine similarity calculation process includes vector dot product calculation and modulus calculation. The dot product is the sum of the products of each dimension of the two vectors, and the modulus is the square root of the sum of the squares of each dimension. Before similarity calculation, the feature vectors are standardized by subtracting the mean and dividing by the standard deviation for each feature. The mean and standard deviation are calculated from the training dataset. The similarity calculation results are stored as a similarity matrix, with rows corresponding to test signals and columns corresponding to preset patterns. The similarity calculation is validated by testing with known pattern signals; for example, the similarity between the input uniform seepage signal and the uniform seepage pattern should be greater than 0.9.

[0039] The optimal matching mode is determined based on the similarity results between the transient seepage signal and several preset typical interface seepage modes. The determination of the optimal matching mode is based on the principle of highest similarity, selecting the preset mode with the highest similarity value as the optimal matching mode. The similarity results are read from the similarity matrix. For example, if a test signal has similarity values ​​of 0.85, 0.92, 0.78, and 0.88 with four preset modes, the fracture seepage mode corresponding to a similarity of 0.92 is selected as the optimal matching mode. During the determination process, a similarity threshold is set, for example, a threshold of 0.7. When all similarity values ​​are below the threshold, the mode is marked as unmatched. The similarity threshold is set based on historical data analysis, for example, analyzing the similarity distribution of 100 known mode signals and taking the 5th percentile of the distribution as the threshold. The output of the optimal matching mode includes the mode type and the matching confidence score. The matching confidence score is calculated by normalizing the similarity values, for example, scaling the similarity values ​​to a range of 0 to 100. The application of the optimal matching pattern provides a reference for subsequent interface characteristic analysis; for example, matching a fracture seepage pattern may indicate the presence of fracture development at the interface. Records of the results, including timestamps and test conditions, are used for quality traceability.

[0040] The morphological deviation is calculated as the Euclidean distance between the waveform morphology feature vector and the feature vector corresponding to a preset benchmark seepage pattern. The preset benchmark seepage pattern is a uniform seepage pattern, as it represents the ideal seepage state. The Euclidean distance is calculated based on the square root of the sum of the squares of the differences in each dimension of the feature vector. For example, for an n-dimensional feature vector, the Euclidean distance is equal to the square root of the sum of the squares of the differences in each dimension. The unit of the morphological deviation is consistent with the unit of the feature vector; for example, the unit is seconds when the feature is time, and hertz when the feature is frequency. The calculation process of the Euclidean distance includes difference calculation, square calculation, summation calculation, and square root calculation, using double-precision floating-point numbers to ensure accuracy. The range of the morphological deviation value is determined according to the feature scale; for example, a typical value is between 0 and 10, with a larger value indicating a greater deviation from the benchmark pattern. Before calculation, the feature vector is standardized using the same method as the similarity calculation to ensure dimensional consistency. Applications of the morphological deviation include interface quality assessment; for example, indicating an interface anomaly when the deviation exceeds a set threshold. The threshold is determined based on statistical methods, such as analyzing the deviation distribution of normal interfaces and taking the average plus three times the standard deviation as the threshold. The record of morphological deviation is used together with other parameters to calculate permeability and complete the comprehensive evaluation of interface characteristics.

[0041] When generating the equivalent permeability of the interface based on the decay rate, nonlinear growth law, and morphological deviation, the decay rate, characteristic parameters of the nonlinear growth law, and morphological deviation are first normalized to form an input parameter set. The normalization process uses a minimum-maximum scaling method, linearly transforming each parameter value to the range of 0 to 1. The normalization of the decay rate is based on its theoretical value range; for example, the decay rate is typically between 0.1 and 10, so the minimum value during normalization is 0.1, and the maximum value is 10. The characteristic parameters of the nonlinear growth law have their normalization range determined according to their type. For example, the characteristic parameters of an exponential function include a scale parameter and an exponential parameter; the typical value range of the scale parameter is 0.5 to 5, and the typical value range of the exponential parameter is 0.1 to 2. The characteristic parameters of a power-law function include a coefficient parameter and a power parameter; the typical value range of the coefficient parameter is 1 to 10, and the typical value range of the power parameter is 0.5 to 3. Normalization of morphological deviation is based on historical test data statistics, such as analyzing the morphological deviation distribution of multiple interfaces, taking the minimum value as 0 and the maximum value as 20. The specific calculation for normalization is to subtract the minimum value from each parameter value and then divide by the difference between the maximum and minimum values. The input parameter set is formed by combining the normalized decay rate, the feature parameters in the normalized nonlinear growth law, and the normalized morphological deviation in a fixed order into a vector form, for example, the vector order is normalized decay rate, normalized scale parameter, normalized exponent parameter, and normalized morphological deviation. During normalization, if a parameter value exceeds a preset range, it is truncated; for example, a parameter value less than the minimum is taken as 0, and a parameter value greater than the maximum is taken as 1. The dimension of the input parameter set is determined according to the number of features; for example, when using decay rate, two feature parameters, and morphological deviation, the input parameter set is a four-dimensional vector. Verification of the normalization process is performed through reverse calculation; for example, the normalized value is converted back to the original value to check for error, which is required to be less than 1%.

