Rock strength identification method and system based on drilling rig triaxial vibration spectrum characteristics

Abstract

The present invention discloses a rock strength identification method and system based on the three-axis vibration spectrum characteristics of a drilling rig. The method comprises: collecting vibration data, performing wavelet denoising on the vibration signal to generate denoised vibration data, extracting time domain eigenvalues ​​of the denoised vibration data, performing Fourier transform on the time domain eigenvalues ​​to generate frequency domain eigenvalues; normalizing the frequency domain eigenvalues ​​and extracting the main components in the frequency domain eigenvalues ​​by principal component analysis; setting a rock mechanics parameter prediction model, and predicting the uniaxial compressive strength, elastic modulus and Poisson's ratio of the rock according to the main components, thereby completing rock mechanics parameter identification.

Description

Technical Field

[0001] The present invention belongs to the technical field of rock mechanics parameter identification, and more specifically, relates to a rock strength identification method and system based on the three-axis vibration spectrum characteristics of a drilling rig. Background Art

[0002] The identification of rock mechanical parameters while drilling has always been a hot topic and difficulty in the field of drilling research, and is also a very important part of the while drilling technology. Traditional real-time identification technology of rock mechanical parameters often uses parameters such as torque, thrust, and drilling speed. These parameters are easy to collect, but the information they contain is often limited. They can only roughly judge the lithology and cannot accurately identify rock mechanical parameters. The drilling process often generates vibrations, and there is a lot of information hidden in the vibration signal to be mined. The identification of rock mechanical parameters through vibration characteristics is also a hot topic in the current while drilling field.

[0003] At present, most of the methods for predicting rock mechanical parameters using drilling parameters are inversion methods. Obtain rock mechanical parameters based on field lithology cataloging and core physical and mechanical tests; establish a standardized dataset of mapping relationships between drilling data and rock mechanical parameters; invert the drilling data based on the established dataset to obtain the corresponding rock mechanical parameters, and then realize the inversion of rock mechanical parameters based on drilling tests. Since the mechanical parameters of the rock are unknown before drilling, the database established based on field drilling data is blind and uncontrollable, which is not conducive to the accumulation of a large amount of data, and becomes a bottleneck for predicting rock mechanical parameters through drilling parameters (including vibration signals), affecting the prediction effect of rock mechanical parameters. Summary of the invention

[0004] In order to solve the above technical problems, the present invention proposes a rock strength identification method based on the three-axis vibration spectrum characteristics of the drilling rig, comprising:

[0005] Collecting vibration data, performing wavelet denoising on the vibration signal to generate denoised vibration data, extracting time domain eigenvalues ​​of the denoised vibration data, performing Fourier transform on the time domain eigenvalues ​​to generate frequency domain eigenvalues;

[0006] Normalizing the frequency domain eigenvalues, and extracting the main components of the frequency domain eigenvalues ​​by principal component analysis;

[0007] A rock mechanics parameter prediction model is set up, and the uniaxial compressive strength, elastic modulus and Poisson's ratio of the rock are predicted based on the main components, thereby completing the rock mechanics parameter identification.

[0008] Furthermore, collecting vibration data includes: preparing a cubic rock sample, performing an indoor rotary drilling test on the cubic rock sample, and collecting the vibration data.

[0009] Furthermore, normalizing the frequency domain eigenvalues ​​includes: scaling the frequency domain eigenvalues ​​to the same scale.

[0010] Furthermore, extracting the main components from the frequency domain eigenvalues ​​by principal component analysis includes: selecting the frequency domain eigenvalues ​​whose contribution rate reaches more than 80% as the main components according to the cumulative contribution rate of the frequency domain eigenvalues.

[0011] Furthermore, the rock mechanics parameter prediction model includes: taking the main components as independent variables, and establishing linear Lasso regression models with uniaxial compressive strength, elastic modulus and Poisson's ratio respectively.

