Well hole sound wave speed obtaining method and device based on frequency domain analysis

Abstract

The invention discloses a well hole sound wave speed obtaining method and device based on frequency domain analysis, and relates to the field of geophysical acoustic logging, and the method comprises the steps: obtaining an array waveform signal obtained by carrying out full wave train sound wave logging on a target layer section; performing Fourier transform on the array waveform signal to obtain a waveform frequency spectrum signal; extracting a noisy frequency dispersion signal from the waveform frequency spectrum signal, and carrying out noise reduction processing on the noisy frequency dispersion signal to obtain an effective frequency dispersion signal; and performing signal analysis processing on the effective frequency dispersion signal to obtain the well hole sound wave speed. According to the method, real-time acquisition of the stratum sound velocity can be realized while the acoustic logging data is acquired, the accuracy and stability of sound velocity acquisition are improved, the problems that a traditional frequency domain processing method needs manual interaction, the processing efficiency is low and the like are solved, and the problems that a sound velocity processing result is inaccurate and unstable are solved; and the defects of a time domain sound velocity real-time processing method are overcome.

Description

Technical Field

[0001] This application relates to the field of geophysical acoustic logging, and specifically to a method and device for obtaining borehole acoustic wave velocity based on frequency domain analysis. Background Art

[0002] Obtaining the elastic wave velocity profile propagating along the borehole wall is the core task of acoustic logging, which provides a necessary data basis for in-depth formation analysis. The automation of acoustic velocity processing is an urgent requirement for on-site logging operations and a key technology for realizing automatic formation analysis and real-time evaluation.

[0003] Time domain processing methods are the mainstream technical means for obtaining formation acoustic velocity. The threshold method is a widely used real-time acoustic velocity processing method. This method takes the waveform amplitude that first reaches the threshold as the arrival time of the formation wave signal, and combines the receiver array spacing of the instrument to calculate the acoustic wave velocity. However, the threshold method cannot distinguish noise from the true formation wave signal, and its processing results are easily affected by noise. The slowness-time coherence (STC) method has good anti-interference ability and high processing efficiency, and is the most widely used acoustic velocity processing method. In actual data processing, the STC method usually requires fine adjustment of various parameters such as time window, filtering, and correlation search based on expert experience to obtain the formation acoustic velocity. Researchers have tried to use methods such as peak search, Kalman filtering, and artificial intelligence (Sun Zhifeng et al., Method and Device for Real-Time Calculation of Formation P-Wave and S-Wave Slownesses; Sun et al., A Stabilized Real-Time Slowness Estimation Method for Compressional Waves by Using Kalman Filtering; Fan Chuan et al., A Method for Calculating Acoustic Travel Time Difference) to automatically determine the processing parameters of STC to achieve automatic extraction of formation acoustic velocity. However, in the face of complex geological environments, these methods are difficult to remove noise interference and accurately identify the correlation changes of array waveforms, resulting in large errors and unstable fluctuations in the acoustic velocity processing results. Therefore, there is an urgent need for a solution that can accurately extract borehole acoustic wave velocity in real time. Summary of the Invention

[0004] In view of the above problems, this application is proposed to provide a method, device, computing device, computer storage medium, and computer program product for obtaining borehole acoustic wave velocity based on frequency domain analysis that overcomes the above problems or at least partially solves the above problems.

[0005] According to one aspect of the embodiments of the present application, a method for obtaining borehole acoustic wave velocity based on frequency domain analysis is provided. The method includes: Obtaining an array waveform signal obtained by performing full-waveform acoustic logging on a target interval; Perform a Fourier transform on the array waveform signal to obtain a waveform spectrum signal; Extract the noisy dispersion signal from the waveform spectrum signal, and perform noise reduction processing on the noisy dispersion signal to obtain an effective dispersion signal; Perform signal analysis processing on the effective dispersion signal to obtain the borehole acoustic velocity.

[0006] Further, performing signal analysis processing on the effective dispersion signal to obtain the borehole acoustic velocity further includes: If the effective dispersion signal is a Stoneley wave dispersion signal and / or a longitudinal wave dispersion signal, perform frequency distribution statistics on the Stoneley wave dispersion signal and / or the longitudinal wave dispersion signal, and determine the velocity corresponding to the maximum frequency as the Stoneley wave velocity and / or the longitudinal wave velocity; If the effective dispersion signal is a flexural wave dispersion signal and / or a helical wave dispersion signal, perform fitting processing on the flexural wave dispersion signal and / or the helical wave dispersion signal to obtain the shear wave velocity.

[0007] Further, performing fitting processing on the flexural wave dispersion signal and / or the helical wave dispersion signal to obtain the shear wave velocity further includes: Input the borehole acoustic parameters into a pre-trained fast dispersion simulation network to obtain the corresponding theoretical simulated dispersion signal, where the borehole acoustic parameters include: the borehole mud velocity, the borehole diameter, and the formation shear wave velocity; Determine the formation shear wave velocity when the difference between the flexural wave dispersion signal and the theoretical simulated dispersion signal is the smallest as the shear wave velocity, and / or determine the formation shear wave velocity when the difference between the helical wave dispersion signal and the theoretical simulated dispersion signal is the smallest as the shear wave velocity.

[0008] Further, the training process of the fast dispersion simulation network includes: Generate a preset number of theoretical simulated dispersion signals based on the dispersion equation of the borehole acoustic field; Obtain the borehole acoustic sample parameters, and use the borehole acoustic sample parameters as the training dataset features and the theoretical simulated dispersion signal as the training dataset label to train the neural network to obtain the fast dispersion simulation network, where the borehole acoustic sample parameters include: the borehole mud velocity, the borehole diameter, and the formation shear wave velocity.

