A fault physical characteristic parameter extraction method based on a double direct shear experiment platform

Abstract

The application discloses a fault physical characteristic parameter extraction method based on a double direct shear experiment platform, and relates to the field of fault physical analysis.The method comprises the following steps: building a double direct shear experiment platform; performing a fault sliding experiment on a to-be-tested fault sample on the double direct shear experiment platform to obtain an experimental data sequence of the to-be-tested fault sample, i.e., an axial strain sequence, a shear stress sequence, a lateral strain sequence and an acoustic wave signal; screening the shear stress sequence based on a Hanning window smoothing and Fourier transform method to obtain a plurality of peak time points; and extracting the experimental data sequence corresponding to the plurality of peak time points in a preset mutation interval to obtain fault physical characteristic parameters, i.e., a shear stress drop, an average shear stress intensity, a sliding rate, a thickness change amount, a transmission coefficient, a longitudinal wave velocity and a tail wave autocorrelation coefficient. The application extracts fault physical characteristic parameters through the double direct shear experiment platform, and improves the reliability of constraints on earthquake nucleation and dynamic rupture prediction.

Description

Technical Field

[0001] This application relates to the field of fault physics analysis technology, and in particular to a method for extracting fault physics characteristic parameters based on a dual-direct shear experimental platform. Background Technology

[0002] Faults are the core structural units for deformation compensation under brittle conditions in the lithosphere. In deeper strata, they can extend to plastic deformation shear zones, leading to significant differences in deformation mechanisms among different fault segments. These structures are widely distributed in the lithosphere and often occur in clusters, fully demonstrating that the complexity and diversity of subsurface structures far exceed expectations.

[0003] Indoor shear experiments are a core method for studying fault slip mechanisms and have significant academic and applied value. Early indoor shear experiments commonly used the following methods: the first was the sawing method; the second involved fracturing cylindrical rock samples within an experimental setup and then measuring the resulting fracture sliding friction; the third often used a rotating shear apparatus for measuring friction; and the fourth was the triaxial shear test. Among these, the triaxial and rotating shear tests, as key tools for studying the mechanical behavior of soil and rock masses and faults, share a significant limitation: their observation methods primarily rely on macroscopic mechanical parameters outside the sample, failing to provide in-situ, synchronous multiphysics field observations within the shear localization zone. Therefore, it is difficult to comprehensively obtain information on key physical characteristics such as displacement, dynamic changes in pore pressure, evolution of local strain fields, and acoustic waves accompanying fault slip, making the experimental process, to some extent, a "black box." This leads to incomplete understanding of fault slip energy conversion and weakening mechanisms, as well as inaccurate subsequent constitutive model construction, thus limiting the reliability of earthquake nucleation and dynamic rupture prediction. Summary of the Invention

[0004] The purpose of this application is to provide a method for extracting fault physical characteristic parameters based on a dual-direct shear test platform. This method can extract fault physical characteristic parameters through the dual-direct shear test platform, thereby improving the reliability of predicting earthquake nucleation and dynamic rupture.

[0005] To achieve the above objectives, this application provides the following solution.

[0006] This application provides a method for extracting fault physical feature parameters based on a dual-direct shear experimental platform, including: A dual-axis shear test platform was constructed; the dual-axis shear test platform is a platform equipped with acoustic wave sensors, displacement sensors, eddy current sensors and mechanical sensors on a biaxial loader; On the dual-direct shear test platform, a fault slip experiment is performed on the fault sample to be tested to obtain the experimental data sequence of the fault sample; the experimental data sequence includes: axial strain sequence, shear stress sequence, lateral strain sequence and acoustic wave signal; the acoustic wave signal includes: acoustic wave signal under stress-free state and acoustic wave signal under applied stress state; Based on the Hanning window smoothing and Fourier transform method, the shear stress sequence is filtered to obtain multiple peak times; Within a preset mutation interval, experimental data sequences corresponding to multiple peak times are extracted to obtain fault physical characteristic parameters; the fault physical characteristic parameters include: shear stress drop, average shear stress intensity, slip rate, thickness change, transmission coefficient, longitudinal wave velocity, and wake autocorrelation coefficient.

[0007] According to the specific embodiments provided in this application, this application has the following technical effects: This application first constructs a dual-axis shear test platform, which is a platform equipped with acoustic sensors, displacement sensors, eddy current sensors, and mechanical sensors on a biaxial loader. A novel shear test platform is employed, introducing acoustic sensors, displacement sensors, eddy current sensors, and mechanical sensors to detect experimental data during fault slip experiments. Axial strain sequences, shear stress sequences, lateral strain sequences, and acoustic signals (acoustic signals under stress-free and stress-applied conditions) are measured. Then, the shear stress sequence is filtered using Hanning window smoothing and Fourier transform methods. Multiple peak times were identified, and within a preset abrupt change interval, the experimental data sequences corresponding to these peak times were extracted to obtain fault physical characteristic parameters (shear stress drop, average shear stress intensity, slip rate, thickness change, transmission coefficient, P-wave velocity, and coma autocorrelation coefficient). This enabled the extraction of fault physical characteristic parameters during the fault slip experiment, making the fault slip experiment process transparent. By extracting fault physical characteristic parameters through a dual-direct shear test platform, the relationship between fault slip energy conversion and weakening mechanism can be mapped, improving the accuracy of subsequent constitutive model construction and thus enhancing the reliability of earthquake nucleation and dynamic rupture prediction. Attached Figure Description

[0008] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0009] Figure 1 This is a schematic diagram of the rotational shearing experiment provided in this application.