[0012] Furthermore, the uniaxial compressive strength, the elastic modulus and the Poisson's ratio are compared with the test results of actual on-site coring samples, the error percentages between the uniaxial compressive strength, the elastic modulus and the Poisson's ratio and the actual values ​​are calculated, and the rock mechanics parameter prediction model is adjusted according to the error percentages.

[0013] The present invention also proposes a rock strength identification system based on the three-axis vibration spectrum characteristics of the drilling rig, comprising:

[0014] An acquisition module is used to acquire vibration data, perform wavelet denoising on the vibration signal to generate denoised vibration data, extract time domain eigenvalues ​​of the denoised vibration data, perform Fourier transform on the time domain eigenvalues ​​to generate frequency domain eigenvalues;

[0015] A main component extraction module is used to normalize the frequency domain eigenvalues ​​and extract the main components in the frequency domain eigenvalues ​​by using a main component analysis method;

[0016] The prediction module is used to set up a rock mechanics parameter prediction model and predict the uniaxial compressive strength, elastic modulus and Poisson's ratio of the rock based on the main components, thereby completing the rock mechanics parameter identification.

[0017] Furthermore, collecting vibration data includes: preparing a cubic rock sample, performing an indoor rotary drilling test on the cubic rock sample, and collecting the vibration data.

[0018] Furthermore, normalizing the frequency domain eigenvalues ​​includes: scaling the frequency domain eigenvalues ​​to the same scale.

[0019] Furthermore, extracting the main components from the frequency domain eigenvalues ​​by principal component analysis includes: selecting the frequency domain eigenvalues ​​whose contribution rate reaches more than 80% as the main components according to the cumulative contribution rate of the frequency domain eigenvalues.

[0020] Compared with the prior art, the above technical solution conceived by the present invention has the following beneficial effects:

[0021] The invention is based on the three-axis vibration spectrum characteristics of the drilling rig and improves the accuracy and practicability of rock mechanical parameter prediction by combining indoor and field tests. It can predict rock mechanical parameters in real time during the drilling process and improve drilling exploration efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 is a method flow chart of a rock strength identification method based on the three-axis vibration spectrum characteristics of a drilling rig involved in Embodiment 1 of the present invention;

[0023] Figure 2 It is a system structure diagram of embodiment 2 of the present invention. DETAILED DESCRIPTION

[0024] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0025] The method provided by the present invention can be implemented in the following terminal environment, and the terminal may include one or more of the following components: a processor, a storage medium, and a display screen. The storage medium stores at least one instruction, and the instruction is loaded and executed by the processor to implement the method described in the following embodiment.

[0026] The processor may include one or more processing cores. The processor uses various interfaces and lines to connect various parts in the entire terminal, and executes various functions of the terminal and processes data by running or executing instructions, programs, code sets or instruction sets stored in the storage medium, and calling data stored in the storage medium.

[0027] The storage medium may include a random access memory (RAM) or a read-only memory (ROM). The storage medium may be used to store instructions, programs, codes, code sets or instructions.

[0028] The display is used to show the interactive cross-section of each application.

[0029] All subscripts in the formulas of the present invention are only used to distinguish parameters and have no actual meaning.

[0030] In addition, those skilled in the art can understand that the structure of the above terminal does not constitute a limitation on the terminal, and the terminal may include more or fewer components, or combine certain components, or arrange the components differently. For example, the terminal also includes components such as a radio frequency circuit, an input unit, a sensor, an audio circuit, and a power supply, which will not be described in detail here.

[0031] Example 1

[0032] like Figure 1 As shown, the rock strength identification method based on the three-axis vibration spectrum characteristics of the drilling rig of the present invention is operated according to the following steps:

[0033] Step 1: Sample preparation

[0034] The rock samples collected on site are prepared into two standard specimens, cubic and cylindrical. The cubic specimen is used for indoor rotation test, and the cylindrical specimen is used for indoor uniaxial test. The two are numbered accordingly.