[0009] Further, performing noise reduction processing on the noisy dispersion signal to obtain the effective dispersion signal further includes: Use a pre-trained dispersion noise reduction model to perform noise reduction processing on the noisy dispersion signal to obtain the effective dispersion signal; Among them, the training process of the dispersion noise reduction model includes: Obtain the noisy sample dispersion signal and the effective sample dispersion signal; Train a deep learning model using noisy sample dispersion signals to obtain a model output signal; Calculate the loss function between the model output signal and the effective sample dispersion signal; Adjust the model parameters of the deep learning model based on the loss function; Iteratively execute the model training steps until a preset training termination condition is met.

[0010] Further, extracting the noisy dispersion signal from the waveform frequency spectrum signal further includes: Extract the noisy dispersion signal from the waveform frequency spectrum signal using the matrix pencil method.

[0011] According to another aspect of the embodiments of the present application, there is provided a borehole acoustic velocity acquisition device based on frequency domain analysis. The device includes: An acquisition module adapted to acquire an array waveform signal obtained by performing full waveform acoustic logging on a target interval; A Fourier transform module adapted to perform a Fourier transform on the array waveform signal to obtain a waveform frequency spectrum signal; An extraction module adapted to extract a noisy dispersion signal from the waveform frequency spectrum signal; A noise reduction processing module adapted to perform noise reduction processing on the noisy dispersion signal to obtain an effective dispersion signal; A signal analysis and processing module adapted to perform signal analysis and processing on the effective dispersion signal to obtain the borehole acoustic velocity.

[0012] According to yet another aspect of the embodiments of the present application, there is provided a computing device, including: a processor, a memory, a communication interface, and a communication bus. The processor, the memory, and the communication interface complete communication with each other through the communication bus; The memory is used to store at least one executable instruction, and the executable instruction causes the processor to execute the operations corresponding to the above-mentioned borehole acoustic velocity acquisition method based on frequency domain analysis.

[0013] According to still another aspect of the embodiments of the present application, there is provided a computer storage medium storing at least one executable instruction, and the executable instruction causes the processor to execute the operations corresponding to the above-mentioned borehole acoustic velocity acquisition method based on frequency domain analysis.

[0014] According to still another aspect of the present application, there is provided a computer program product including at least one executable instruction, and the executable instruction causes the processor to execute the operations corresponding to the above-mentioned borehole acoustic velocity acquisition method based on frequency domain analysis.

[0015] This application transforms the array waveform from the time domain to the frequency domain, and calculates the noisy borehole mode wave dispersion signal through the waveform spectrum. Secondly, a specifically trained and optimized deep learning model is used to efficiently clean the dispersion signal data without manual interaction. Finally, through a dispersion correction technology driven by both data and model, the longitudinal wave, transverse wave, and Stoneley wave velocities in the cable and logging-while-drilling environments are automatically obtained.

[0016] By converting the array waveform signal obtained from full-wave acoustic logging to frequency-domain processing to obtain the velocities of various mode waves, and solving problems such as the need for manual interaction and slow processing efficiency in traditional frequency-domain processing methods, the acoustic logging data can be processed while being collected, realizing the real-time acquisition of formation acoustic velocity. This application can be applied to the real-time acquisition of various acoustic velocities such as longitudinal waves, transverse waves, and Stoneley waves in cable logging and logging-while-drilling scenarios, and can become a powerful tool for on-site data processing and real-time analysis and decision-making. At the same time, it overcomes the problems of inaccurate and unstable acoustic velocity processing results in the existing technology, and solves the deficiencies of the time-domain acoustic velocity real-time processing method.

[0017] The above description is only an overview of the technical solution of this application. In order to be able to more clearly understand the technical means of this application, it can be implemented in accordance with the content of the specification. And in order to make the above and other purposes, features, and advantages of this application more obvious and understandable, the following specifically illustrates the specific implementation manners of this application. Brief Description of the Drawings

[0018] By reading the detailed description of the preferred embodiments below, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only for the purpose of showing the preferred embodiments and are not considered to be a limitation of this application. And throughout the drawings, the same reference numerals are used to represent the same components. In the drawings: Figure 1 It shows a schematic flow chart of a method for obtaining borehole acoustic velocity based on frequency-domain analysis according to an embodiment of this application; Figure 2 It shows a schematic diagram of an array acoustic waveform signal; Figure 3 It shows a schematic diagram of obtaining a waveform spectrum signal by performing a Fourier transform on the array acoustic waveform signal; Figure 4 It shows a schematic diagram of the noisy dispersion signal extracted from the waveform spectrum signal; Figure 5 It shows a schematic diagram of the structure of the dispersion noise reduction model; Figure 6 It shows a schematic diagram of the effect of the dispersion noise reduction model on denoising the noisy dispersion signal; Figure 7It shows a schematic diagram of the effect of the dispersion noise reduction model on reducing the noise of the dispersion signal in the well section; Figure 8 It shows a schematic diagram of the longitudinal wave velocity obtained by processing the logging-while-drilling monopole acoustic logging signal in a soft formation using the method of the present application; Figure 9 It shows a schematic diagram of the Stoneley wave velocity obtained by processing the logging-while-drilling monopole acoustic logging signal in a soft formation using the method of the present application; Figure 10 It shows a schematic diagram of the formation shear wave velocity obtained by processing the wireline dipole acoustic logging signal in hard and soft formations using the method of the present application; Figure 11 It shows a schematic diagram of the flow of the borehole acoustic velocity acquisition method based on frequency domain analysis according to another embodiment of the present application; Figure 12 It shows a structural block diagram of the borehole acoustic velocity acquisition device based on frequency domain analysis according to an embodiment of the present application; Figure 13 It shows a schematic diagram of the structure of a computing device according to an embodiment of the present application. Detailed implementation manners

[0019] Hereinafter, the exemplary embodiments of the present disclosure will be described in more detail with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be completely conveyed to those skilled in the art.

