A method for calibrating geological stratification parameters based on acoustic wave profiles and core tests

Abstract

The present invention relates to the technical field of geological exploration and geotechnical engineering testing, and specifically relates to a method for calibrating geological stratification parameters based on acoustic profiles and core tests, comprising the following steps: S1: Obtain the acoustic profile data of the target area and perform time-frequency domain preprocessing on the acoustic profile; S2: Divide the stratigraphic interfaces to generate an acoustic stratification profile with boundary markers; S3: Implement core sampling and perform three-dimensional CT scanning and mechanical parameter tests on the core samples; S4: Establish an acoustic velocity-mechanical parameter mapping relationship model and correct the transitional values of the interlayer parameters; S5: Construct a three-dimensional stratigraphic parameter matrix based on the corrected transitional parameter values and output a calibration report of geological stratification parameters with credibility indicators. With the present invention, by integrating acoustic profile and core test data, continuous calibration of stratigraphic mechanical parameters and credibility quantification are achieved, effectively improving the accuracy and reliability of geological stratification modeling.

Description

Technical Field

[0001] The present invention relates to the technical field of geological exploration and geotechnical engineering testing, and particularly relates to a method for calibrating geological stratification parameters based on acoustic profiles and core tests. Background Art

[0002] In the fields of geological exploration and engineering survey, the acoustic profile technology is widely used in underground structure detection and stratum identification due to its high resolution, non-destructive nature, and continuous acquisition ability. However, the acoustic profile data essentially reflects the physical response characteristics of the stratum medium and is difficult to directly reflect the specific lithological mechanical properties. To accurately master the key stratum mechanical parameters in engineering design, it is usually necessary to conduct in-situ tests on samples in combination with core tests. However, due to the limited number of core samples and the distribution positions being restricted by costs and construction conditions, it is impossible to fully cover all stratification areas, resulting in the inability to continuously calibrate the stratum mechanical parameters, seriously affecting the integrity and accuracy of geological modeling.

[0003] In the prior art, there is a lack of a systematic method to accurately align and fuse acoustic profile data with core parameters, especially in terms of parameter spatial extension and multi-source data consistency verification. At the same time, traditional parameter interpolation methods lack physical constraints, are prone to introducing non-geologically reasonable transition zones, and fail to establish a clear mapping relationship from acoustic response to rock mechanical characteristics. Therefore, there is an urgent need for a method for calibrating geological stratification parameters based on acoustic profiles and core tests to solve the problems of discontinuous, unreliable, and non-quantifiable geological stratification parameters. Summary of the Invention

[0004] Based on the above purpose, the present invention provides a method for calibrating geological stratification parameters based on acoustic profiles and core tests.

[0005] A method for calibrating geological stratification parameters based on acoustic profiles and core tests includes the following steps:

[0006] S1: Obtain the acoustic profile data of the target area, and perform time-frequency domain preprocessing on the acoustic profile to generate a preprocessed acoustic profile containing time-frequency characteristics;

[0007] S2: Based on the time-frequency characteristic differences of the preprocessed acoustic profile, use the sliding window energy ratio algorithm to divide the stratum interfaces and generate an acoustic stratification profile with boundary markers;

[0008] S3: Conduct core sampling at the corresponding positions of the key layers of the acoustic stratification profile, and perform three-dimensional CT scans and mechanical parameter tests on the core samples to generate a core parameter dataset containing spatial coordinates;

[0009] S4: Establish a mapping relationship model between acoustic wave velocity and mechanical parameters, perform time-frequency feature matching on the core drilling parameter dataset and the acoustic wave stratigraphic profile, and correct the transitional values of interlayer parameters by using the robust estimation algorithm;

[0010] S5: Construct a three-dimensional formation parameter matrix based on the corrected transitional parameter values, verify the parameter continuity by using the wave impedance inversion method, and output a geological stratification parameter calibration report with a credibility index.

[0011] Optionally, the S1 specifically includes:

[0012] S11: Use a towed multi-channel acoustic wave detection device to carry out detection operations along a straight detection profile set in the target area. The detection device is configured with 12 receiving channels and 1 high-energy sound source, the center frequency of the sound source is set at 60 kHz, and the detection depth coverage range is 0 - 50 meters; in an environment where the water depth change is no more than ±2 meters, uniformly advance along the profile direction at a stable towing speed of 0.5 m / s, emit a pulse signal every 0.2 seconds, and synchronously collect the reflected signals of each channel to continuously obtain the raw acoustic wave profile data covering the target area;

[0013] S12: Denoise the obtained raw acoustic wave profile data, use a band-pass filter to remove signal interference with a frequency less than 10 Hz or greater than 500 Hz, and output the denoised pure acoustic wave profile data;

[0014] S13: Perform Fourier transform on the pure acoustic wave profile data to obtain frequency domain distribution characteristics; then apply short-time Fourier transform to the frequency domain characteristics, set the short-time window length to 50 ms and the overlap rate to 50%, extract the spectral characteristics of the profile signal changing with time, and form a time-frequency feature matrix;

[0015] S14: Use Hilbert transform to extract the envelope of the time-frequency feature matrix, and perform normalization processing on the extracted envelope signal to generate a preprocessed acoustic wave profile containing complete time-frequency features.

[0016] Optionally, the S14 specifically includes:

[0017] S141: For each time-domain signal sequence in the time-frequency feature matrix obtained in S13, perform Hilbert transform respectively to obtain the analytical expression of the corresponding signal;

[0018] S142: Calculate the envelope amplitude of each time-domain signal sequence, combine the Hilbert transform result obtained in S141 with the original signal to form an analytical signal;

[0019] S143: Use the extreme value normalization method to normalize the envelope amplitude obtained in S142 to obtain a standardized time-frequency feature data matrix with a value range within [0, 1].