[0035] The size of the cubic specimen is 100mm*100mm*100mm, and the size of the cylindrical specimen is Φ50mm*100mm. Each rock sample on site can be used to make a cubic specimen and a cylindrical specimen, totaling 100 groups of rock samples, 100 cubic specimens and 100 cylindrical specimens.

[0036] Step 2: Indoor test

[0037] 2.1: Preparation before rotary drilling test

[0038] Before the test, the acceleration sensor is installed first. The sensor uses the 2.4GHz AX-3D acceleration sensor produced by Beanair of Germany. The sensor can use wireless transmission to transmit data to the computer through the wireless gateway, which is convenient for real-time viewing and storage in the computer.

[0039] 2.2: Indoor rotary drilling test method

[0040] During the test, the prepared cubic rock sample was placed in the indoor rotation testing machine and the acceleration sensor was turned on.

[0041] 2.3: Vibration signal acquisition

[0042] Collect vibration signals during the drilling process, save all data and number them for later use.

[0043] 2.4 Indoor uniaxial test

[0044] The prepared standard Φ50mm*100mm cylindrical specimen is placed in the RMT uniaxial testing machine to monitor the axial and lateral deformation of the specimen during the test. Through this experiment, the uniaxial compressive strength, elastic modulus, Poisson's ratio and other parameters of the rock sample can be obtained to provide calibration data for the model.

[0045] Step 3: Data Processing

[0046] 3.1: Vibration signal preprocessing

[0047] The vibration signals collected in the second step include data on the preparation moment and the end of drilling. To avoid its influence on the rock drilling process data, it is necessary to extract the pure drilling process data according to the recorded time and export it to MATLAB for data preprocessing.

[0048] 3.2: Vibration signal noise reduction

[0049] Since the vibration signal is mixed with environmental noise during acquisition, in order to more accurately analyze the relationship between the vibration signal and the rock mechanical parameters, MATLAB programming is used to perform wavelet denoising on the vibration signal. The threshold processing rules, soft threshold, wavelet packet decomposition and other parameters in the wavelet denoising toolbox are set to perform denoising. The waveform signal after denoising is saved and the denoised data is exported.

[0050] 3.3: Vibration time domain eigenvalue extraction

[0051] The exported processed vibration data is used to extract its time domain characteristic values ​​through MATLAB programming: mean value, standard deviation, mean square error, maximum value, minimum value, peak-to-peak value, skewness, kurtosis, amplitude factor, waveform factor, impact factor, and margin factor.

[0052] 3.4: Vibration frequency domain eigenvalue extraction

[0053] After the time domain signal is transformed into a frequency domain signal through Fourier transform, its frequency domain eigenvalues ​​can be extracted: centroid frequency, mean square frequency, root mean square frequency, frequency variance, and frequency standard deviation.

[0054] 3.5: Experimental data integration

[0055] Through the second and third steps, the mechanical parameters and vibration data of the same rock can be obtained. The vibration characteristic values ​​and rock mechanical parameters of the same rock are matched one by one and integrated into an EXCEL table as a data set for subsequent use.

[0056] Step 4: Build a prediction model

[0057] Since there is a high correlation between vibration eigenvalues ​​and rock mechanical parameters, the principal component regression (PCR) method is used to establish a prediction model with vibration eigenvalues ​​as input values ​​and uniaxial compressive strength, elastic modulus, Poisson's ratio, etc. as output values. Principal component regression is a statistical method that combines principal component analysis (PCA) and linear regression, aiming to deal with multicollinearity problems and reduce data dimensions. The specific operations are as follows:

[0058] 4.1: Performing Principal Component Analysis (PCA)

[0059] Since PCA is sensitive to the scale of the original data, all features need to be scaled to the same scale. Common methods include z-score standardization or range normalization.