[0020] Figure 1 It shows a schematic diagram of the flow of the borehole acoustic velocity acquisition method based on frequency domain analysis according to an embodiment of the present application. As Figure 1 shown, the method includes the following steps: Step S101, acquire the array waveform signal obtained by performing full waveform acoustic logging on the target interval.

[0021] Specifically, in the target interval, full waveform acoustic logging is performed, and the entire acoustic array is collected by N receivers. The array waveform signal obtained by the nth receiver can be expressed as x n (t), thereby obtaining the array waveform signal. The array waveform signal collected here is a signal in the time domain.

[0022] Figure 2 It shows a schematic diagram of the array acoustic waveform signal. As Figure 2 shown, it schematically shows the array waveform signals collected by 8 receivers.

[0023] To make the spectrum of the signal after Fourier transform dense enough, before performing the discrete Fourier transform, zero-padding processing can be performed on the array waveform signal in the time domain so that the total number of data points of the array waveform signal after zero-padding processing reaches a preset threshold. For example, zero-padding the array waveform signal in the time domain until the total number of data points of the entire array waveform signal reaches 2048. Through the zero-padding operation, the spectrum resolution can be improved during the subsequent Fourier transform.

[0024] Step S102: Perform a Fourier transform on the array waveform signal to obtain a waveform spectrum signal.

[0025] Specifically, the discrete Fourier transform can be used to process the array waveform signal, transforming the array waveform signal from the time domain to the frequency domain. After the transformation is completed, the waveform spectrum signal X n (ω) corresponding to the array waveform signal can be obtained. By transforming the array waveform signal from the time domain to the frequency domain, it is convenient to utilize the characteristic that different signals have different frequency characteristics for signal denoising processing, separating the noise signal and the effective signal in the frequency domain, thereby providing a basis for accurately obtaining the sound speed subsequently.

[0026] For example, the discrete Fourier transform formula (1) can be used to perform transformation processing on the array waveform signal to obtain the waveform spectrum signal: (1) where X n (ω) is the waveform spectrum signal, x n (t) is the array waveform signal, and N is the total number of receivers.

[0027] In actual production, all array waveform signals are discrete time-domain signals. Therefore, the discrete Fourier transform needs to be used to transform them into discrete waveform spectrum signals.

[0028] Figure 3 Shows a schematic diagram of the waveform spectrum signal obtained after performing a Fourier transform on the array waveform signal in Figure 2 . After the Fourier transform processing, the array waveform signals collected by 8 receivers are transformed from the time domain to the frequency domain.

[0029] Step S103: Extract the noisy dispersion signal from the waveform spectrum signal and perform noise reduction processing on the noisy dispersion signal to obtain an effective dispersion signal.

[0030] After obtaining the waveform spectrum signal X n (ω), methods such as filtering, wavelet transform, and machine learning can be used to extract from the waveform spectrum signal X nExtract the noisy dispersion signal \(v(\omega)\) from \((\omega)\). Extracting the dispersion signal from the waveform spectrum signal facilitates revealing the differences in the propagation characteristics of different frequency components in the signal, thereby obtaining the sound velocities of various mode waves and facilitating noise reduction processing.

[0031] The noise of the signal is not only in the frequency-domain amplitude spectrum but also in the frequency-domain phase spectrum of the array waveform. It is difficult to achieve a good noise reduction effect by only performing noise reduction processing on the full waveform spectrum signal. Compared with the spectrum signal of the full band, the dispersion signal can synthesize the frequency-domain amplitude spectrum and phase spectrum of the signal, directly map the spectrum to the frequency-velocity domain, and can more directly characterize the wave velocity dispersion effect and correct it compared with other methods to obtain more accurate sound velocity information.

[0032] In an alternative embodiment, the matrix pencil method can be used to obtain the noisy dispersion signal from the waveform spectrum signal. The array waveform signal contains waveform signals of various modes such as Stoneley wave, longitudinal wave, flexural wave, and helical wave. Therefore, the waveform spectrum signal can be expressed as a linear superposition of the eigenmodes of \(M\) mode waves. The specific expression can be seen in Equation (2):

[0033] where \(X\) n \((\omega)\) is the spectrum of the waveform collected by the \(n\)th receiver at the angular frequency \(\omega\), \(M\) is the number of mode waves, \(A\) m \((\omega)\), and \(v\) m \((\omega)\) are respectively the amplitude, phase, and noisy dispersion signal of the \(m\)th mode wave, \(d\) is the receiver spacing, and \(N\) is the total number of receivers.

[0034] The phase contains dispersion information. The matrix pencil method obtains the dispersion by solving the waveform phase difference between adjacent receivers.

[0035] First, construct two \((N - M)\times M\) matrices, \(Y_1\) and \(Y_2\). The specific expressions of \(Y_1\) and \(Y_2\) can be seen in Equation (3) and Equation (4): (3) (4) where \(X\) n is the abbreviation of \(X\) n \((\omega)\) ( \(n = 1, 2, 3, 4, \cdots, M, M + 1, \cdots N - M, N - M + 1, \cdots, N - 1, N\)), representing the waveform spectrum signal of the \(n\)th receiver at the angular frequency \(\omega\). Then, solve the generalized eigenvalues of \(Y_1\) and \(Y_2\). The specific expression for the solution is: (5) where is the Moore-Penrose inverse of Y1, λ is the generalized eigenvalue, and I is the M-order identity matrix.