[0020] S144: Combine all the normalized envelope signals in the original channel order to form a preprocessed acoustic wave profile containing complete time-frequency characteristics.

[0021] Optionally, the specific steps of S2 are as follows:

[0022] S21: Based on the preprocessed acoustic wave profile, set a sliding window with a length of 20 ms along the time axis of acoustic wave propagation, and slide it successively with a step size of 2 ms to intercept the signal data within each window.

[0023] S22: Calculate the short-time energy value of the signal data within each sliding window by summing the squares of the amplitudes of each sampling point within the window to obtain the energy value corresponding to the window.

[0024] S23: Calculate the energy ratio between adjacent sliding windows successively.

[0025] S24: Compare the energy ratio obtained in S23 with the preset formation interface determination threshold of 3.0. When the energy ratio exceeds the threshold, mark the window boundary position corresponding to that place as the formation interface, and scan the entire acoustic wave profile successively to obtain the positions of all formation interfaces that meet the conditions.

[0026] S25: Superimpose all the marked formation interface positions onto the preprocessed acoustic wave profile to generate an acoustic wave stratification profile with boundary marks.

[0027] Optionally, the specific steps of S3 are as follows:

[0028] S31: According to the acoustic wave stratification profile obtained in S25, select the layer with the highest energy ratio as the key layer, and record the three-dimensional spatial coordinate information of this key layer.

[0029] S32: Set the drilling position at the three-dimensional spatial coordinate position recorded for the key layer, and use a core drilling rig to drill core samples vertically. The drilling speed is set to 0.2 m / min, the core diameter is 50 mm, and the length of the target formation position to be drilled is 2 m to obtain columnar core samples.

[0030] S33: Conduct industrial-grade three-dimensional CT scanning on the obtained columnar core samples. The scanning parameters are set as: scanning voltage 120 kV, scanning current 180 mA, and spatial scanning resolution 0.1 mm to obtain the three-dimensional structure image data inside the core samples.

[0031] S34: Cut the columnar core samples into standard specimens with a specimen size of diameter 50 mm and height 100 mm, conduct uniaxial compressive strength tests on the standard specimens, load them at a constant loading rate of 0.5 MPa / s until the specimens are damaged, and measure the compressive strength, elastic modulus, and Poisson's ratio parameters of each specimen.

[0032] S35: Integrate the spatial three-dimensional coordinates of the obtained core samples, the structural image data obtained by CT scanning, and the mechanical parameter test data to generate a core parameter dataset with spatial coordinate information.

[0033] Optionally, the specific steps of S4 are as follows:

[0034] S41: Extract the acoustic wave propagation velocity data between each stratum interface, obtain the average acoustic wave propagation velocity within each stratification segment, and form a velocity vector set.

[0035] S42: Extract the core sample parameters corresponding to the spatial position of the stratification segment in S35, obtain the compressive strength, elastic modulus, and Poisson's ratio corresponding to each segment, and form a mechanical parameter vector set.

[0036] S43: Based on the correspondence between the velocity vector set and the mechanical parameter vector set, use the multiple linear regression method to construct a mapping relationship model between the acoustic wave velocity and the mechanical parameters, and output the predicted mechanical parameter values corresponding to each acoustic wave velocity value.

[0037] S44: Apply the established mapping relationship model to all stratification segments in the acoustic wave stratification profile, combine their average acoustic wave velocity, calculate the initial mechanical parameter transition data, and perform spatial registration with the corresponding positions in the core dataset.

[0038] S45: For the deviations in the initial parameter transition data caused by local anomalies or measurement errors, use the robust estimation algorithm to correct the parameters and form the corrected mechanical parameter transition values.

[0039] Optionally, the specific steps of S43 are as follows:

[0040] S431: Use the multiple linear regression method to establish three linear mapping models with the compressive strength, elastic modulus, and Poisson's ratio as the dependent variables and the acoustic wave velocity as the independent variable respectively. The regression equations are expressed as follows:

[0041] ;

[0042] ;

[0043] ; In the formula, represents the acoustic wave propagation velocity of the i-th segment; , , respectively represent the compressive strength, elastic modulus, and Poisson's ratio corresponding to the i-th segment; , , are the slope coefficients of each regression model; , , is the intercept constant term of each regression model;

[0044] S432: Use the least squares method to fit the regression coefficients, minimize the squared error between the predicted values and the actual mechanical parameter values, and output the predicted mechanical parameter triples corresponding to each acoustic velocity value .

[0045] Optionally, the S44 specifically includes:

[0046] S441: Apply the established mapping relationship model between acoustic velocity and mechanical parameters to the average acoustic velocity values of each layer segment in the acoustic stratigraphic section, substitute each acoustic velocity value into the prediction model, and obtain the corresponding initial mechanical parameters, including compressive strength, elastic modulus, and Poisson's ratio;

[0047] S442: Associate the calculated set of initial mechanical parameters with the spatial coordinates in the acoustic stratigraphic section to form a preliminary parameter transition data set;

[0048] S443: Extract the spatial positions of each sample in the core parameter data set, and judge whether it meets the registration condition according to the spatial distance between the center position of the acoustic layer segment and the center position of the core sample; when the spatial distance is within the preset error range, it is considered that the acoustic prediction parameter and the core measured parameter form a valid matching pair;

[0049] S444: Output all registered prediction parameter and measured parameter pairs.