[0060] 4.1.1 Standardization

[0061] This patent adopts the z-score standardization method to standardize the time domain characteristic values ​​of the vibration signal, including the mean value, standard deviation, mean square error, maximum value, minimum value, peak value, skewness, kurtosis, amplitude factor, waveform factor, impact factor, margin factor, and the frequency domain characteristic values ​​of the vibration signal, such as the center of gravity frequency, mean square frequency, root mean square frequency, frequency variance, and frequency standard deviation.

[0062] Calculates the arithmetic mean of all observations in the input parameter. and standard deviation , then the z-score standardized value of each observation It can be calculated by the following formula:

[0063] ,

[0064] Perform the above steps for each feature in the dataset separately to convert the entire dataset into a standard form.

[0065] The dataset is then divided into training set, test set, and validation set.

[0066] 4.1.2 Calculating the Correlation Matrix

[0067] Calculate the covariance. The specific formula is as follows:

[0068] ,

[0069] in, For any input parameter , For another different input parameter , For input parameters The average value of Another input parameter The average value of For parameters The deviation, For parameters deviation.

[0070] Calculate the correlation coefficient , the specific formula is as follows:

[0071] ,

[0072] in The input parameters are 、 The standard deviation of .

[0073] Based on the correlation coefficients, a correlation matrix is ​​constructed.

[0074] 4.1.3 Calculating Eigenvalues ​​and Eigenvectors

[0075] Solving the equations yields the eigenvalues ​​and corresponding eigenvectors of the correlation matrix.

[0076] 4.1.4 Data Conversion

[0077] The cumulative contribution rate of the eigenvalues ​​was calculated, and the first few principal components with a contribution rate of more than 80% were selected. The original data were transformed using the selected eigenvectors to obtain a low-dimensional principal component representation.

[0078] 4.2: Principal Components Regression Model

[0079] 4.2.1 Building a regression model

[0080] The transformed principal components were used as independent variables to establish linear Lasso regression models with the three dependent variables of uniaxial compressive strength, elastic modulus and Poisson's ratio.

[0081] Preferably, this embodiment optimizes the Lasso regression model to improve prediction accuracy:

[0082] ,

[0083] in, is the number of output variables, For the output variable, which is one of uniaxial compressive strength, elastic modulus, and Poisson's ratio, is the transpose of the regression coefficient, For the The main ingredients, is the adjustment factor, is the number of regression coefficients, For the The regression coefficients of the principal components are is a positive definite matrix representing an additional complexity penalty term that is used to capture the dependencies between the variables in the principal components.

[0084] Preferably, the present embodiment proposes the following model for predicting uniaxial compressive strength, elastic modulus and Poisson's ratio:

[0085] ,

[0086] ,

[0087] in, For in time The output variables, such as uniaxial compressive strength, elastic modulus, or Poisson's ratio, is the number of main components, For in time The Output variables With in time The Main ingredients The regression coefficient between For time The Main ingredients The nonlinear interactive feature mapping function is used to capture the complex nonlinear relationship between vibration features. For in time The Output variables The error term, is the number of kernel functions, For the The adjustment factor of the kernel function, For in time The Main ingredients In the The center of the kernel function, is the number of cross terms, For time The Main ingredients The transpose of For in time The Main ingredients In the The reference point in the intersection term, the reference point Specifically, it refers to the sample points used when calculating high-order cross terms. These points are used as benchmarks in the feature space to capture the nonlinear relationship between input features. For the The adjustment factor for the linear part of the cross terms can be used to capture the main components The linear relationship with itself, is the first adjustment factor to control the complexity of the cross terms, For in time The Main ingredients In the The reference point in the cross term, For the The adjustment factor for the linear part of the cross terms, is the second adjustment factor that controls the complexity of the cross terms.

[0088] 4.2.2 Training Model

[0089] The regression model is trained on the training set and the model prediction ability is evaluated on the test set. The R² (coefficient of determination), mean square error (MSE), and root mean square error (RMSE) are analyzed to select appropriate parameters.