[0010] Figure 2 This is a schematic diagram of the triaxial shearing experiment provided in this application.

[0011] Figure 3 This is a flowchart illustrating a method for extracting fault physical characteristic parameters based on a dual-direct shear experimental platform, as provided in this application.

[0012] Figure 4 The diagram shows a top view and a north-south cross-sectional view of the dual straight shear test platform provided in this application.

[0013] Figure 5 This is a schematic diagram of the east-west cross-section of the dual straight shear experimental platform provided in this application.

[0014] Figure 6 A schematic diagram of the double-shear test method provided in this application before the start of the experiment.

[0015] Figure 7 This is a schematic diagram of the double-shear test method provided in this application after the experiment is completed.

[0016] Figure 8 A waveform diagram of the shear stress sequence provided in this application, wherein, Figure 8 (a) is a schematic diagram of the stress decrease rate curve. Figure 8 (b) is a schematic diagram of the shear stress curve.

[0017] Figure 9 This is a waveform diagram of the shear stress drop rate sequence provided in this application, wherein, Figure 9 (a) is a schematic diagram of the shear stress curve. Figure 9 (b) is a schematic diagram of the stress decrease rate curve.

[0018] Figure 10 This is a schematic diagram of the waveform of detected shear stress abrupt change provided in this application, wherein, Figure 10 (a) is a schematic diagram of normal stress. Figure 10 (b) is a schematic diagram of shear stress. Figure 10 (c) is a schematic diagram of normalized T. Figure 10 (d) is a schematic diagram of longitudinal waves. Figure 10 (e) is a schematic diagram of the tailwave autocorrelation coefficient C. Figure 10 (f) is a schematic diagram of the instability event when shear stress, normalized T, longitudinal wave and tail wave autocorrelation coefficient C occur.

[0019] Figure 11 A waveform diagram illustrating the extraction of acoustic emission characteristic information of the acoustic signal provided in this application.

[0020] Figure 12A waveform diagram illustrating typical extraction results of acoustic emission feature information of the acoustic signal provided in this application, wherein, Figure 12 (a) is a schematic diagram of the waveforms of the received wave and the incident wave. Figure 12 (b) is a waveform diagram of the amplitude. Figure 12 (c) is a schematic diagram showing the changes in ultrasonic characteristic parameters under expansion and compaction conditions of the fault sample to be tested. Detailed Implementation

[0021] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0022] The characteristics of faults cannot be fully described by simple geometric models. Simple geometric models are only suitable for macroscopic analysis at the first-level, while faults (especially those with displacements exceeding tens of meters) often evolve into complex fault zones with highly heterogeneous geometric morphology, material composition, and internal structure. This complexity not only makes faults play a key regulatory role in the migration of underground fluids (oil, gas, groundwater, etc.), but also significantly affects the rupture behavior of active faults. Furthermore, as a direct cause of earthquakes, faults naturally become a core link connecting structural geology and seismology. Because faults play an irreplaceable role in revealing underground structures, regulating fluid migration, and analyzing the causes of earthquakes, multiple branches of geological science, including structural geology, geomechanics, seismology, engineering geology, petroleum geology, and Quaternary geology, need to conduct systematic and in-depth research on them.

[0023] Indoor shear experiments, through controlled simulation of underground temperature, pressure, and lithology, can accurately obtain key parameters such as frictional intensity, stick-slip characteristics, and dilatation / contraction effects during fault slippage, revealing the influence of different factors (such as water content, particle size distribution, and confining pressure) on slippage stability. The experimental results provide fundamental data for fault mechanics model construction, supporting theoretical research on deformation mechanisms and earthquake genesis in structural geology. They also provide scientific basis for oil and gas exploration and development (such as reservoir stimulation and wellbore stabilization), engineering site selection, and earthquake hazard risk assessment, effectively compensating for the limitations of quantitative analysis in field observations. This serves as a crucial bridge connecting theoretical research and practical applications.