[0050] Optionally, the S45 specifically includes:

[0051] S451: For each set of registered prediction parameters and measured parameters in S444, calculate the difference between the two;

[0052] S452: Judge each deviation value according to the set residual threshold; when the deviation amplitude is within the allowable range, retain the deviation; when it exceeds the range, adjust the deviation to the threshold size and retain its positive or negative sign;

[0053] S453: Subtract the adjusted deviation amount from the original prediction parameter to obtain the corrected mechanical parameter transition value;

[0054] S454: Summarize all the corrected mechanical parameter transition values in the formation order to form a complete transition parameter sequence.

[0055] Optionally, the S5 specifically includes:

[0056] S51: Using the corrected mechanical parameter transition values as input, discretize the three-dimensional space covered by the acoustic wave stratigraphic profile into equilateral voxel grids, and fill the compressive strength, elastic modulus, and Poisson's ratio in each voxel using the hierarchical Kriging interpolation method according to the spatial coordinates of the voxel center points to construct a three-dimensional formation parameter matrix with a resolution of 0.25m × 0.25m × 0.25m;

[0057] S52: For each voxel in the three-dimensional formation parameter matrix, calculate the wave impedance value based on the acoustic wave propagation velocity and rock density constant corresponding to the voxel; then use forward convolution modeling to generate a synthetic acoustic wave record, and perform a channel-by-channel comparison with the original acoustic wave profile to calculate the residual energy distribution between the two;

[0058] S53: Apply the Gauss–Newton iterative inversion method to minimize the residual energy, iteratively update the wave impedance value of the voxel and the gradient between adjacent voxels until the residual energy converges to below the preset threshold of 2%;

[0059] S54: Calculate the credibility index for each layer segment based on the final residual energy and the gradient smoothness between voxels, and classify the credibility into three grades: A, B, and C;

[0060] S55: Generate a geological stratification parameter calibration report, the report content includes the three-dimensional formation parameter matrix visualization slice and the credibility grade of each layer segment, and output and archive it in the form of an electronic signature.

[0061] Advantages of the present invention:

[0062] In the present invention, by establishing a multiple linear mapping relationship model between the acoustic wave propagation velocity and rock mechanical parameters, and combining the measured data of core samples for registration and correction, high-precision prediction and spatial extension of formation parameters are realized, effectively solving the problem of parameter missing caused by sparse core data; at the same time, by accurately identifying the formation interface through the sliding window energy ratio algorithm, and combining Hilbert envelope extraction and normalization processing, the stratification recognition ability of the acoustic wave profile is improved.

[0063] In the present invention, robust estimation and wave impedance inversion algorithms are introduced to eliminate anomalies and optimize the continuity of the initial prediction parameters, and the calibration results are quantitatively evaluated through a credibility grading mechanism, enhancing the engineering usability and interpretability of the parameter matrix. Description of the drawings

[0064] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings in the following description are only those of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0065] Figure 1 Schematic diagram of the geological stratification parameter calibration method according to the embodiment of the present invention;

[0066] Figure 2 Schematic diagram of the mapping relationship model construction and transition value correction method according to the embodiment of the present invention. Specific embodiments

[0067] The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. At the same time, it should be noted here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments. For some well-known technologies, those skilled in the art can also adopt other alternative methods for implementation; moreover, the accompanying drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0068] As Figure 1 - Figure 2 shown, a geological stratification parameter calibration method based on acoustic wave profile and core drilling test includes the following steps:

[0069] S1: Obtain the acoustic wave profile data of the target area, and perform time-frequency domain preprocessing on the acoustic wave profile to generate a preprocessed acoustic wave profile containing time-frequency characteristics;

[0070] S2: Based on the time-frequency characteristic differences of the preprocessed acoustic wave profile, use the sliding window energy ratio algorithm to divide the formation interface and generate an acoustic wave stratification profile with boundary marks;

[0071] S3: Perform core sampling at the corresponding positions of the key horizons of the acoustic wave stratification profile, and perform three-dimensional CT scanning and mechanical parameter tests on the core samples to generate a core parameter dataset containing spatial coordinates;

[0072] S4: Establish an acoustic wave velocity-mechanical parameter mapping relationship model, perform time-frequency characteristic matching between the core parameter dataset and the acoustic wave stratification profile, and correct the interlayer parameter transition value by using the robust estimation algorithm;

[0073] S5: Construct a three-dimensional formation parameter matrix based on the corrected parameter transition value, verify the parameter continuity by using the wave impedance inversion method, and output a geological stratification parameter calibration report with a credibility index.

[0074] S1 specifically includes:

[0075] S11: Conduct detection operations along the straight detection profile set in the target area using a towed multi-channel acoustic detection device. The detection device is configured with 12 receiving channels and 1 high-energy sound source. The center frequency of the sound source is set to 60 kHz, and the detection depth coverage range is 0 - 50 meters. In an environment where the water depth change is no more than ±2 meters, advance uniformly along the profile direction at a stable towing speed of 0.5 m / s, emit a pulse signal every 0.2 seconds, and synchronously collect the reflected signals of each channel to continuously obtain the acoustic raw profile data covering the target area.

[0076] S12: Denoise the obtained acoustic raw profile data, use a band-pass filter to remove signal interference with a frequency less than 10 Hz or greater than 500 Hz, and output the pure acoustic profile data after denoising.

[0077] S13: Perform Fourier transform on the pure acoustic profile data to obtain the frequency domain distribution characteristics. Then apply short-time Fourier transform to the frequency domain characteristics, set the short-time window length to 50 ms and the overlap rate to 50%, extract the spectral characteristics of the profile signal changing with time, and form a time-frequency characteristic matrix.