[0090] Step 5: Field Trial

[0091] 5.1: Experimental preparation

[0092] Before the field test, the acceleration sensor was fixed on the drilling frame of the drilling rig.

[0093] 5.2: Vibration signal acquisition

[0094] The vibration signal of the drilling rig during the drilling process is collected through the acceleration sensor, and all the data are saved and numbered.

[0095] 5.3: Vibration signal preprocessing

[0096] The vibration signals collected in the second step include data on the preparation moment and the end of drilling. To avoid its influence on the rock drilling process data, it is necessary to extract the pure drilling process according to the recorded time and export it to MATLAB for data preprocessing.

[0097] 5.4: Vibration signal noise reduction

[0098] Since the vibration signal is mixed with environmental noise during acquisition, in order to more accurately analyze the relationship between the vibration signal and the rock mechanical parameters, MATLAB programming is used to perform wavelet denoising on the vibration signal. The threshold processing rules, soft threshold, wavelet packet decomposition and other parameters in the wavelet denoising toolbox are set to perform denoising. The waveform signal after denoising is saved and the denoised data is exported.

[0099] 5.5: Vibration time domain eigenvalue extraction

[0100] The exported processed vibration data is used to extract its time domain characteristic values ​​through MATLAB programming: mean value, standard deviation, mean square error, maximum value, minimum value, peak-to-peak value, skewness, kurtosis, amplitude factor, waveform factor, impact factor, and margin factor.

[0101] 5.6: Vibration frequency domain eigenvalue extraction

[0102] After the time domain signal is transformed into a frequency domain signal by Fourier transform, its frequency domain eigenvalues ​​are extracted: centroid frequency, mean square frequency, root mean square frequency, frequency variance, and frequency standard deviation.

[0103] Step 6: Predict outcomes and refine the model

[0104] 6.1: Prediction results

[0105] The vibration characteristic values ​​obtained in the fifth step are brought into the prediction model to predict the rock mechanical parameters. The prediction results may not be accurate and may deviate from the actual results.

[0106] 6.2: Improve the model

[0107] The prediction results are compared with the rock mechanical parameters of the rock samples obtained from actual coring, so as to correct and improve the Lasso regression model.

[0108] Example 2

[0109] like Figure 2 As shown, the embodiment of the present invention also proposes a rock strength identification system based on the three-axis vibration spectrum characteristics of the drilling rig, comprising:

[0110] An acquisition module is used to acquire vibration data, perform wavelet denoising on the vibration signal to generate denoised vibration data, extract time domain eigenvalues ​​of the denoised vibration data, perform Fourier transform on the time domain eigenvalues ​​to generate frequency domain eigenvalues;

[0111] Specifically, collecting vibration data includes: preparing a cubic rock sample, performing an indoor rotary drilling test on the cubic rock sample, and collecting the vibration data.

[0112] A main component extraction module is used to normalize the frequency domain eigenvalues ​​and extract the main components in the frequency domain eigenvalues ​​by using a main component analysis method;

[0113] Specifically, normalizing the frequency domain eigenvalues ​​includes: scaling the frequency domain eigenvalues ​​to the same scale.

[0114] Specifically, extracting the main components from the frequency domain eigenvalues ​​by the principal component analysis method includes: selecting the frequency domain eigenvalues ​​whose contribution rate reaches more than 80% as the main components according to the cumulative contribution rate of the frequency domain eigenvalues.

[0115] The prediction module is used to set up a rock mechanics parameter prediction model and predict the uniaxial compressive strength, elastic modulus and Poisson's ratio of the rock based on the main components, thereby completing the rock mechanics parameter identification.

[0116] Specifically, the rock mechanics parameter prediction model includes: taking the main components as independent variables, and establishing linear Lasso regression models with uniaxial compressive strength, elastic modulus and Poisson's ratio respectively.