[0042] The input parameter set is substituted into the permeability calculation formula obtained through sample training. The permeability calculation formula is obtained through a sample training process. Sample training pre-obtains a sample dataset containing attenuation rate, nonlinear growth law characteristic parameters, morphological deviation, and true permeability through standard sample testing. Standard sample testing uses standard rock samples with known permeability. The true permeability of the standard rock samples is measured using steady-state or transient methods. For example, the steady-state method uses Darcy's law to calculate permeability, while the transient method uses pressure pulse decay. The construction of the sample dataset involves collecting test data from multiple standard samples. Each sample's data includes attenuation rate, nonlinear growth law characteristic parameters, morphological deviation, and true permeability. The number of samples is determined according to statistical requirements, such as at least 30 samples to ensure representativeness. Preprocessing of the sample dataset includes outlier removal, where outliers are defined as data points deviating from other data points by more than three standard deviations. The permeability calculation formula is established using multiple regression analysis, which is based on the least squares principle and fits the mathematical relationship between the input parameters and the true permeability. The specific steps of multiple regression analysis include selecting a regression model, such as a linear or multinomial model. Model selection is determined through cross-validation, which divides the dataset into training and test sets. The training set is used to fit the model, and the test set is used to evaluate the model's predictive accuracy. The coefficients in the regression model are obtained by minimizing the sum of squared prediction errors. The minimization process uses the gradient descent algorithm, with a learning rate of 0.01 and 1000 iterations. The permeability calculation formula is, for example, a linear equation, where the equivalent permeability equals coefficient 1 multiplied by the normalized decay rate, coefficient 2 multiplied by the normalized scale parameter, coefficient 3 multiplied by the normalized exponential parameter, coefficient 4 multiplied by the normalized morphological deviation, and a constant term. The formula is validated using the coefficient of determination, which must be greater than 0.9. When substituting the normalized input parameter set into the permeability calculation formula, each element of the input parameter set corresponds to a variable in the formula. The substitution process is numerical, using methods such as matrix multiplication or element-wise operations. The equivalent permeability value is obtained from the calculation. The equivalent permeability value is a dimensionless value or a value with units, such as millidarcy.

[0043] The equivalent permeability value of the interface is output by solving the permeability calculation formula. The solution process is a mathematical calculation; the values ​​of the input parameter set are substituted into the permeability calculation formula, and arithmetic operations are performed to obtain the result. Arithmetic operations include addition, multiplication, and possibly higher-order operations, depending on the form of the formula. For example, linear formulas involve multiplication and addition, while polynomial formulas involve exponentiation. During the solution process, numerical precision is guaranteed by using double-precision floating-point numbers, and the calculation error is controlled within 0.1%. The output format of the equivalent permeability value is a numeric value, retaining three decimal places, for example, an output value of 1.234 millidarcy. The output value is verified by comparing it with known samples; for example, the parameter set of a known sample is input to check whether the output matches the true permeability. If the output value is outside a reasonable range, such as a negative or abnormally large permeability value, a recalculation or a check of the input parameters is triggered. The equivalent permeability value is used to evaluate the permeability characteristics of the interface; for example, a higher value indicates better permeability. The output process includes data recording and display. Recorded data includes test time, input parameter values, and equivalent permeability values, displayed via a graphical interface or digital output device. The solution process is automated through a computer program written in Python or MATLAB, ensuring repeatable calculations. The entire solution process ensures the accuracy and reliability of the equivalent permeability values, providing a basis for engineering decisions.

[0044] All calculations involved in the embodiments are dimensionless numerical calculations, and the preset parameters and thresholds in the calculations are set by those skilled in the art according to the actual situation.

[0045] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0046] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and inventive constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0047] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0048] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0049] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0050] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for testing the permeability of rock and soil, characterized in that, Includes the following steps: S1. Construct a sealed test environment containing the interface between high-performance concrete and bedrock while maintaining in-situ stress. S2. Apply progressively increasing fluid pressure steps to the sealing test environment; S3. Monitor the transient seepage signal through the interface under each fluid pressure step; S4. Analyze the decay rate of the transient seepage signal required to reach stability under adjacent pressure steps, and the nonlinear growth law of the initial signal amplitude. S5. Identify the matching degree between the waveform shape of the transient seepage signal and a variety of preset typical interface seepage modes, and quantify the deviation of its shape from the preset reference seepage mode. S6. Based on the decay rate, nonlinear growth law, and morphological deviation, the equivalent penetration rate of the interface is generated.