[0024] The first method, sawing, creates a simulated fault structure by cutting a cylindrical specimen made of rock, steel, or ceramic along its vertical axis at a 30 or 45-degree angle. This fault surface can be roughened or filled with fine-grained abrasive (i.e., fault grooves). The second method uses two hydraulically driven pistons to provide normal and shear stresses, independently controlled. Typically, the normal stress generated by the horizontal axis is kept constant via a servo feedback system, while the vertical axis is driven at a predetermined displacement rate and shear force is measured synchronously. This biaxial, dual-direct-shear configuration allows for simultaneous bare rock and cuttings testing. Fluid testing can be conducted by immersing the specimen assembly in fluid or within a pressure vessel. The latter requires a custom-made rubber sheath to isolate the entire specimen assembly from the confining fluid (oil). The confining pressure acts on the normal stress of the cuttings layer and must therefore be considered. Fluid enters the cuttings layer through holes in three pressure blocks connected to the pressure vessel port, providing fluid coverage to the cuttings sample. To achieve full-area fluid distribution on the cuttings sample, a porous plate is inserted into the specimen surface. In this way, the fluid pressure on both sides of the rock debris layer can be precisely controlled, thereby enabling the measurement of the vertical permeability of the fault. For example... Figure 1 As shown, the third method uses a rotary shearing device that employs servo control to drive a punch to press cylindrical or annular samples together. The drive can be hydraulic or electric, making it suitable for achieving continuous large-displacement shearing, particularly helpful in studying the frictional constitutive relations and seismic physical mechanisms during high-speed fault sliding. For exposed rock samples, aluminum support rings are typically used to provide additional support. Figure 2 As shown, the fourth method, triaxial shear testing, can accurately simulate the static behavior of deep geological environments and obtain material strength parameters, but it cannot conduct in-situ, synchronous multiphysics field observations within the shear localization zone.

[0025] Compared to triaxial shear tests and rotational shear tests, the dual direct shear test platform of this application, by combining multi-sensor synchronous monitoring and in-situ observation of internal physical fields, can effectively overcome the above-mentioned defects. It fills the gap in previous methods that were limited to macroscopic mechanical parameters (such as load and displacement) outside the sample and could not conduct in-situ, synchronous multi-physics field observations inside the shear localization zone. This allows for the revelation of the mechanical and kinematic characteristics of the entire process of fault instability nucleation and dynamic sliding induced by pore pressure disturbance. It enables multi-dimensional, high-resolution data capture of the entire fault sliding process, providing more comprehensive experimental evidence for systematically revealing the fault instability mechanism, thereby improving the reliability of earthquake nucleation and dynamic rupture prediction.

[0026] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, this application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0027] In one exemplary embodiment, such as Figure 3As shown, a method for extracting fault physical feature parameters based on a dual-direct shear experimental platform is provided. The method includes the following steps S1 to S4.

[0028] Step S1: Construct a dual-direction shear test platform; the dual-direction shear test platform is a platform equipped with acoustic wave sensors, displacement sensors, eddy current sensors and mechanical sensors on a biaxial loader.

[0029] As one feasible approach, the biaxial loader comprises three granite blocks arranged in sequence.

[0030] Two fault samples are to be tested, with one sample placed between adjacent granite blocks. An acoustic sensor is placed on the side of any granite block. A displacement sensor is placed on the bottom of the middle granite block. An eddy current sensor is placed on the side of the last granite block. Two mechanical sensors are placed on the surfaces of the two fault samples.

[0031] The twin-axis loader is a horizontal twin-axis loader (see...) Figure 4 (Central top view). Two layers of simulated fault gouge (i.e., the fault sample to be tested) on the dual direct shear test platform are sandwiched between three granite blocks (see...). Figure 4 (North-South cross-section). Roughen each of the four sliding surfaces using sandpaper.

[0032] Two types of displacement monitoring elements were deployed on the top surface of the fault sample to be tested: a high-resolution eddy current sensor (i.e., an eddy current sensor) and a linear variable differential transformer (i.e., a displacement sensor). The linear variable differential transformer (LVDT) is a commonly used linear displacement sensor that converts mechanical displacement into an electrical signal output through electromagnetic induction, offering advantages such as high accuracy, high reliability, and long lifespan. The high-resolution eddy current sensor is installed on the inner side of the right-hand granite block (closer to the fault gouge layer) to accurately measure the thickness variation of the fault gouge layer; the LVDT is installed at the bottom of the middle granite block to monitor displacement along the fault (see...). Figure 4 North-South Cross Section and Figure 5Two additional linear variable differential transformers were installed between the loading plate and the machine frame, respectively positioned at the connection points of the loading plate and the machine frame in the east-west and north-south directions. These transformers were used to measure the displacement of the two servo-controlled loading cylinders, effectively distinguishing between "machine self-deformation" and "the true mechanical response of the fault sample under test." The data acquisition frequency for stress, fault thickness, and slip displacement was 1 kHz to ensure the capture of details of the dynamic process. To monitor the contact state of grains within the fault mud layer, an active ultrasonic monitoring system was installed on the sides of the fault sample under test: on each of the two opposite north-south sample sides, a P-wave piezoelectric ceramic ultrasonic transducer (i.e., acoustic sensor) coated with ultrasonic coupling agent was attached. One transducer served as an active ultrasonic source emitting P-wave signals, while the other served as a receiving probe for transmitting acoustic signals. Simultaneously, an S-wave transducer was installed on the top surface of the fault sample under test, near the active ultrasonic source, to calibrate the excitation time of each ultrasonic pulse and ensure the time synchronization of the acoustic signals.

[0033] like Figure 4 and Figure 5 As shown, the dual-direct shear experimental platform also includes: a loading system, a sample chamber, a multi-parameter sensing system, and auxiliary units. The location and function of each part are as follows: 1. Loading System: Located on a biaxial loader, it consists of horizontal (left and right sides of the fault sample to be tested) and vertical (upper and lower sides of the fault sample to be tested) loading units, which respectively provide shear stress. With normal stress Simulate the formation stress environment to drive fault slip.