[0036] According to formula (5), M generalized eigenvalues λ can be obtained. It can be proven that λ and are equal and in one-to-one correspondence, that is, (6) According to formula (6), the phase difference can be obtained as: (7) where Im(λ m ) and Re(λ m ) are the imaginary part and real part of λ m respectively. According to the expression , the phase difference should always be less than or equal to 0. And the value range of formula (7) is (-π, π]. Therefore, for positive results, that is, the [0, π] part in the value range, subtracting 2π can change it to [-2π, -π], so as to map the value range of formula (7) from (-π, π] to the range of [-2π, -π].

[0037] According to formula (6) and formula (7), the noisy dispersion signal corresponding to the m-th mode wave can be obtained as: (8) Figure 4 shows a schematic diagram of extracting the noisy dispersion signal from the Figure 3 shown waveform spectrum signal. Among them, channel 1 is the noisy dispersion signal corresponding to the 1st mode wave, and channel 2 is the noisy dispersion signal corresponding to the 2nd mode wave.

[0038] Perform noise reduction processing on the noisy dispersion signal v(ω) to obtain the effective dispersion signal v T (ω). For example, wavelet transform method, adaptive threshold method, and EMD method can be used for noise reduction processing.

[0039] In an alternative embodiment, a dispersion noise reduction model is pre-trained. Therefore, the pre-trained dispersion noise reduction model can be used to perform noise reduction processing on the noisy dispersion signal to obtain the effective dispersion signal. By performing noise reduction processing on the noisy dispersion signal, noise interference can be suppressed, effective signal features can be enhanced, and the accuracy, reliability, and effectiveness of subsequent analysis can be improved.

[0040] Among them, the training process of the dispersion noise reduction model includes: obtaining the noisy sample dispersion signal and the effective sample dispersion signal; training the deep learning model with the noisy sample dispersion signal to obtain the model output signal; calculating the loss function between the model output signal and the effective sample dispersion signal; adjusting the model parameters of the deep learning model based on the loss function; and iteratively executing the model training steps until the preset training termination condition is met.

[0041] Specifically, after a large number of processes are performed using step S103, a data set can be constructed. Among them, the data set contains the noisy sample dispersion signal and the effective sample dispersion signal. The noisy sample dispersion signal v(ω) is used as the data set feature, and the effective sample dispersion signal v T (ω) is used as the data set label. Based on this data set, the deep learning model is trained to train a dispersion noise reduction model for intelligent dispersion noise reduction. The trained dispersion noise reduction model can be used to automatically reduce the noise of the noisy dispersion signal.

[0042] Specifically, the noisy sample dispersion signal is input into the deep learning model. After the calculations and feature extractions of each layer inside the model (which can be called forward propagation processing), the model output signal is obtained. To evaluate the difference between the model output signal and the real signal, it is necessary to calculate the loss function between the model output signal and the effective sample dispersion signal. This function can quantify the error magnitude between the two. Based on the calculated loss function value, the model parameters of the deep learning model are adjusted using the backpropagation algorithm and an optimizer (such as stochastic gradient descent, Adam optimizer, etc.). By continuously updating the parameters, the value of the loss function is reduced, making the model output signal closer to the effective sample dispersion signal. Then, the above model training steps are iteratively executed, continuously inputting the noisy sample dispersion signal, calculating the loss function, and adjusting the model parameters until the preset training termination condition is met, such as the value of the loss function drops below a certain threshold, the maximum number of training epochs is reached, or the performance on the validation set no longer improves, etc. At this time, it is considered that the deep learning model training is completed, and thus the dispersion noise reduction model is obtained.

[0043] Before model training, the deep learning model is first initialized. The deep learning model includes a one-dimensional convolutional layer, a batch normalization layer, an activation function, a pooling layer, and a deconvolution layer. The structure of the deep learning model is as Figure 5As shown in the figure, the forward propagation process of the deep learning model is as follows: The encoder composed of a one-dimensional convolutional layer and others performs local feature extraction and dimensional adjustment on the input noisy sample dispersion signal; the pooling layer performs an operation of taking the maximum value in the local area of the features output by the encoder to reduce the spatial size of the feature map, thereby reducing the computational complexity and extracting the main features. The combination of multiple encoders is used to extract signal features with different resolutions, and then the transposed convolutional layer processes them to restore the length of the noisy sequence; finally, the Sigmoid activation function limits the output of the transposed convolutional layer within the range of [0, 1].

[0044] When using the dispersion noise reduction model for noise reduction processing, its processing process is similar to the forward propagation process during training, which will not be elaborated here.

[0045] The output of the dispersion noise reduction model has the same shape as the noisy dispersion signal v(ω), and the magnitude of its output value represents the validity probability of the data points in the noisy dispersion signal v(ω). The set of data points with a validity probability greater than 0.9 constitutes the denoised effective dispersion signal v T (ω). Figure 6 Shows the effect of the dispersion noise reduction model on Figure 4 the noisy dispersion signal shown in the figure after noise reduction processing, Figure 7 Shows the effect diagram of the dispersion noise reduction model on the noisy dispersion signal of the well section. After converting the unit of the noisy dispersion signal from m / s in the velocity domain to μs / ft in the slowness domain, it is then displayed. Figure 7 In the first trace is the depth, the second trace is the array waveform, the third trace is the noisy dispersion signal extracted from the waveform, and the fourth trace is the effective dispersion signal after noise reduction using the dispersion noise reduction model.

[0046] Step S104, perform signal analysis processing on the effective dispersion signal to obtain the borehole acoustic velocity.