[0078] S14: Use Hilbert transform to extract the envelope of the time-frequency characteristic matrix, and perform normalization processing on the extracted envelope signal to generate a preprocessed acoustic profile containing complete time-frequency characteristics. Through the specific implementation of the above steps S11 to S14, interference factors are effectively removed, the time-frequency domain characteristics of the acoustic data are clarified, and the accuracy and reliability of subsequent formation division are ensured.

[0079] S14 specifically includes:

[0080] S141: For each time-domain signal sequence in the time-frequency characteristic matrix obtained in S13, perform Hilbert transform respectively to obtain the analytical expression of the corresponding signal. The Hilbert transform calculation formula is: , where represents the imaginary part of the analytical signal obtained by Hilbert transform; represents the signal amplitude of the time-domain signal to be processed at the integration variable ; t represents the target time point for solving the Hilbert transform result; is the integration variable, representing the integration moment of the signal; is the differential of the integration variable ; P.V. is the Cauchy principal value integral operator;

[0081] S142: Calculate the envelope amplitude of each time-domain signal sequence, combine the Hilbert transform result obtained in S141 with the original signal to form an analytical signal. The specific calculation formula is as follows: , where represents the instantaneous envelope amplitude of the time-domain signal; is the original amplitude of the time-domain signal at time t; is the imaginary part of the analytic signal obtained by Hilbert transform;

[0082] S143: Use the extreme value normalization method to normalize the envelope amplitude obtained in S142 to obtain a standardized time-frequency feature data matrix with a value range in [0,1]. The specific calculation formula is as follows: , where is the normalized envelope amplitude; is the maximum value of the envelope amplitude during the entire analysis period; is the minimum value of the envelope amplitude during the entire analysis period;

[0083] S144: Combine all the normalized envelope signals in the original channel order to form a preprocessed acoustic wave profile containing complete time-frequency features; Through the implementation of the above steps S141 to S144, the envelope features extracted by Hilbert transform make the formation structure features hidden in the acoustic wave signal more prominent. At the same time, the normalization process eliminates the interference of signal amplitude differences between different measurement channels and improves the accuracy and reliability of subsequent geological boundary recognition.

[0084] S2 specifically includes:

[0085] S21: Based on the preprocessed acoustic wave profile, set a sliding window with a length of 20 ms along the time axis of acoustic wave propagation, and slide it successively with a step size of 2 ms to intercept the signal data within each window;

[0086] S22: Calculate the short-time energy value of the signal data within each sliding window, sum the squares of the amplitudes of each sampling point within the window to obtain the energy value of the corresponding window. The calculation formula is as follows: , where represents the energy value of the signal within the current sliding window; represents the signal amplitude of the i-th sampling point within the current sliding window; N represents the number of sampling points within the current sliding window;

[0087] S23: Calculate the energy ratio between adjacent sliding windows in turn, and use the energy value of the window with larger energy divided by the energy value of the window with smaller energy as the energy ratio. The calculation formula is as follows: , where: represents the energy ratio between the j-th window and the j + 1-th window; represents the energy value within the j-th sliding window; represents the energy value within the j + 1-th sliding window;

[0088] S24: Compare the energy ratio obtained in S23 with the preset formation interface determination threshold of 3.0. When the energy ratio exceeds the threshold, mark the window boundary position corresponding thereto as the formation interface, and sequentially scan the entire acoustic profile to obtain all formation interface positions that meet the conditions;

[0089] S25: Superimpose all the marked formation interface positions onto the preprocessed acoustic profile to generate an acoustic stratification profile with boundary marks. Through the implementation of the above steps S21 to S25, the sliding window energy ratio algorithm is used to accurately identify the positions with significant energy differences between adjacent formations, effectively improving the recognition accuracy of the formation interface, thereby providing a clear spatial reference basis for subsequent core sampling and formation parameter calibration.

[0090] S3 specifically includes:

[0091] S31: According to the acoustic stratification profile obtained in S25, select the layer with the highest energy ratio and the most obvious continuity as the key layer, and record the three-dimensional spatial coordinate information of this key layer;

[0092] S32: Set the drilling position at the three-dimensional spatial coordinate position recorded for the key layer, and use the core drilling equipment to perform core sampling in the vertical direction. The drilling speed is set to 0.2 m / min, the core diameter is 50 mm, and the length of the target formation position drilled is 2 m to obtain a columnar core sample;

[0093] S33: Perform industrial-grade three-dimensional CT scanning on the obtained columnar core sample. The scanning parameters are set as: scanning voltage 120 kV, scanning current 180 mA, and spatial scanning resolution 0.1 mm to obtain the three-dimensional structural image data inside the core sample;

[0094] S34: Cut the columnar core sample into standard specimens with a specimen size of diameter 50 mm and height 100 mm. Perform a uniaxial compressive strength test on the standard specimens, load at a constant loading rate of 0.5 MPa / s until the specimens are damaged, and measure the compressive strength, elastic modulus, and Poisson's ratio parameters of each specimen;

[0095] S35: Integrate the three-dimensional spatial coordinates of the obtained core sample, the structural image data obtained by CT scanning, and the mechanical parameter test data to generate a core parameter dataset with spatial coordinate information. Through the implementation of the above steps S31 to S35, clarify the core sampling positions corresponding to the key layers, accurately obtain the internal structure of the formation and detailed mechanical property parameters, realize the accurate spatial alignment of the core parameters and the acoustic stratification profile, and provide high-precision basic data guarantee for subsequent geological parameter calibration.