[0117] Specifically, the uniaxial compressive strength, the elastic modulus and the Poisson's ratio are compared with the test results of actual on-site coring samples, the error percentages between the uniaxial compressive strength, the elastic modulus and the Poisson's ratio and the actual values ​​are calculated, and the rock mechanics parameter prediction model is adjusted according to the error percentages.

[0118] Example 3

[0119] The embodiment of the present invention further provides a storage medium storing a plurality of instructions, wherein the instructions are used to implement the rock strength identification method based on the three-axis vibration spectrum characteristics of the drilling rig.

[0120] Optionally, in this embodiment, the above storage medium may be located in any computer terminal in a computer terminal group in a computer network, or in any mobile terminal in a mobile terminal group.

[0121] Optionally, in this embodiment, the storage medium is configured to store program codes for executing the following steps: collecting vibration data, performing wavelet denoising on the vibration signal to generate denoised vibration data, extracting time domain eigenvalues ​​of the denoised vibration data, performing Fourier transform on the time domain eigenvalues ​​to generate frequency domain eigenvalues;

[0122] Specifically, collecting vibration data includes: preparing a cubic rock sample, performing an indoor rotary drilling test on the cubic rock sample, and collecting the vibration data.

[0123] Normalizing the frequency domain eigenvalues, and extracting the main components of the frequency domain eigenvalues ​​by principal component analysis;

[0124] Specifically, normalizing the frequency domain eigenvalues ​​includes: scaling the frequency domain eigenvalues ​​to the same scale.

[0125] Specifically, extracting the main components from the frequency domain eigenvalues ​​by the principal component analysis method includes: selecting the frequency domain eigenvalues ​​whose contribution rate reaches more than 80% as the main components according to the cumulative contribution rate of the frequency domain eigenvalues.

[0126] A rock mechanics parameter prediction model is set up, and the uniaxial compressive strength, elastic modulus and Poisson's ratio of the rock are predicted based on the main components, thereby completing the rock mechanics parameter identification.

[0127] Specifically, the rock mechanics parameter prediction model includes: taking the main components as independent variables, and establishing linear Lasso regression models with uniaxial compressive strength, elastic modulus and Poisson's ratio respectively.

[0128] Specifically, the uniaxial compressive strength, the elastic modulus and the Poisson's ratio are compared with the test results of actual on-site coring samples, the error percentages between the uniaxial compressive strength, the elastic modulus and the Poisson's ratio and the actual values ​​are calculated, and the rock mechanics parameter prediction model is adjusted according to the error percentages.

[0129] Example 4

[0130] An embodiment of the present invention also proposes an electronic device, including a processor and a storage medium connected to the processor, wherein the storage medium stores a plurality of instructions, and the instructions can be loaded and executed by the processor so that the processor can execute a rock strength identification method based on the three-axis vibration spectrum characteristics of the drilling rig.

[0131] Specifically, the electronic device of this embodiment may be a computer terminal, and the computer terminal may include: one or more processors, and a storage medium.

[0132] Among them, the storage medium can be used to store software programs and modules, such as a rock strength identification method based on the three-axis vibration spectrum characteristics of a drilling rig in an embodiment of the present invention, and the corresponding program instructions / modules. The processor executes various functional applications and data processing by running the software programs and modules stored in the storage medium, that is, realizing the above-mentioned rock strength identification method based on the three-axis vibration spectrum characteristics of the drilling rig. The storage medium may include a high-speed random storage medium, and may also include a non-volatile storage medium, such as one or more magnetic storage systems, flash memory, or other non-volatile solid-state storage media. In some instances, the storage medium may further include a storage medium remotely arranged relative to the processor, and these remote storage media may be connected to the terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0133] The processor can call the information and application stored in the storage medium through the transmission system to execute the steps of: collecting vibration data, performing wavelet denoising on the vibration signal to generate denoised vibration data, extracting time domain eigenvalues ​​of the denoised vibration data, performing Fourier transform on the time domain eigenvalues ​​to generate frequency domain eigenvalues;

[0134] Specifically, collecting vibration data includes: preparing a cubic rock sample, performing an indoor rotary drilling test on the cubic rock sample, and collecting the vibration data.