[0047] The effective dispersion signal is the dispersion signal after noise reduction processing, which does not contain noise. Therefore, by performing signal analysis processing on the effective dispersion signal, the borehole acoustic velocity can be accurately obtained. The signal analysis processing here mainly includes: frequency distribution statistics and fitting processing. Different processing methods are adopted for the effective dispersion signals of different mode waves.

[0048] For example, if the effective dispersion signal is the Stoneley wave dispersion signal and / or the longitudinal wave dispersion signal, then perform frequency distribution statistics on the Stoneley wave dispersion signal and / or the longitudinal wave dispersion signal, and determine the velocity corresponding to the maximum frequency as the Stoneley wave velocity and / or the longitudinal wave velocity. Figure 8 Shows the schematic diagram of the longitudinal wave velocity obtained by using the method of the present application to process the logging-while-drilling monopole acoustic logging signal in a soft formation; Figure 9Shows a schematic diagram of the Stoneley wave acoustic velocity obtained by processing the monopole acoustic logging while drilling signal in a soft formation using the method of the present application.

[0049] If the effective dispersion signal is a flexural wave dispersion signal and / or a helical wave dispersion signal, then perform fitting processing on the flexural wave dispersion signal and / or the helical wave dispersion signal to obtain the shear wave acoustic velocity. For example, nonlinear least squares method or neural network can be used for fitting processing, and other methods suitable for fitting processing are also included in the protection scope of the present application.

[0050] In order to improve the fitting efficiency, a pre-trained fast dispersion simulation network can be used to obtain the theoretical simulated dispersion signal. Specifically, during logging, borehole acoustic parameters will be obtained. Among them, the borehole acoustic parameters mainly include: the highly sensitive parameter of dispersion, the borehole mud acoustic velocity v f , the borehole diameter R, and the formation shear wave acoustic velocity v s . Input the borehole acoustic parameters into the pre-trained fast dispersion simulation network to obtain the corresponding theoretical simulated dispersion signal v sim (ω). Determine the formation shear wave acoustic velocity when the difference between the measured effective dispersion signal and the theoretical simulated dispersion signal v sim (ω) is the smallest as the shear wave acoustic velocity. For the dipole logging data, the effective dispersion signal refers to the flexural wave dispersion signal, and for the quadrupole logging data, it refers to the helical wave dispersion signal.

[0051] For example, use the Euclidean norm of the theoretical simulated dispersion signal v sim (ω) and the effective dispersion signal v T (ω) (flexural wave dispersion signal or helical wave dispersion signal) as the objective function to represent the difference between the two. Use the grid search method to find the best formation shear wave acoustic velocity that can minimize the objective function, and the searched result is the shear wave acoustic velocity. Figure 10 Shows a schematic diagram of the formation shear wave acoustic velocity obtained by processing the wireline dipole acoustic logging signal in hard and soft formations using the method of the present application. In Figure 10 , the part above 3681.5 m is the hard formation, and the part below is the soft formation.

[0052] In an optional implementation manner, the training process of the fast dispersion simulation network includes: generating a preset number of theoretical simulated dispersion signals based on the dispersion equation of the borehole sound field; obtaining borehole acoustic sample parameters, and using the borehole acoustic sample parameters as the training dataset features and the theoretical simulated dispersion signals as the training dataset labels to train the neural network to obtain the fast dispersion simulation network, where the borehole acoustic sample parameters include: borehole mud acoustic velocity, borehole diameter, and formation shear wave acoustic velocity.

[0053] When performing network training, it is necessary to first prepare the dataset required for training. The dataset required for training mainly includes: borehole acoustic sample parameters and theoretical simulated dispersion signals. Among them, the theoretical simulated dispersion signals can be generated based on the dispersion equation of the borehole acoustic field. The dispersion equation of the borehole acoustic field can be referred to in formula (9): (9) Among them, k is the wave number, ω is the angular frequency, B represents the waveguide in the borehole, the influence of borehole mud and logging instruments, and F represents the relevant parameters of the elastic isotropic formation, including the longitudinal and transverse wave velocities and formation density, which vary with the radial distance r. By solving the numerical solution of the dispersion equation at each frequency ω, the wave number k of the flexural wave and / or helical wave can be obtained. The velocity is inversely proportional to the wave number k, and thus the corresponding velocity can be obtained, and then the theoretical simulated dispersion signal v sim (ω) of the flexural wave and / or helical wave can be obtained. These signals reflect the propagation characteristics of acoustic waves in the borehole acoustic field at different frequencies.

[0054] The borehole acoustic sample parameters are highly sensitive parameters of dispersion, mainly including: the borehole mud sound velocity v f , the borehole diameter R, and the formation shear wave velocity v s . These parameters have an important influence on the dispersion characteristics of the borehole acoustic field. Taking the borehole acoustic sample parameters as the features of the dataset and the theoretical simulated dispersion signals as the dataset labels (i.e., the target reference for model learning), the neural network is trained. Specifically, the training dataset is input into the neural network, and by adjusting the internal parameters of the network, the model can learn the mapping relationship between the borehole acoustic sample parameters and the theoretical simulated dispersion signals. After multiple rounds of iterative training, a dispersion fast simulation network is finally obtained. This network can quickly and accurately simulate the theoretical simulated dispersion signals under different borehole conditions, providing strong support for subsequent borehole acoustic analysis, formation property evaluation, etc. The dispersion fast simulation network is a fully connected neural network with a single hidden layer, which is used to quickly generate theoretical simulated dispersion signals and improve the fitting efficiency.