2. The method for testing the permeability of soil and rock mass according to claim 1, characterized in that, Constructing a sealed test environment containing a high-performance concrete-bedrock interface while maintaining in-situ stress conditions includes: A three-dimensional stress field is applied to the interface between high-performance concrete and bedrock using a confining pressure loading device to simulate the in-situ stress state. An elasto-plastic sealing layer is installed around the interface and interfacial sealing pressure is applied; The pressure inside the sealed test environment is kept in dynamic equilibrium with the three-dimensional stress field applied by the confining pressure loading device through the pressure balancing pipeline.

3. The method for testing the permeability of rock and soil according to claim 2, characterized in that, The elasto-plastic sealing layer is made of a composite material that matches the deformation characteristics of high-performance concrete and bedrock, and the interface sealing pressure is directionally adjusted according to the principal stress direction of the three-dimensional stress field.

4. The method for testing the permeability of soil and rock mass according to claim 1, characterized in that, Applying progressively increasing fluid pressure steps to the sealing test environment, including: Initial fluid pressure is applied to the sealing test environment using a pressure control device; Maintain the initial fluid pressure for a preset duration; The fluid pressure is gradually increased in stages according to a preset pressure increment to form a multi-stage progressive fluid pressure step. During each fluid pressure step transition, the rate of change of fluid pressure is kept constant.

5. The method for testing the permeability of soil and rock mass according to claim 1, characterized in that, Monitor the transient seepage signal across the interface under each fluid pressure step, including: A differential pressure sensor array is symmetrically arranged on both sides of the interface to capture local pressure gradient changes in the interface region; Set the sampling frequency of the differential pressure sensor array to be one order of magnitude higher than the frequency of fluid pressure step change; The voltage signal output by the differential pressure sensor array is recorded synchronously during the duration of each fluid pressure step. The voltage signal is converted into a fluid pressure data sequence as a transient seepage signal through a calibration curve.

6. The method for testing the permeability of soil and rock mass according to claim 1, characterized in that, The analysis included the decay rate of the transient seepage signal requiring time to stabilize under adjacent pressure steps, and the nonlinear growth law of the initial signal amplitude, including: For each fluid pressure step, the transient seepage signal corresponding to the signal is identified by the moving window variance method to determine the time point when the signal reaches stability. The ratio of the time required for the signal to reach stability under adjacent pressure steps is calculated as the attenuation rate. Extract the initial amplitude of the transient seepage signal at the start of each fluid pressure step; Establish a nonlinear functional relationship between the initial amplitude and the step change of fluid pressure; The characteristic parameters in the nonlinear function relationship are fitted using the least squares method to obtain the nonlinear growth law.

7. The method for testing the permeability of soil and rock mass according to claim 6, characterized in that, Establishing a nonlinear functional relationship between the initial amplitude and the fluid pressure step change includes: constructing a data sequence with the fluid pressure step value as the independent variable and the initial signal amplitude as the dependent variable; fitting the data sequence into an exponential or power-law function using the nonlinear least squares method; and determining the characteristic parameters in the function through the fitting process to complete the establishment of the nonlinear functional relationship.

8. The method for testing the permeability of rock and soil according to claim 1, characterized in that, The waveform morphology of the transient seepage signal is matched with several preset typical interface seepage modes to identify the degree of matching, and the deviation of its morphology from the preset reference seepage mode is quantified, including: Extract the waveform morphology feature vector of the transient seepage signal; Calculate the similarity between the waveform morphology feature vector and the feature vectors corresponding to each preset typical interface seepage mode; Based on the similarity results, the optimal matching mode between the transient seepage signal and a variety of preset typical interface seepage modes is determined. The Euclidean distance between the waveform morphology feature vector and the feature vector corresponding to the preset benchmark seepage pattern is calculated as the morphological deviation.

9. The method for testing the permeability of rock and soil according to claim 1, characterized in that, Based on the decay rate, nonlinear growth law, and morphological deviation, the equivalent permeability of the interface is generated, including: The decay rate, characteristic parameters of the nonlinear growth law, and morphological deviation are normalized to form the input parameter set. Substitute the input parameter set into the penetration rate calculation formula obtained through sample training; The equivalent permeability value of the interface is output by solving the permeability calculation formula.

10. A method for testing the permeability of rock and soil according to claim 9, characterized in that, The process of substituting the input parameter set into the penetration rate calculation formula obtained through sample training includes: obtaining a sample dataset containing attenuation rate, nonlinear growth law characteristic parameters, morphological deviation degree and true penetration rate through standard sample testing; using multiple regression analysis to train the sample dataset to establish the penetration rate calculation formula; and substituting the normalized input parameter set into the penetration rate calculation formula to calculate the equivalent penetration rate value.

Images