[0135] Normalizing the frequency domain eigenvalues, and extracting the main components of the frequency domain eigenvalues ​​by principal component analysis;

[0136] Specifically, normalizing the frequency domain eigenvalues ​​includes: scaling the frequency domain eigenvalues ​​to the same scale.

[0137] Specifically, extracting the main components from the frequency domain eigenvalues ​​by the principal component analysis method includes: selecting the frequency domain eigenvalues ​​whose contribution rate reaches more than 80% as the main components according to the cumulative contribution rate of the frequency domain eigenvalues.

[0138] A rock mechanics parameter prediction model is set up, and the uniaxial compressive strength, elastic modulus and Poisson's ratio of the rock are predicted based on the main components, thereby completing the rock mechanics parameter identification.

[0139] Specifically, the rock mechanics parameter prediction model includes: taking the main components as independent variables, and establishing linear Lasso regression models with uniaxial compressive strength, elastic modulus and Poisson's ratio respectively.

[0140] Specifically, the uniaxial compressive strength, the elastic modulus and the Poisson's ratio are compared with the test results of actual on-site coring samples, the error percentages between the uniaxial compressive strength, the elastic modulus and the Poisson's ratio and the actual values ​​are calculated, and the rock mechanics parameter prediction model is adjusted according to the error percentages.

[0141] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0142] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0143] In the several embodiments provided by the present invention, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the system embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0144] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0145] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0146] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only storage medium (ROM, Read-Only Memory), random access storage medium (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0147] Obviously, the above embodiments are merely examples for the purpose of clear explanation, and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived therefrom are still within the scope of protection of the invention.

Claims

1. A rock strength identification method based on the three-axis vibration spectrum characteristics of a drilling rig, characterized in that: include: Collecting vibration data, performing wavelet denoising on the vibration data to generate denoised vibration data, extracting time domain eigenvalues ​​of the denoised vibration data, performing Fourier transform on the time domain eigenvalues ​​to generate frequency domain eigenvalues; Normalizing the frequency domain eigenvalues, and extracting the main components of the frequency domain eigenvalues ​​by principal component analysis; A rock mechanics parameter prediction model is set, and the uniaxial compressive strength, elastic modulus and Poisson's ratio of the rock are predicted according to the main components, so as to complete the rock strength identification, wherein the rock mechanics parameter prediction model includes: ， ， in, For in time The output variables, is the number of main components, For in time The Output variables With in time The Main ingredients The regression coefficient between For time The Main ingredients The nonlinear interactive feature mapping function is used to capture the complex nonlinear relationship between vibration features. For in time The Output variables The error term, is the number of kernel functions, For the The adjustment factor of the kernel function, For in time The Main ingredients In the The center of the kernel function, is the number of cross terms, For time The Main ingredients The transpose of For in time The Main ingredients In the The reference point in the intersection term, the reference point Specifically, it refers to the sample points used when calculating high-order cross terms. These points are used as benchmarks in the feature space to capture the nonlinear relationship between input features. For the The adjustment factor for the linear part of the cross terms, is the first adjustment factor to control the complexity of the cross terms, For in time The Main ingredients In the The reference point in the cross term, For the The adjustment factor for the linear part of the cross terms, is the second adjustment factor that controls the complexity of the cross terms.

2. A rock strength identification method based on the three-axis vibration spectrum characteristics of a drilling rig as claimed in claim 1, characterized in that: Collecting vibration data includes: preparing a cubic rock sample, performing an indoor rotary drilling test on the cubic rock sample, and collecting the vibration data.