[0034] 2. Sample Chamber: Used to place the fault sample to be tested. Located in the center of the device, it consists of the surrounding rock mass and the fault zone. It is the core area of ​​the experiment and is used to contain the "surrounding rock-fault" sample and provide shear surface constraints.

[0035] 3. Multi-parameter sensing system: including acoustic wave sensors, displacement sensors, eddy current sensors and mechanical sensors.

[0036] Among them, the acoustic sensor is an acoustic emission probe (attached to the surface / inside of the fracture sample to be tested) + function generator: actively excites / passively receives micro-fracture signals to capture acoustic information.

[0037] The displacement sensor is a displacement sensor (next to the fault zone), and the eddy current sensor is (on both sides of the shear plane): respectively measuring the sliding displacement and the evolution of the fault zone thickness.

[0038] Mechanical sensors (loaded surface / inside sample): Real-time acquisition of mechanical data such as shear stress and normal stress.

[0039] 4. Auxiliary Unit: The data acquisition system connects each sensor and the loading unit via cables to synchronously integrate data from multiple fields; the rigid boundary (such as a granite block) is located outside the sample chamber, providing a boundary constraint / sealing environment.

[0040] Step S2: On the dual-direct shear test platform, a fault slip test is performed on the fault sample to be tested to obtain the experimental data sequence of the fault sample to be tested; the experimental data sequence includes: axial strain sequence, shear stress sequence, lateral strain sequence, and acoustic signal; the acoustic signal includes: acoustic signal under stress-free state and acoustic signal under applied stress state.

[0041] Specifically, such as Figure 6 and Figure 7 As shown, the fault slip experiment first uses two axially hydraulically driven pistons on a double-direct shear test platform to provide independently controlled normal and shear stresses to the fault sample. During the double-shear experiment, the fault sample to be tested is first installed in the double-direct shear test platform. Then, a set normal stress is applied in the normal direction of the fault sample to simulate the surrounding rock pressure and kept constant. Next, shear stress is gradually applied along the shear direction until the fault sample fails along the fault plane. During the experiment, the changes in shear stress and displacement are recorded in real time, and the failure phenomenon is observed.

[0042] A double-shear experiment (i.e., fault slip experiment) is conducted. The double-shear experiment extracts two parts of information about the physical characteristics of the fault sample under test through experimental data sequences. The first part is mechanical and kinematic information, and the second part is acoustic information. Among them, the mechanical and kinematic information includes the evolution of shear strength, slip rate, and fault thickness (i.e., the change in thickness).

[0043] Specifically, by extracting and combining the sliding displacement data of the LVDT on the top surface of the sample with the cylinder displacement data of the LVDT between the loading plate and the frame, the interference of machine deformation can be eliminated, and the true sliding amount of the fault can be obtained, thereby deriving mechanical parameters such as shear strain. The "fault mud layer thickness change" collected by the eddy current sensor can reflect the compression / expansion behavior of the fault during loading. The stress data, linked with the above displacement and thickness data, can construct the "stress-strain-displacement" response relationship of the fault and analyze its tribological characteristics. For shear strength, if a series of sudden sliding events of the fault are encountered, the shear stress drop and the change in average shear strength for each sliding time are extracted. For acoustic information, the transmission coefficient, longitudinal wave velocity, and wake wave autocorrelation coefficient are extracted. Acoustic information extraction, by analyzing the propagation time, amplitude, and waveform evolution of the P-wave received signal, can invert the microscopic state of grains within the fault mud layer, such as contact stiffness and porosity. The "pulse excitation time data" of the S-wave transducer is used to calibrate the emission time of the P-wave, improving the accuracy of microscopic state inversion. Furthermore, by correlating the "acoustic signal characteristics" with the "mechanical / kinematic data," the coupling mechanism between fault friction intensity, sliding behavior, and grain contact state can be revealed.

[0044] Among them, the displacement sensor is used to acquire the axial strain sequence, the mechanical sensor is used to acquire the shear stress sequence, the eddy current sensor is used to acquire the lateral strain sequence, and the acoustic wave sensor is used to acquire the acoustic wave signal.

[0045] Step S3: Based on the Hanning window smoothing and Fourier transform method, the shear stress sequence is screened to obtain multiple peak times.

[0046] As one feasible approach, step S3 specifically includes steps S31 to S34: Step S31: Based on the shear stress sequence, the shear stress drop rate sequence is calculated using the Hanning window smoothing and Fourier transform methods.

[0047] Specifically, based on the shear stress sequence, the evolution curve of the target signal (with the shear stress sequence as the core) over time is first constructed (see...). Figure 8 Simultaneously, the "Hanning window smoothing + Fourier transform differentiation" method is used to calculate the time derivative of the signal (i.e., the shear stress drop rate sequence) (see...). Figure 9 ).

[0048] As an implementable approach, the expression for Hanning window smoothing is: .

[0049] in, For the first Hanning window weights for each signal data point; This is the Hanning window filtering function; For the length of the window, It is an even number; This represents the number of signal data points.

[0050] The expression for the Fourier transform method is: ; in, This is a sequence of shear stress drop rates; For Fast Fourier Transform; It is a shear stress sequence.