[0096] S4 specifically includes:

[0097] S41: Extract the acoustic wave propagation velocity data between each stratigraphic interface, obtain the average acoustic wave propagation velocity within each stratification segment, and form a velocity vector set;

[0098] S42: Extract the core sample parameters corresponding to the spatial position of the stratification segment in S35, obtain the compressive strength, elastic modulus, and Poisson's ratio corresponding to each segment, and form a mechanical parameter vector set;

[0099] S43: Based on the correspondence between the velocity vector set and the mechanical parameter vector set, use the multiple linear regression method to construct a mapping relationship model between the acoustic wave velocity and the mechanical parameters, and output the predicted mechanical parameter values corresponding to each acoustic wave velocity value;

[0100] S44: Apply the established mapping relationship model to all stratification segments in the acoustic wave stratigraphic profile, combine its average acoustic wave velocity, calculate the initial mechanical parameter transition data, and perform spatial registration with the corresponding positions in the core data set;

[0101] S45: For the deviations in the initial parameter transition data caused by local anomalies or measurement errors, use the robust estimation algorithm to correct the parameters and form the corrected mechanical parameter transition values; Through the implementation of the above steps S41 to S45, a stable parameter mapping relationship is constructed by combining the velocity information and the core mechanical characteristics, and the robust estimation is introduced to improve the robustness of the model to abnormal data, laying an accurate data foundation for the construction of the three-dimensional stratigraphic parameter matrix.

[0102] S43 specifically includes:

[0103] S431: Use the multiple linear regression method to establish three linear mapping models with the compressive strength, elastic modulus, and Poisson's ratio as the dependent variables and the acoustic wave velocity as the independent variable respectively. The regression equations are expressed as follows:

[0104] ;

[0105] ;

[0106] ; In the formula, represents the acoustic wave propagation velocity of the i-th segment; , , represent the compressive strength, elastic modulus, and Poisson's ratio corresponding to the i-th segment respectively; , , are the slope coefficients of each regression model; , , are the intercept constant terms of each regression model;

[0107] S432: Fit the regression coefficients using the least squares method, minimize the squared error between the predicted values and the actual mechanical parameter values, and output the predicted mechanical parameter triples corresponding to each acoustic wave velocity value , for subsequent parameter transition construction; By implementing the above steps S431 to S432, a linear response relationship between the acoustic wave velocity and the key mechanical parameters is established, a stable and interpretable geomechanical parameter prediction model is constructed, significantly improving the parameter compensation ability of the acoustic wave profile in the non-core section area, and providing a complete prediction basis for robust correction and three-dimensional parameter matrix construction.

[0108] S44 specifically includes:

[0109] S441: Apply the established mapping relationship model between the acoustic wave velocity and the mechanical parameters to the average acoustic wave velocity values of each stratified segment in the acoustic wave stratified profile, and substitute each acoustic wave velocity value into the prediction model to obtain the corresponding initial mechanical parameters, including the uniaxial compressive strength , elastic modulus and Poisson's ratio , and their calculation formulas are as follows: ; ; ; In the formula, is the average acoustic wave propagation velocity of the i-th segment; , , are the predicted uniaxial compressive strength, elastic modulus and Poisson's ratio respectively; , , are the slope coefficients in the mapping model; , , are the intercept constant terms;

[0110] S442: Associate each set of calculated initial mechanical parameters with the spatial coordinates in the acoustic wave stratified profile to form a preliminary parameter transition dataset;

[0111] S443: Extract the spatial positions of each sample in the core drilling parameter dataset, and judge whether it meets the registration condition according to the spatial distance between the center position of the acoustic wave segment and the center position of the core drilling sample; When the spatial distance is within the preset error range, it is considered that the acoustic wave predicted parameter and the core drilling measured parameter form a valid matching pair;

[0112] Specifically, adopt the Euclidean distance matching method. Let the barycenter coordinates of the acoustic wave segment be , and the center coordinates of the core drilling sample be . When the following distance constraint condition is satisfied: , then register the acoustic wave predicted parameter with the core drilling sample in space; where is the spatial distance between the acoustic wave layer segment and the core sample, is the registration distance threshold (set to 0.5 m);

[0113] S444: Output all pairs of registered predicted parameters and measured parameters, which are used as the input basis for the subsequent robust estimation process, ensuring that the predicted data is highly consistent with the actual core position; Through the implementation of the above steps S441 to S444, combined with the average acoustic wave velocity and the established mapping relationship, the prediction generation of mechanical parameters is realized, and through the spatial registration technology, it is accurately aligned with the measured core data, providing a unified and reliable input basis for parameter anomaly identification and model correction.

[0114] S45 specifically includes:

[0115] S451: For each set of registered predicted parameters and measured parameters in S444, calculate the difference between the two to measure the model prediction deviation;

[0116] S452: According to the set residual threshold, judge each deviation value; When the deviation amplitude is within the allowable range, retain the deviation; When it exceeds the range, adjust the deviation to the threshold size and retain its positive and negative signs to limit the influence of outliers;

[0117] S453: Subtract the adjusted deviation amount from the original predicted parameters to obtain the corrected mechanical parameter transition value, ensuring that the parameters of each layer segment are closer to the measured values;

[0118] S454: Summarize all the corrected mechanical parameter transition values in the formation order to form a complete transition parameter sequence, which is directly used for the construction of the subsequent three-dimensional parameter matrix; Through the above steps, the abnormal fluctuations in the initial predicted parameters are effectively suppressed, the high-precision alignment between the model output and the measured data is realized, and the stability and reliability of the formation parameter calibration are significantly improved.