[0055] In summary, this application transforms the array waveform from the time domain to the frequency domain, and calculates the noisy borehole mode wave dispersion signal through the waveform spectrum. Secondly, a specifically trained and optimized deep learning model is used to achieve efficient data cleaning of the dispersion signal without manual interaction. Finally, through the dispersion correction technology driven by both data and model, the longitudinal wave, transverse wave, and Stoneley wave velocities in the cable and logging-while-drilling environments are automatically obtained.

[0056] By converting the array waveform signal obtained from full-wave train acoustic logging into the frequency domain for processing to obtain the velocities of various mode waves, and solving problems such as the need for manual interaction and slow processing efficiency in traditional frequency domain processing methods, it enables the processing of acoustic logging data during acquisition, achieving real-time acquisition of formation acoustic velocity. This application can be applied to the real-time acquisition of various acoustic velocities such as longitudinal waves, transverse waves, and Stoneley waves in wireline logging and logging-while-drilling scenarios, and can become a powerful tool for on-site data processing and real-time analysis and decision-making. At the same time, it overcomes the problems of inaccurate and unstable acoustic velocity processing results in the prior art, and solves the deficiencies of the time-domain acoustic velocity real-time processing method. Figure 11 FIG. shows a schematic flow chart of a borehole acoustic velocity acquisition method based on frequency domain analysis according to another embodiment of the present application, as Figure 11 shown, in the target interval, full-wave train acoustic ranging is performed, and an array waveform signal is collected by a receiver, and then the array waveform signal is subjected to fast Fourier transform processing to obtain a waveform spectrum signal. Based on a preset dispersion extraction method, a mode wave dispersion signal ( Figure 1 the noisy dispersion signal in the shown embodiment) is extracted from the waveform spectrum signal, and the above mode wave dispersion signal is subjected to intelligent noise reduction processing by using a dispersion noise reduction model to obtain an effective dispersion signal. For the effective dispersion signal, it can be determined whether fitting processing is required. If fitting processing is required, a theoretical simulated dispersion signal can be obtained by using a dispersion fast simulation network, and the effective dispersion signal and the theoretical simulated dispersion signal are subjected to fitting processing to determine the borehole mode wave acoustic velocity; if fitting processing is not required, the borehole mode wave acoustic velocity can be determined by frequency statistics.

[0057] Figure 12 FIG. shows a structural block diagram of a borehole acoustic velocity acquisition device based on frequency domain analysis according to an embodiment of the present application, as Figure 12 shown, the device includes: An acquisition module 1201, adapted to acquire an array waveform signal obtained by performing full-wave train acoustic logging on a target interval; A Fourier transform module 1202, adapted to perform Fourier transform on the array waveform signal to obtain a waveform spectrum signal; An extraction module 1203, adapted to extract a noisy dispersion signal from the waveform spectrum signal; A noise reduction processing module 1204, adapted to perform noise reduction processing on the noisy dispersion signal to obtain an effective dispersion signal; A signal analysis and processing module 1205, adapted to perform signal analysis and processing on the effective dispersion signal to obtain the borehole acoustic velocity.

[0058] Optionally, the signal analysis and processing module is further adapted to: if the effective dispersion signal is a Stoneley wave dispersion signal and / or a longitudinal wave dispersion signal, perform frequency distribution statistics on the Stoneley wave dispersion signal and / or the longitudinal wave dispersion signal, and determine the velocity corresponding to the maximum frequency as the Stoneley wave velocity and / or the longitudinal wave velocity; If the effective dispersion signal is a flexural wave dispersion signal and / or a helical wave dispersion signal, perform fitting processing on the flexural wave dispersion signal and / or the helical wave dispersion signal to obtain the shear wave velocity.

[0059] Optionally, the signal analysis and processing module is further adapted to: input the borehole acoustic parameters into a pre-trained fast dispersion simulation network to obtain the corresponding theoretical simulated dispersion signal, where the borehole acoustic parameters include: the borehole mud velocity, the borehole diameter, and the formation shear wave velocity; Determine the formation shear wave velocity when the difference between the flexural wave dispersion signal and the theoretical simulated dispersion signal is the smallest as the shear wave velocity, and / or determine the formation shear wave velocity when the difference between the helical wave dispersion signal and the theoretical simulated dispersion signal is the smallest as the shear wave velocity.

[0060] Optionally, the device further includes: a training module for the fast dispersion simulation network, adapted to generate a preset number of theoretical simulated dispersion signals based on the dispersion equation of the borehole acoustic field; Obtain the borehole acoustic sample parameters, use the borehole acoustic sample parameters as the features of the training data set and the theoretical simulated dispersion signal as the labels of the training data set to train the neural network, and obtain the fast dispersion simulation network, where the borehole acoustic sample parameters include: the borehole mud velocity, the borehole diameter, and the formation shear wave velocity.

[0061] Optionally, the noise reduction processing module is further adapted to: perform noise reduction processing on the noisy dispersion signal using a pre-trained dispersion noise reduction model to obtain an effective dispersion signal; The device further includes: a training module for the dispersion noise reduction model, adapted to obtain the noisy sample dispersion signal and the effective sample dispersion signal; Use the noisy sample dispersion signal to train the deep learning model to obtain the model output signal; Calculate the loss function between the model output signal and the effective sample dispersion signal; Based on the loss function, adjust the model parameters of the deep learning model; Iteratively execute the model training steps until the preset training termination condition is met.

[0062] Optionally, the extraction module is further adapted to: extract the noisy dispersion signal from the waveform spectrum signal using the matrix pencil method.

[0063] In summary, the present application transforms the array waveform from the time domain to the frequency domain, and calculates the noisy borehole mode wave dispersion signal through the waveform spectrum. Secondly, a specifically trained and optimized deep learning model is used to efficiently clean the dispersion signal data without manual interaction. Finally, through a dispersion correction technology driven by both data and model, the longitudinal wave, transverse wave, and Stoneley wave velocities in the cable and logging-while-drilling environments are automatically obtained.