[0051] The time differential curve of shear stress exhibits a significant negative sharp peak at the moment of shear force abrupt change, and this peak segment corresponds to... Figure 8 The sudden drop and rise in the rate of shear stress reduction curve (such as the dark-marked segment marked -0.14167MPa / s) is the core characteristic of the sudden change in shear stress.

[0052] Step S32: Perform a positive transformation on the negative peak value in the shear stress drop rate sequence to obtain the transformed shear stress drop rate sequence (positive value curve).

[0053] Specifically, the negative peak is converted to a positive peak (for) Figure 8 The shear stress decrease rate curve is transformed, with the negative peak on the curve representing the instant of rapid stress decrease. .

[0054] in, This is the transformed shear stress drop rate sequence; This is a sequence of shear stress drop rates.

[0055] Step S33: Based on the preset minimum peak height and preset minimum peak spacing, the transformed shear stress drop rate sequence is filtered using a peak-finding function to obtain multiple initial peak times.

[0056] Specifically, preset value filtering: After obtaining the transformed shear stress drop rate sequence (positive curve), a peak detection algorithm is used for filtering to automatically detect significant instability events. A preset minimum peak height (0.065 MPa / s) + preset minimum peak spacing (500 data points) is set. Peaks are filtered using a peak-finding function, outputting multiple initial peak times that satisfy the above preset values ​​(corresponding to the event occurrence time, such as...). Figure 9 The marked moments of shear stress abrupt change and multiple initial peak heights (i.e., stress drop rate amplitudes, such as...) Figure 8 The 0.14167 MPa / s marked in the figure generates a catalog of instability events.

[0057] Step S34: Reconstruct multiple initial peak times to obtain multiple peak times.

[0058] Specifically, the necessary post-processing of the preset value screening results maps the candidate events detected on the positive transformation curve back to the true negative peaks in the original differential signal, thereby obtaining precise multiple peak moments (i.e., the instants when stress decreases most drastically). This step ensures the physical authenticity of the event moments and provides a reliable time reference for all subsequent window analyses based on those moments.

[0059] Step S4: Within the preset mutation interval, extract the experimental data sequences corresponding to multiple peak times to obtain fault physical characteristic parameters; the fault physical characteristic parameters include: shear stress drop, average shear stress intensity, slip rate, thickness change, transmission coefficient, longitudinal wave velocity and wake autocorrelation coefficient.

[0060] After determining a sudden change moment in the shear stress (i.e., the peak moment), the characteristic points of the response signal are located within the interval before and after the sudden change (i.e., from the end of the calm period to the outbreak of the event to the return to calm) using the peak moment as the anchor point, and the core parameters are extracted by differentiation.

[0061] As one feasible approach, step S4 specifically includes steps S41 to S48: Step S41: Within the preset abrupt change interval, extract the stress sequences corresponding to multiple peak times to obtain the peak shear strength and residual shear strength. The preset abrupt change interval includes: the interval before the abrupt change and the interval after the abrupt change.

[0062] The maximum value of the average effective stress is extracted in the interval before the abrupt change (the first peak value along the time axis from the point of maximum slope, corresponding to...). Figure 8 The green-filled diamond-shaped marker represents the peak shear strength.

[0063] The minimum value of the average effective stress is extracted in the interval after the abrupt change (the first valley value along the time axis from the point of maximum slope, corresponding to...). Figure 8 The purple-filled diamond-shaped marker represents the residual shear strength.

[0064] Step S42: Calculate the shear stress drop and average shear stress intensity based on the peak shear strength and residual shear strength.

[0065] Specifically, Figure 10 This paper presents the response of transmitted ultrasound to high-frequency normal stress oscillations (NSO) under velocity weakened wave (SVW) fault gouge conditions. The figure compares the evolution characteristics of normalized T, P-wave V, and wake correlation coefficient C with shear stress after the NSO is initiated (i.e., normal stress is applied). The data in this set are from experiments with a frequency of 1 Hz and an amplitude of 20%. To eliminate the interference of elastic effects, Figure 10In (c) and (e) of 10, the corrected normalized T, the wake autocorrelation coefficient C, and the original measured data are plotted simultaneously. At the instant the normal stress oscillation begins, the normalized T, P-wave V, and wake autocorrelation coefficient all show an instantaneous decrease, and the shear stress decreases synchronously; subsequently, the normalized T continues to decrease, the wake autocorrelation coefficient C continues to increase, while the P-wave does not show obvious stage evolution. The background values ​​of the three parameters all fluctuate synchronously in a sinusoidal manner with the applied normal stress oscillation. As the average shear stress gradually increases and tends to stabilize, the average values ​​of the normalized T, P-wave V, and wake autocorrelation coefficient C are all lower than the levels before oscillation. This phenomenon indicates that after the oscillation stage, the average particle contact area of ​​the fault decreases, or the overall stiffness decreases. In other words, the fault expansion effect appears instantaneously at the start of the normal stress oscillation and continues to develop in the subsequent stages.