[0119] The steps for parameter correction using the robust estimation algorithm are as follows:

[0120] First, for each set of registered predicted mechanical parameters and measured parameters, calculate the residual value, and represent the predicted parameter as and represent the measured parameter as , then the i-th residual is calculated as the difference between the predicted value and the measured value, that is, the formula is: ;

[0121] Then, for each residual , apply the Huber influence function to calculate the correction component , where the threshold is set. When the absolute value of the residual does not exceed the threshold, take ; When the absolute value of the residual is greater than the threshold, take ; Specifically expressed as:

[0122] ;

[0123] In the formula: is the correction component of the i-th residual; is the preset residual threshold; is the sign function of;

[0124] Next, according to the Huber correction component update the prediction parameters to form the i-th corrected mechanical parameter transition value , which is calculated as the original predicted value minus the correction component, that is ;

[0125] Finally, summarize all in the order of layers to form a complete sequence of corrected mechanical parameter transition values.

[0126] S5 specifically includes:

[0127] S51: Taking the corrected mechanical parameter transition value as the input, discretize the three-dimensional space covered by the acoustic stratigraphic profile into equilateral voxels, and fill the compressive strength, elastic modulus, and Poisson's ratio in each voxel according to the spatial coordinates of the voxel center points, and construct a three-dimensional formation parameter matrix with a resolution of 0.25m×0.25m×0.25m;

[0128] S52: For each voxel in the three-dimensional formation parameter matrix, calculate the wave impedance value based on the acoustic wave propagation velocity corresponding to the voxel and the rock density constant; then use forward convolution modeling to generate a synthetic acoustic wave record, and compare it with the original acoustic wave profile channel by channel to calculate the residual energy distribution between the two;

[0129] The specific steps of the above S52 are as follows:

[0130] S521: For each voxel in the three-dimensional formation parameter matrix, extract the acoustic wave propagation velocity value at its center position and the set rock density constant, and calculate its corresponding wave impedance value. The formula is as follows: In the formula, represents the wave impedance value of the k-th voxel; represents the rock density constant of the k-th voxel (unit is ); represents the longitudinal wave propagation velocity of the k-th voxel (unit is m / s);

[0131] S522: According to the wave impedance sequence Arrangement along the detection path direction, calculating the reflection coefficient sequence between adjacent voxels , the reflection coefficient is calculated as follows: , where represents the reflection coefficient on the k-th interface; , are the wave impedance values on both sides of the interface respectively;

[0132] S523: Perform one-dimensional convolution on the reflection coefficient sequence and the standard wavelet signal to generate the synthetic acoustic record on this path, and its calculation expression is as follows: ; where represents the synthetic acoustic record; is the discrete form of the reflection coefficient sequence; is the standard wavelet used in the acquisition system; represents the convolution operation;

[0133] S524: For each detection path node, compare the generated synthetic acoustic record with the measured original acoustic profile data at the same position in a cross-section, and calculate its residual signal , and calculate the residual energy E in the form of time-domain squared error, and the expression is as follows: , where E is the residual energy of this path node; is the synthetic acoustic signal; is the original measured acoustic data; t is the sampling time point.

[0134] S53: Apply the Gauss–Newton iterative inversion method to minimize the residual energy, and iteratively update the voxel wave impedance value and the gradient between adjacent voxels until the residual energy converges to less than the preset threshold of 2%;

[0135] The specific steps of S53 are as follows:

[0136] S531: Define the objective function. For each path node, set the goal to minimize the residual energy between the synthetic acoustic record and the measured acoustic record; let the synthetic acoustic wave be , the measured acoustic wave be , and the wave impedance parameter vector be , then the residual vector is: ; the objective function is defined as the sum of squares of the residual vector as: ;

[0137] S532: Take the derivative of the wave impedance vector Z to construct the Jacobian matrix of the residual function with respect to the parameters , that is: , where represents the partial derivative of the i-th residual with respect to the j-th wave impedance parameter;

[0138] S533: The Gauss-Newton iteration method is used for minimization optimization. Let the current iteration step be k, then the parameter update amount is solved according to the following linear equations: ; The updated wave impedance parameter is: ;

[0139] S534: During the iteration process, in order to avoid mutations between the wave impedance values of adjacent voxels, a gradient constraint term is added to penalize the square of the impedance difference between adjacent voxels, making the solution have spatial smoothness, and an improved objective function is formed: , where is the smoothing coefficient (with a value range of 0.1 - 1.0), and this regularization term is incorporated into the Jacobian matrix solution in each iteration;

[0140] S535: Repeat the iteration step. If the residual energy change rate between two consecutive iterations satisfies:

[0141] , that is, the residual energy change rate is lower than 2%, then it is judged that convergence has occurred, and the current wave impedance distribution is output as the final inversion result.

[0142] S54: Calculate the credibility index for each layer segment based on the final residual energy and the gradient smoothness between voxels, and classify the credibility into three levels: A, B, and C. Among them, level A corresponds to a residual energy lower than 1% and a gradient smoothness higher than 95%; level B corresponds to a residual energy between 1% - 2% and a gradient smoothness between 90% - 95%; level C corresponds to a residual energy higher than 2% or a gradient smoothness lower than 90%;

[0143] S55: Generate a geological stratification parameter calibration report. The report content includes visual slices of the three-dimensional formation parameter matrix and the credibility level of each layer segment, and is output and archived in an electronically signed manner; Through the implementation of steps S51 to S55, high-precision interpolation of the correction parameters in three-dimensional space is achieved, the parameter continuity is strictly verified using wave impedance inversion, and the results are output in a quantified credibility form, thus ensuring the integrity and reliability of the geological stratification parameter calibration.