[0064] By converting the array waveform signal obtained from full-waveform acoustic logging to frequency-domain processing to obtain the velocities of various mode waves, and solving problems such as the need for manual interaction and slow processing efficiency in traditional frequency-domain processing methods, the acoustic logging data can be processed while being collected, realizing the real-time acquisition of formation acoustic velocities. The present application can be applied to the real-time acquisition of various acoustic velocities such as longitudinal waves, transverse waves, and Stoneley waves in wireline logging and logging-while-drilling scenarios, can become a powerful tool for on-site data processing and real-time analysis and decision-making, and at the same time overcomes the problems of inaccurate and unstable acoustic velocity processing results in the prior art, and solves the deficiencies of the time-domain acoustic velocity real-time processing method.

[0065] The embodiment of the present application provides a non-volatile computer storage medium, and the computer storage medium stores at least one executable instruction or computer program, and the executable instruction or computer program can enable the processor to execute the operations corresponding to the borehole acoustic velocity acquisition method based on frequency-domain analysis in any of the above method embodiments.

[0066] The embodiment of the present application provides a computer program product, and the computer program product includes at least one executable instruction or computer program, and the executable instruction or computer program can enable the processor to execute the operations corresponding to the borehole acoustic velocity acquisition method based on frequency-domain analysis in any of the above method embodiments.

[0067] Figure 13 The structure diagram of a computing device according to an embodiment of the present application is shown, and the specific implementation of the computing device is not limited in the specific embodiment of the present application.

[0068] As Figure 13 shown, the computing device may include: a processor 1302, a communication interface 1304, a memory 1306, and a communication bus 1308.

[0069] Among them: the processor 1302, the communication interface 1304, and the memory 1306 communicate with each other through the communication bus 1308.

[0070] The communication interface 1304 is used to communicate with network elements of other devices such as clients or other servers.

[0071] A processor 1302 is configured to execute a program 1310, and specifically, can execute the relevant steps in the above embodiments of the method for obtaining borehole acoustic velocity based on frequency domain analysis.

[0072] Specifically, the program 1310 may include program code, and the program code includes computer operation instructions.

[0073] The processor 1302 may be a central processing unit (CPU), or a specific integrated circuit (ASIC) (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present application. One or more processors included in the computing device may be of the same type of processor, such as one or more CPUs; or may be of different types of processors, such as one or more CPUs and one or more ASICs.

[0074] A memory 1306 is configured to store the program 1310. The memory 1306 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk memory.

[0075] The program 1310 is specifically configured to cause the processor 1302 to execute the method for obtaining borehole acoustic velocity based on frequency domain analysis in any of the above method embodiments. For the specific implementation of each step in the program 1310, reference may be made to the corresponding steps and descriptions in the corresponding units in the above embodiments of the method for obtaining borehole acoustic velocity based on frequency domain analysis, which will not be elaborated herein. Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described devices and modules may refer to the corresponding process descriptions in the foregoing method embodiments, which will not be repeated herein.

[0076] The algorithms and displays provided herein are not inherently related to any particular computer, virtual system, or other device. Various general-purpose systems may also be used in conjunction with the teachings provided herein. The structure required to construct such systems will be apparent from the above description. In addition, the present application is not directed to any specific programming language. It should be understood that the content of the present application described herein can be implemented using various programming languages, and the description of the specific language above is for the purpose of disclosing the best mode of the present application.

[0077] In the specification provided herein, a large number of specific details are set forth. However, it can be understood that the embodiments of the present application may be practiced without these specific details. In some instances, well-known methods, structures, and technologies have not been shown in detail so as not to obscure the understanding of this specification.

[0078] Similarly, it should be understood that, in order to streamline the present disclosure and assist in understanding one or more of the various inventive aspects, in the foregoing description of the exemplary embodiments of the present application, the various features of the present application are sometimes grouped together into a single embodiment, figure, or description thereof. However, the disclosed method should not be construed as reflecting an intention that the claimed application requires more features than are expressly recited in each claim. Rather, as reflected by the claims, the inventive aspects lie in less than all the features of the single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into the detailed description, where each claim stands on its own as a separate embodiment of the present application.

[0079] Those skilled in the art can understand that the modules in the devices in the embodiments can be adaptively changed and disposed in one or more devices different from the embodiments. The modules or units or components in the embodiments can be combined into one module or unit or component, and in addition, they can be divided into multiple sub-modules or sub-units or sub-components. Except that at least some of such features and / or processes or units are mutually exclusive, any combination can be used to combine all the features disclosed in this specification (including the accompanying claims, abstract, and drawings) and all the processes or units of any method or device so disclosed. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstract, and drawings) can be replaced by an alternative feature that provides the same, equivalent, or similar purpose.

[0080] In addition, those skilled in the art can understand that although some of the embodiments herein include certain features included in other embodiments rather than other features, the combination of the features of different embodiments means that it is within the scope of the present application and forms different embodiments. For example, in the claims, any one of the claimed embodiments can be used in any combination.

[0081] The various component embodiments of the present application can be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art should understand that a microprocessor or a digital signal processor (DSP) can be used in practice to implement some or all of the functions of some or all of the components according to the embodiments of the present application. The present application can also be implemented as a device or apparatus program (for example, a computer program and a computer program product) for executing part or all of the methods described herein. Such a program implementing the present application can be stored on a computer-readable medium, or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

[0082] It should be noted that the above embodiments are illustrative of the present application rather than restrictive of the present application, and those skilled in the art can design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in the claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The present application can be implemented by means of hardware including several different elements and by means of a suitably programmed computer. In a unit claim listing several devices, several of these devices can be embodied by the same item of hardware. The use of the words first, second, and third, etc. does not denote any order. These words can be interpreted as names.