[0066] Specifically, if a sudden change in shear stress is detected based on the experimental data sequence (i.e., an instability event is defined as a situation where multiple indicators show abnormalities, such as shear stress approaching its peak value, abnormal wave velocity, and a sharp increase in acoustic emission activity), then it is considered an instability event. Figure 10 As shown in (f), further calculations are performed: Shear stress drop: The difference between the peak shear strength (i.e., the maximum value of the average effective stress extracted in the pre-abrupt interval) and the residual shear strength (i.e., the minimum value of the average effective stress extracted in the post-abrupt interval) is used to quantify the stress drop magnitude of the instability event.

[0067] Average shear stress intensity: The arithmetic mean of the peak shear strength (i.e., the maximum value of the average effective stress extracted in the interval before the sudden change) and the residual shear strength (i.e., the minimum value of the average effective stress extracted in the interval after the sudden change) is taken as the comprehensive characterization of the average bearing capacity during the instability process.

[0068] Step S43: Within the preset abrupt change range, extract the axial strain sequence corresponding to multiple peak times to obtain the sliding rate.

[0069] As an feasible approach, the formula for calculating the sliding rate is: .

[0070] in, For the first The sliding rate of each signal data point; For the first Axial strain of each signal data point in the interval after the abrupt change; For the first Axial strain of each signal data point in the interval before the sudden change; For the first The time of each signal data point in the interval after the sudden change; For the first The time interval of each signal data point before the mutation.

[0071] Step S44: Within the preset abrupt change range, extract the lateral strain sequence corresponding to multiple peak times to obtain the thickness change.

[0072] As an feasible approach, the formula for calculating the thickness variation is: ; in, For the first The thickness change of each signal data point; The initial thickness of the fault sample to be tested; For the first Lateral strain of each signal data point in the interval after the abrupt change; For the first Lateral strain of each signal data point in the interval before the mutation.

[0073] Step S45: Based on the acoustic signal, determine the peak-to-peak value of the acoustic wave under no stress and the peak-to-peak value of the acoustic wave under applied stress.

[0074] As an implementable method, step S46 specifically includes: determining the peak-to-peak value within the vibration period corresponding to the first maximum peak value in the acoustic signal under stress-free state as the peak-to-peak value of the acoustic signal under stress-free state; and determining the peak-to-peak value within the vibration period corresponding to the first maximum peak value in the acoustic signal under applied stress as the peak-to-peak value of the acoustic signal under applied stress.

[0075] Specifically, only the acoustic wave emission / reception system (active excitation and passive reception acoustic emission probes) in the dual direct shear test platform is activated, and no stress is applied to the fracture sample under test (both normal stress and shear stress are 0), so the fracture sample under test is in an initial relaxed state; the peak-to-peak value of the first maximum peak value of the received waveform at this time is collected within the vibration period. Figure 11 The first peak with the highest amplitude following the blue line in the black curve is denoted as... .

[0076] Under applied stress, normal stress is applied to the surrounding rock mass on both sides of the sample along the "normal stress loading direction" (reference). Figure 6 Shear stress is applied to the fracture surface region of the sample along the "shear stress loading direction". After the stress is loaded to the experimental set value and the sample is in a stable shear state, the peak-to-peak value of the first maximum peak value of the received waveform within the vibration period is collected. Figure 11 The first peak with the highest amplitude following the blue line in the red curve is denoted as... .

[0077] Step S46: Calculate the transmission coefficient based on the peak-to-peak value of the sound wave under no stress and the peak-to-peak value of the sound wave under applied stress.

[0078] As a feasible approach, the transmission coefficient needs to be calculated based on acoustic signals under both "stress-free" and "shear stress" conditions. The formula for calculating the transmission coefficient is: .

[0079] in, Transmission coefficient; The peak-to-peak value of the sound wave under applied stress; This represents the peak-to-peak value of the sound wave under stress-free conditions.

[0080] Step S47: Calculate the longitudinal wave velocity based on the acoustic wave signal under a preset fixed width and stress-free state and the acoustic wave signal under applied stress state.

[0081] Specifically, such as Figure 12 As shown, in a preset fixed width (Effective width between cross-sections calibrated before the experiment) By identifying the acoustic signal under no stress and the acoustic signal under applied stress (taking the moment when the rising edge of the waveform reaches the amplitude threshold), the arrival time difference between the two is calculated. The formula for calculating the longitudinal wave velocity is: .

[0082] in, For longitudinal wave velocity; Preset fixed width; The time interval is the time between the acoustic signal under no stress and the acoustic signal under applied stress.

[0083] Step S48: Based on the acoustic signal under no stress and the acoustic signal under applied stress, calculate the tail wave autocorrelation coefficient using the Pearson correlation coefficient formula.

[0084] Specifically, such as Figure 8 The red curve in the middle represents the tail wave segment of the received waveform after the P-wave (longitudinal wave) and S-wave (transverse wave) have completely decayed and their amplitudes have stabilized, serving as the acoustic signal under stress-free conditions.

[0085] As an feasible approach, the formula for calculating the wake autocorrelation coefficient is: .

[0086] in, This is the tail wave autocorrelation coefficient; For the first The acoustic signal of each signal data point under stress-free conditions; The number of signal data points; This represents the average value of the acoustic signal under stress-free conditions. For the first The acoustic signal of each signal data point under applied stress; This is the average value under applied stress.