3. A rock strength identification method based on the three-axis vibration spectrum characteristics of a drilling rig as claimed in claim 1, characterized in that: Normalizing the frequency domain eigenvalues ​​includes: scaling the frequency domain eigenvalues ​​to the same scale.

4. A rock strength identification method based on the three-axis vibration spectrum characteristics of a drilling rig as claimed in claim 1, characterized in that: Extracting the main components from the frequency domain eigenvalues ​​by the principal component analysis method includes: selecting frequency domain eigenvalues ​​with a contribution rate of more than 80% as the main components according to the cumulative contribution rate of the frequency domain eigenvalues.

5. The rock strength identification method based on the three-axis vibration spectrum characteristics of the drilling rig according to claim 1 is characterized in that: The rock mechanics parameter prediction model comprises: taking the main components as independent variables, and establishing linear Lasso regression models with uniaxial compressive strength, elastic modulus and Poisson's ratio respectively.

6. A rock strength identification method based on the three-axis vibration spectrum characteristics of a drilling rig as claimed in claim 1, characterized in that: The uniaxial compressive strength, the elastic modulus and the Poisson's ratio are compared with the test results of actual on-site coring samples, the error percentages between the uniaxial compressive strength, the elastic modulus and the Poisson's ratio and the actual values ​​are calculated, and the rock mechanics parameter prediction model is adjusted according to the error percentages.

7. A rock strength identification system based on the three-axis vibration spectrum characteristics of a drilling rig, characterized in that: include: An acquisition module is used to acquire vibration data, perform wavelet denoising on the vibration data to generate denoised vibration data, extract time domain eigenvalues ​​of the denoised vibration data, perform Fourier transform on the time domain eigenvalues ​​to generate frequency domain eigenvalues; A main component extraction module is used to normalize the frequency domain eigenvalues ​​and extract the main components in the frequency domain eigenvalues ​​by using a main component analysis method; The prediction module is used to set a rock mechanics parameter prediction model and predict the uniaxial compressive strength, elastic modulus and Poisson's ratio of the rock according to the main components, thereby completing the rock strength identification, wherein the rock mechanics parameter prediction model includes: ， ， in, For in time The output variables, is the number of main components, For in time The Output variables With in time The Main ingredients The regression coefficient between For time The Main ingredients The nonlinear interactive feature mapping function is used to capture the complex nonlinear relationship between vibration features. For in time The Output variables The error term, is the number of kernel functions, For the The adjustment factor of the kernel function, For in time The Main ingredients In the The center of the kernel function, is the number of cross terms, For time The Main ingredients The transpose of For in time The Main ingredients In the The reference point in the intersection term, the reference point Specifically, it refers to the sample points used when calculating high-order cross terms. These points are used as benchmarks in the feature space to capture the nonlinear relationship between input features. For the The adjustment factor for the linear part of the cross terms, is the first adjustment factor to control the complexity of the cross terms, For in time The Main ingredients In the The reference point in the cross term, For the The adjustment factor for the linear part of the cross terms, is the second adjustment factor that controls the complexity of the cross terms.

8. A rock strength identification system based on the three-axis vibration spectrum characteristics of a drilling rig as claimed in claim 7, characterized in that: Collecting vibration data includes: preparing a cubic rock sample, performing an indoor rotary drilling test on the cubic rock sample, and collecting the vibration data.

9. A rock strength identification system based on the three-axis vibration spectrum characteristics of a drilling rig as claimed in claim 7, characterized in that: Normalizing the frequency domain eigenvalues ​​includes: scaling the frequency domain eigenvalues ​​to the same scale.

10. A rock strength identification system based on the three-axis vibration spectrum characteristics of a drilling rig as claimed in claim 7, characterized in that: Extracting the main components from the frequency domain eigenvalues ​​by the principal component analysis method includes: selecting frequency domain eigenvalues ​​with a contribution rate of more than 80% as the main components according to the cumulative contribution rate of the frequency domain eigenvalues.

Images