[0144] This invention covers any substitutions, modifications, equivalent methods, and solutions made within the essence and scope of this invention. To enable the public to have a thorough understanding of this invention, specific details are described in detail in the following preferred embodiments of this invention. However, those skilled in the art can fully understand this invention without the description of these details. Additionally, well-known methods, processes, procedures, components, and circuits are not described in detail to avoid unnecessary confusion to the essence of this invention.

[0145] The above are only the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A method for calibrating geological stratification parameters based on acoustic wave profiles and core drilling tests, characterized in that, It includes the following steps: S1: Obtain the acoustic wave profile data of the target area, and perform time-frequency domain preprocessing on the acoustic wave profile to generate a preprocessed acoustic wave profile containing time-frequency features; S2: Based on the time-frequency feature differences of the preprocessed acoustic wave profile, use the sliding window energy ratio algorithm to divide the formation interface and generate an acoustic wave stratified profile with boundary markers; S3: Implement core sampling at the corresponding positions of the key horizons in the acoustic wave stratified profile, and perform three-dimensional CT scanning and mechanical parameter tests on the core samples to generate a core parameter dataset containing spatial coordinates; S4: Establish a mapping relationship model between acoustic wave velocity and mechanical parameters, perform time-frequency feature matching between the core parameter dataset and the acoustic wave stratified profile, and correct the transitional values of interlayer parameters by using the robust estimation algorithm; The specific steps of S4 include: S41: Extract the acoustic wave propagation velocity data between each formation interface, obtain the average acoustic wave propagation velocity within each stratified section, and form a velocity vector set; S42: Extract the core sample parameters corresponding to the spatial positions of the stratified sections in S3, obtain the compressive strength, elastic modulus, and Poisson's ratio corresponding to each section, and form a mechanical parameter vector set; S43: Based on the corresponding relationship between the velocity vector set and the mechanical parameter vector set, use the multiple linear regression method to construct a mapping relationship model between acoustic wave velocity and mechanical parameters, and output the predicted mechanical parameter values corresponding to each acoustic wave velocity value; S44: Apply the established mapping relationship model to all stratified sections in the acoustic wave stratified profile, combine its average acoustic wave velocity, calculate the initial mechanical parameter transition data, and perform spatial registration with the corresponding positions in the core dataset; S45: For the deviations in the initial parameter transition data caused by local anomalies or measurement errors, use the robust estimation algorithm to correct the parameters and form the corrected mechanical parameter transition values; S5: Based on the corrected parameter transition values, construct a three-dimensional formation parameter matrix, use the wave impedance inversion method to verify the parameter continuity, and output a geological stratification parameter calibration report with a credibility index; The specific steps of S5 include: S51: Take the corrected mechanical parameter transition values as the input, discretize the three-dimensional space covered by the acoustic wave stratified profile into equilateral voxel grids, and use the stratified Kriging interpolation method to fill the compressive strength, elastic modulus, and Poisson's ratio values in each voxel according to the spatial coordinates of the voxel center points, and construct a three-dimensional formation parameter matrix with a resolution of 0.25m×0.25m×0.25m; S52: For each voxel in the three-dimensional formation parameter matrix, calculate the wave impedance value according to the acoustic wave propagation velocity and the rock density constant corresponding to the voxel; then use forward convolution modeling to generate a synthetic acoustic wave record, and perform a channel-by-channel comparison with the original acoustic wave profile to calculate the residual energy distribution between the two; S53: Apply the Gauss–Newton iterative inversion method to minimize the residual energy, iteratively update the voxel wave impedance values and the gradients between adjacent voxels until the residual energy converges to less than the preset threshold of 2%; S54: Calculate the credibility index of each section according to the final residual energy and the gradient smoothness between voxels, and classify the credibility into three levels: A, B, and C; S55: Generate a calibration report for geological stratification parameters. The report content includes visual slices of the 3D stratigraphic parameter matrix and the confidence level of each layer segment, and is output and archived in an electronically signed manner.

2. The geological stratification parameter calibration method based on acoustic wave profile and core test according to claim 1, wherein, The specific steps of S1 are as follows: S11: Use a towed multi-channel acoustic detection device to conduct detection operations along the straight detection profile set in the target area. The detection device is configured with 12 receiving channels and 1 high-energy sound source. The center frequency of the sound source is set to 60 kHz, and the detection depth coverage range is 0 - 50 meters. In an environment where the water depth change is no more than ±2 meters, it is uniformly advanced along the profile direction at a stable towed speed of 0.5 m / s. A pulse signal is emitted every 0.2 seconds, and the reflected signals of each channel are synchronously collected to continuously obtain the acoustic raw profile data covering the target area. S12: Denoise the obtained acoustic raw profile data. Use a band-pass filter to remove signal interference with frequencies less than 10 Hz or greater than 500 Hz, and output the pure acoustic profile data after denoising. S13: Perform a Fourier transform on the pure acoustic profile data to obtain the frequency domain distribution characteristics. Then apply a short-time Fourier transform to the frequency domain characteristics, set the short-time window length to 50 ms and the overlap rate to 50%, and extract the spectral characteristics of the profile signal changing with time to form a time-frequency characteristic matrix. S14: Use the Hilbert transform to extract the envelope of the time-frequency characteristic matrix, and perform normalization processing on the extracted envelope signal to generate a preprocessed acoustic profile containing complete time-frequency characteristics.