[0087] A comprehensive analysis of the physical properties of the fault, combining mechanical / kinematic information with acoustic wave information, reveals the following evolutionary pattern during the instability process: 1. Initial state before instability: high integrity, high stiffness, and uniform density.

[0088] In the initial loading stage, the fault is in a high stress accumulation phase. Acoustic data shows that the transmission coefficient in the stress-free state... Longitudinal wave velocity Stable above 3400m / s, and the tail wave autocorrelation coefficient It remained above 0.9. This indicates that the fault initially had good integrity, high stiffness, and a uniform and dense internal structure, providing a material basis for subsequent sudden instability.

[0089] 2. Instability triggering and sliding behavior: Strong instability is accompanied by expanded sliding. As the shear stress loading reached the threshold, the fault experienced dynamic instability. Based on the mechanical parameters, the shear stress drop in this event was determined to be... MPa, which is considered a strong instability event. Within the 0.5s time window of instability, the slip velocity increased more than 10 times compared to the steady-state value, while the thickness of the fault gouge layer also changed significantly. Initial thickness of sample The 0.1% indicates that the internal fractures of the fault expanded during the instability process, resulting in a significant dilatation effect, which belongs to the expansion type of slip.

[0090] 3. Post-instability structural evolution: fragmentation damage, stiffness reduction and non-uniform fragmentation The strong instability event led to drastic deterioration of the fault's internal structure. Acoustic parameters show that the P-wave velocity increased significantly after the instability. The speed dropped below 3300 m / s, confirming a significant decrease in fault stiffness; transmission coefficient A sharp drop to below 0.4 indicates fragmentation damage to the internal structure of the fault, resulting in a significant decrease in acoustic wave propagation capability; the coma autocorrelation coefficient... The decrease from above 0.9 to below 0.3 quantitatively proves that the internal structure of the fault has changed from uniform and dense to non-uniform and fragmented.

[0091] 4. Overall Conclusion In this double-shear experiment, the fault, initially characterized by high integrity, high stiffness, and homogeneous density, underwent a strong instability (expansion-type slip) event, ultimately evolving into a physical state of "fragmentation + low stiffness + non-homogeneity." This result successfully establishes a direct correspondence between macroscopic mechanical behavior (strong instability, expansion-type slip) and internal structural evolution (stiffness decay, fragmentation, non-homogenization), which can be used to explain the mechanical mechanisms of actual fault instability and the response patterns of geophysical fields (such as wave velocity changes and seismic wave attenuation).

[0092] The beneficial effects of the fault physical feature parameter extraction method based on the dual direct shear test platform proposed in this application are mainly reflected in the following aspects: The core of this application lies in providing a dual-direct shear experimental platform equipped with acoustic, displacement, eddy current, and mechanical sensors on a biaxial loader. Furthermore, after conducting fault slip experiments on the test fault sample on this platform, high-resolution feature parameter extraction of mechanical, kinematic, and acoustic information is performed. This achieves joint analysis of fault physical feature parameters, from the macroscopic stress-strain-displacement response relationship to microscopic parameters such as wave velocity, transmission coefficient, and wake correlation. This method, relying on the synergistic arrangement of the biaxial loader and multi-parameter sensors, constructs a dual-direct shear experimental platform capable of in-situ, synchronous multi-physics field observation of shear localization zones, overcoming the limitations of traditional rotating / triaxial shear methods that only measure macroscopic mechanical parameters. By conducting fault slip experiments on a dual-direct shear experimental platform, the "black box" defects of traditional experiments are effectively overcome. This method can comprehensively capture the mechanical, acoustic, kinematic, and thickness evolution information of fault slip, and more completely reveal its physical mechanisms such as instability nucleation and slip weakening. It fully maps the relationship between fault slip energy conversion and weakening mechanisms, providing comprehensive data support and analytical capabilities for systematically revealing the evolution of the internal structure of faults. This method has important value for earthquake risk prediction and reservoir stability assessment, improves the accuracy of subsequent constitutive model construction, and enhances the reliability of earthquake nucleation and dynamic rupture prediction.

[0093] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0094] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for extracting fault physical characteristic parameters based on a dual-direct shear test platform, characterized in that, include: A dual-axis shear test platform was constructed; the dual-axis shear test platform is a platform equipped with acoustic wave sensors, displacement sensors, eddy current sensors and mechanical sensors on a biaxial loader; On the dual-direct shear test platform, a fault slip experiment was performed on the fault sample to be tested to obtain the experimental data sequence of the fault sample to be tested. The experimental data sequence includes: axial strain sequence, shear stress sequence, lateral strain sequence, and acoustic signal; the acoustic signal includes: acoustic signal under no stress and acoustic signal under applied stress. Based on the Hanning window smoothing and Fourier transform method, the shear stress sequence is filtered to obtain multiple peak times; Within a preset mutation interval, experimental data sequences corresponding to multiple peak times are extracted to obtain fault physical characteristic parameters; the fault physical characteristic parameters include: shear stress drop, average shear stress intensity, slip rate, thickness change, transmission coefficient, longitudinal wave velocity, and wake autocorrelation coefficient.