Claims

1. A method for obtaining borehole acoustic velocity based on frequency-domain analysis, characterized in that, The method includes: Obtaining an array waveform signal obtained by performing full-wave train acoustic logging on a target interval; Performing Fourier transform on the array waveform signal to obtain a waveform frequency spectrum signal; Extracting a noisy dispersion signal from the waveform frequency spectrum signal, and performing noise reduction processing on the noisy dispersion signal to obtain an effective dispersion signal; If the effective dispersion signal is a Stoneley wave dispersion signal and / or a longitudinal wave dispersion signal, then performing frequency distribution statistics on the Stoneley wave dispersion signal and / or the longitudinal wave dispersion signal, and determining the velocity corresponding to the maximum frequency as the Stoneley wave velocity and / or the longitudinal wave velocity; If the effective dispersion signal is a flexural wave dispersion signal and / or a helical wave dispersion signal, then inputting wellbore acoustic parameters into a pre-trained fast dispersion simulation network to obtain a corresponding theoretical simulated dispersion signal, where the wellbore acoustic parameters include: wellbore mud velocity, well diameter, and formation shear wave velocity; determining the formation shear wave velocity when the difference between the flexural wave dispersion signal and the theoretical simulated dispersion signal is the smallest as the shear wave velocity, and / or determining the formation shear wave velocity when the difference between the helical wave dispersion signal and the theoretical simulated dispersion signal is the smallest as the shear wave velocity.

2. The method for obtaining borehole acoustic velocity based on frequency domain analysis according to claim 1, wherein The training process of the fast dispersion simulation network includes: Generating a preset number of theoretical simulated dispersion signals based on the dispersion equation of the wellbore sound field; Obtaining wellbore acoustic sample parameters, using the wellbore acoustic sample parameters as the features of the training data set and the theoretical simulated dispersion signal as the label of the training data set to train a neural network to obtain a fast dispersion simulation network, where the wellbore acoustic sample parameters include: wellbore mud velocity, well diameter, and formation shear wave velocity.

3. The method for obtaining borehole acoustic velocity based on frequency domain analysis according to claim 1 or 2, characterized in that, The performing noise reduction processing on the noisy dispersion signal to obtain an effective dispersion signal further includes: Using a pre-trained dispersion noise reduction model to perform noise reduction processing on the noisy dispersion signal to obtain an effective dispersion signal; Wherein, the training process of the dispersion noise reduction model includes: Obtaining a noisy sample dispersion signal and an effective sample dispersion signal; Training a deep learning model using the noisy sample dispersion signal to obtain a model output signal; Calculating a loss function between the model output signal and the effective sample dispersion signal; Adjusting the model parameters of the deep learning model based on the loss function; Iteratively executing the model training step until a preset training termination condition is satisfied.

4. The method for obtaining borehole acoustic wave velocity based on frequency domain analysis according to claim 1 or 2, characterized in that The extracting a noisy dispersion signal from the waveform frequency spectrum signal further includes: Using the matrix pencil method to extract a noisy dispersion signal from the waveform frequency spectrum signal.

5. An apparatus for obtaining borehole acoustic velocity based on frequency domain analysis, characterized in that, The apparatus includes: An obtaining module, adapted to obtain an array waveform signal obtained by performing full-wave train acoustic logging on a target interval; A Fourier transform module, adapted to perform Fourier transform on the array waveform signal to obtain a waveform frequency spectrum signal; An extracting module, adapted to extract a noisy dispersion signal from the waveform frequency spectrum signal; A noise reduction processing module, adapted to perform noise reduction processing on the noisy dispersion signal to obtain an effective dispersion signal; A signal analysis and processing module is adapted to perform frequency distribution statistics on the Stoneley wave dispersion signal and / or the longitudinal wave dispersion signal if the effective dispersion signal is the Stoneley wave dispersion signal and / or the longitudinal wave dispersion signal, and determine the velocity corresponding to the maximum frequency as the Stoneley wave velocity and / or the longitudinal wave velocity; if the effective dispersion signal is the flexural wave dispersion signal and / or the helical wave dispersion signal, input the borehole acoustic parameters into a pre-trained fast dispersion simulation network to obtain the corresponding theoretical simulated dispersion signal, where the borehole acoustic parameters include: the borehole mud velocity, the borehole diameter, and the formation shear wave velocity; determine the formation shear wave velocity when the difference between the flexural wave dispersion signal and the theoretical simulated dispersion signal is the smallest as the shear wave velocity, and / or determine the formation shear wave velocity when the difference between the helical wave dispersion signal and the theoretical simulated dispersion signal is the smallest as the shear wave velocity.

6. A computing device, characterized in that, Comprising: A processor, a memory, a communication interface, and a communication bus, and the processor, the memory, and the communication interface complete communication with each other through the communication bus; The memory is used to store at least one executable instruction, and the executable instruction causes the processor to perform operations corresponding to the borehole acoustic wave velocity acquisition method based on frequency domain analysis as described in any one of claims 1-4.

7. A computer storage medium, characterized in that, At least one executable instruction is stored in the computer storage medium, and the executable instruction causes the processor to perform operations corresponding to the borehole acoustic wave velocity acquisition method based on frequency domain analysis as described in any one of claims 1-4.

8. A computer program product, characterized in that, Comprising at least one executable instruction, and the executable instruction causes the processor to perform operations corresponding to the borehole acoustic wave velocity acquisition method based on frequency domain analysis as described in any one of claims 1-4.

Images