3. A method for calibrating geological stratification parameters based on acoustic profiles and core tests according to claim 2, characterized in that, The specific steps of S14 are as follows: S141: For each time-domain signal sequence in the time-frequency characteristic matrix obtained in S13, perform a Hilbert transform respectively to obtain the analytical expression of the corresponding signal. S142: Calculate the envelope amplitude of each time-domain signal sequence, combine the Hilbert transform result obtained in S141 with the original signal to form an analytical signal. S143: Use the extreme value normalization method to normalize the envelope amplitude obtained in S142 to obtain a standardized time-frequency characteristic data matrix with a value range within [0, 1]. S144: Combine all the normalized envelope signals in the original channel order to form a preprocessed acoustic profile containing complete time-frequency characteristics.

4. A method for calibrating geological stratification parameters based on acoustic profiles and core tests according to claim 1, characterized in that The specific steps of S2 are as follows: S21: Based on the preprocessed acoustic profile, set a sliding window with a length of 20 ms along the time axis direction of acoustic wave propagation, and slide it in turn with a step size of 2 ms to intercept the signal data within each window. S22: Calculate the short-time energy value of the signal data within each sliding window, sum the squares of the amplitudes of each sampling point within the window to obtain the energy value of the corresponding window. S23: Calculate the energy ratio between adjacent sliding windows in turn. S24: Compare the energy ratio obtained in S23 with the preset stratigraphic interface determination threshold of 3.

0. When the energy ratio exceeds the threshold, mark the corresponding window boundary position as the stratigraphic interface, and scan the entire acoustic profile in turn to obtain all the stratigraphic interface positions that meet the conditions. S25: Superimpose all the marked stratigraphic interface positions onto the preprocessed acoustic profile to generate an acoustic stratification profile with boundary marks.

5. A method for calibrating geological stratification parameters based on acoustic profiles and core tests according to claim 4, characterized in that, The specific steps of S3 are as follows: S31: According to the acoustic wave stratification profile obtained in S25, select the layer with the highest energy ratio as the key layer, and record the three-dimensional spatial coordinate information of this key layer; S32: Layout the drilling positions at the three-dimensional spatial coordinate positions recorded for the key layer. Use core drilling equipment to conduct core sampling in the vertical direction. Set the drilling speed to 0.2 m / min, the core diameter to 50 mm, and the drilling length of the target formation position to 2 m to obtain columnar core samples; S33: Conduct industrial-grade three-dimensional CT scanning on the obtained columnar core samples. Set the scanning parameters as: scanning voltage 120 kV, scanning current 180 mA, and spatial scanning resolution 0.1 mm to obtain the three-dimensional structural image data inside the core samples; S34: Cut the columnar core samples into standard specimens with a specimen size of diameter 50 mm and height 100 mm. Conduct uniaxial compressive strength tests on the standard specimens. Load at a constant loading rate of 0.5 MPa / s until the specimens are damaged, and measure the compressive strength, elastic modulus, and Poisson's ratio parameters of each specimen; S35: Integrate the three-dimensional spatial coordinates of the obtained core samples, the structural image data obtained from CT scanning, and the mechanical parameter test data to generate a core parameter dataset with spatial coordinate information.

6. The geological stratification parameter calibration method based on acoustic wave profile and core drilling test according to claim 1, wherein The specific steps of S43 are as follows: S431: Adopt the multiple linear regression method. Respectively take the compressive strength, elastic modulus, and Poisson's ratio as the dependent variables and the acoustic wave velocity as the independent variable to establish three linear mapping models. The regression equation expressions are respectively: ； ； ; wherein, represents the acoustic wave propagation velocity of the i-th layer segment; respectively represent the compressive strength, elastic modulus and Poisson's ratio corresponding to the i-th layer segment; is the slope coefficient of each regression model; is the intercept constant term of each regression model; S432: Fit the regression coefficients using the least squares method, minimize the squared error between the predicted values and the actual mechanical parameter values, and output the predicted mechanical parameter triples corresponding to each acoustic velocity value 。 7. A method for calibrating geological stratification parameters based on acoustic profiles and core tests according to claim 6, characterized in that, The specific steps of S44 are as follows: S441: Apply the established mapping relationship models between acoustic wave velocity and mechanical parameters to the average acoustic wave velocity values of each stratification segment in the acoustic wave stratification profile. Substitute each acoustic wave velocity value into the prediction model to obtain the corresponding initial mechanical parameters, including compressive strength, elastic modulus, and Poisson's ratio; S442: Associate the calculated initial mechanical parameters of each group with the spatial coordinates in the acoustic wave stratification profile to form a preliminary parameter transition dataset; S443: Extract the spatial positions of each sample in the core parameter dataset. According to the spatial distance between the center position of the acoustic wave segment and the center position of the core sample, judge whether it meets the registration condition; when the spatial distance is within the preset error range, it is considered that the acoustic wave prediction parameter and the core measured parameter form a valid matching pair; S444: Output all the registered prediction parameter and measured parameter pairs.

8. A method for calibrating geological stratification parameters based on acoustic profiles and core tests according to claim 7, characterized in that, The specific steps of S45 are as follows: S451: For each group of registered prediction parameters and measured parameters in S444, calculate the difference between the two; S452: According to the set residual threshold, judge each deviation value; when the deviation amplitude is within the allowable range, retain the deviation; when it exceeds the range, adjust the deviation to the threshold size and retain its positive or negative sign; S453: Subtract the adjusted deviation amount from the original prediction parameter to obtain the corrected mechanical parameter transition value; S454: Summarize all the corrected mechanical parameter transition values in the formation order to form a complete transition parameter sequence.

Images