2. The method for extracting fault physical feature parameters based on a dual-direct shear experimental platform according to claim 1, characterized in that, The biaxial loader includes three granite blocks arranged in sequence; The test fault samples are two in total, with one test fault sample placed between adjacent granite blocks; the acoustic sensor is placed on the side of any granite block; the displacement sensor is placed on the bottom of the middle granite block; the eddy current sensor is placed on the side of the last granite block; and the mechanical sensor is two in total, respectively placed on the surface of the two test fault samples.

3. The method for extracting fault physical feature parameters based on a dual-direct shear experimental platform according to claim 1, characterized in that, Based on the Hanning window smoothing and Fourier transform method, the shear stress sequence is filtered to obtain multiple peak times and corresponding peak heights, specifically including: Based on the shear stress sequence, the shear stress drop rate sequence is calculated using the Hanning window smoothing and Fourier transform methods. The negative peak value in the shear stress drop rate sequence is transformed in the positive direction to obtain the transformed shear stress drop rate sequence. Based on the preset minimum peak height and preset minimum peak spacing, a peak-finding function is used to filter the transformed shear stress drop rate sequence to obtain multiple initial peak times; Multiple initial peak times are restored to obtain multiple peak times.

4. The method for extracting fault physical feature parameters based on a dual-direct shear experimental platform according to claim 3, characterized in that, The expression for the Hanning window smoothing is: ； in, For the first Hanning window weights for each signal data point; This is the Hanning window filtering function; For the length of the window, It is an even number; The number of signal data points; The expression for the Fourier transform method is: ； in, This is a sequence of shear stress drop rates; For Fast Fourier Transform; It is a shear stress sequence.

5. The method for extracting fault physical feature parameters based on a dual-direct shear experimental platform according to claim 1, characterized in that, Within a pre-defined mutation interval, experimental data sequences corresponding to multiple peak times are extracted to obtain fault physical characteristic parameters, specifically including: Within the preset abrupt change interval, stress sequences corresponding to multiple peak times are extracted to obtain peak shear strength and residual shear strength; the preset abrupt change interval includes: a pre-abrupt change interval and a post-abrupt change interval; Based on the peak shear strength and the residual shear strength, calculate the shear stress drop and the average shear stress intensity; Within the preset abrupt change range, the axial strain sequence corresponding to multiple peak times is extracted to obtain the sliding rate; Within the preset abrupt change range, the lateral strain sequence corresponding to multiple peak times is extracted to obtain the thickness change. Based on the acoustic signal, the peak-to-peak value of the acoustic wave under no stress and the peak-to-peak value of the acoustic wave under applied stress are determined. The transmission coefficient is calculated based on the peak-to-peak value of the sound wave under stress-free conditions and the peak-to-peak value of the sound wave under applied stress conditions. The longitudinal wave velocity is calculated based on the acoustic wave signal under a preset fixed width and stress-free state and the acoustic wave signal under applied stress state. Based on acoustic signals under stress-free and stress-applied conditions, the Pearson correlation coefficient formula is used to calculate the coma autocorrelation coefficient.

6. The method for extracting fault physical feature parameters based on a dual-direct shear experimental platform according to claim 5, characterized in that, Based on the acoustic signal, the peak-to-peak value of the acoustic wave under no stress and the peak-to-peak value of the acoustic wave under applied stress are determined, specifically including: The peak-to-peak value within the vibration period corresponding to the first maximum peak value in the acoustic signal under stress-free conditions is determined as the peak-to-peak value of the acoustic wave under stress-free conditions. The peak-to-peak value within the vibration period corresponding to the first maximum peak value in the acoustic signal under applied stress is determined as the peak-to-peak value of the acoustic wave under applied stress.

7. The method for extracting fault physical feature parameters based on a dual-direct shear experimental platform according to claim 5, characterized in that, The formula for calculating the sliding rate is: ； in, For the first The sliding rate of each signal data point; For the first Axial strain of each signal data point in the interval after the abrupt change; For the first Axial strain of each signal data point in the interval before the sudden change; For the first The time of each signal data point in the interval after the sudden change; For the first The time interval of each signal data point before the mutation.

8. The method for extracting fault physical feature parameters based on a dual-direct shear experimental platform according to claim 5, characterized in that, The formula for calculating the thickness change is: ； in, For the first The thickness change of each signal data point; The initial thickness of the fault sample to be tested; For the first Lateral strain of each signal data point in the interval after the abrupt change; For the first Lateral strain of each signal data point in the interval before the mutation.

9. The method for extracting fault physical feature parameters based on a dual-direct shear experimental platform according to claim 5, characterized in that, The formula for calculating the transmission coefficient is: ； in, Transmission coefficient; The peak-to-peak value of the sound wave under applied stress; This represents the peak-to-peak value of the sound wave under stress-free conditions.

10. The method for extracting fault physical feature parameters based on a dual-direct shear experimental platform according to claim 5, characterized in that, The formula for calculating the tail wave autocorrelation coefficient is as follows: ； in, This is the tail wave autocorrelation coefficient; For the first The acoustic signal of each signal data point under stress-free conditions; The number of signal data points; This represents the average value of the acoustic signal under stress-free conditions. For the first The acoustic signal of each signal data point under applied stress; This is the average value under applied stress.

Images