Dispersion correction method and device for dipole acoustic logging and computing equipment
By acquiring monopole and dipole logging data, calculating shear wave velocity error, generating initial dispersion data and extracting characteristic dispersion curves, and using a dispersion forward model to match the target dispersion curve, the problem of long time consumption and insufficient accuracy in dispersion correction in existing technologies is solved, achieving efficient and accurate dispersion correction.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHINA OILFIELD SERVICES LTD
- Filing Date
- 2025-10-21
- Publication Date
- 2026-06-26
AI Technical Summary
Existing dispersion correction methods are too time-consuming and cannot extract accurate shear wave information from weak signal strata lacking low frequencies, resulting in insufficient dispersion correction accuracy.
By acquiring monopole and dipole logging data, the relative error of shear wave velocity is calculated. Initial dispersion data is generated only when the error is greater than a threshold. Characteristic dispersion curves are extracted, and the dispersion forward model is used to match the target forward dispersion curve to generate dispersion correction results.
It improves the accuracy and efficiency of dispersion correction, and is suitable for different geological environments, especially maintaining high-precision dispersion correction in weak signal formations.
Smart Images
Figure CN121386008B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of exploration data processing technology, specifically to a dispersion correction method, apparatus, and computing equipment for dipole acoustic logging. Background Technology
[0002] Dipole acoustic logging is an acoustic logging technique that uses a dipole transducer to excite acoustic waves and measure their propagation characteristics in the wellbore. Because dipole waveforms exhibit significant dispersion effects, dispersion correction is a crucial step in dipole acoustic logging data processing.
[0003] Currently, the commonly used dispersion correction method is to perform dispersion correction on all formations. However, existing dispersion correction methods are too time-consuming and cannot extract accurate shear wave information from weak signal formations lacking low frequencies, resulting in insufficient dispersion correction accuracy. Summary of the Invention
[0004] In view of the above problems, this application is made in order to provide a dispersion correction method, apparatus and computing equipment for dipole acoustic logging that overcomes or at least partially solves the above problems.
[0005] According to a first aspect of this application, a dispersion correction method for dipole acoustic logging is provided, comprising:
[0006] Acquire monopole logging data and dipole logging data of the formation at the target depth, and calculate the first shear wave velocity based on the monopole logging data and the second shear wave velocity based on the dipole logging data;
[0007] If the relative error between the second shear wave velocity and the first shear wave velocity is greater than the error threshold, then initial dispersion data is generated based on the dipole logging data.
[0008] Extract the characteristic dispersion curve from the initial dispersion data;
[0009] From the forward dispersion curves output by the dispersion forward model, determine the target forward dispersion curve that best matches the characteristic dispersion curve;
[0010] Dispersion correction results are generated based on the target forward dispersion curve.
[0011] In one optional implementation, extracting the characteristic dispersion curve from the initial dispersion data includes:
[0012] The initial dispersion data is clustered to generate multiple clusters;
[0013] Identify the longest cluster in the frequency direction from the plurality of clusters;
[0014] The feature dispersion curve is obtained by fitting the longest cluster.
[0015] In one optional implementation, fitting the longest cluster to obtain the feature dispersion curve includes:
[0016] Identify the target frequency corresponding to the maximum velocity value in the longest cluster;
[0017] The dispersion data points in the longest cluster whose frequency is greater than or equal to the target frequency are taken as feature dispersion data points;
[0018] The characteristic dispersion curve is obtained by fitting the characteristic dispersion data points.
[0019] In one alternative implementation, the dispersion forward model is trained as follows:
[0020] Training samples are constructed based on formation parameters and theoretical dispersion curves generated from dispersion equations.
[0021] A dispersion forward model is pre-constructed based on a deep neural network;
[0022] The dispersion forward model is trained using the training samples to obtain a trained dispersion forward model.
[0023] In one optional implementation, determining the target forward dispersion curve that best matches the characteristic dispersion curve from the forward dispersion curves output from the dispersion forward model includes:
[0024] Identify the target frequency band of the characteristic dispersion curve;
[0025] The least squares result of the characteristic dispersion curve and the forward dispersion curve within the target frequency band is used as the iterative objective function;
[0026] The target forward dispersion curve is iteratively searched from the forward dispersion curves based on the iterative objective function.
[0027] In one optional implementation, the iterative search for the target forward dispersion curve from the forward dispersion curve based on the objective function includes:
[0028] Determine the formation parameters for the current iteration step.
[0029] Input the formation parameters of the current iteration step into the dispersion forward model and obtain the forward dispersion curve of the current iteration step output by the dispersion forward model.
[0030] Calculate the iterative objective function value for the current iteration step based on the forward dispersion curve of the current iteration step and the characteristic dispersion curve.
[0031] Based on the value of the iterative objective function, determine whether the iteration termination condition is met; if yes, end the iteration and take the forward dispersion curve of the current iteration step as the target forward dispersion curve; if no, determine the formation parameters for the next iteration step.
[0032] According to a second aspect of this application, a dispersion correction device for dipole acoustic logging is provided, comprising:
[0033] The acquisition module is used to acquire monopole logging data and dipole logging data of the formation at the target depth, and to calculate a first shear wave velocity based on the monopole logging data and a second shear wave velocity based on the dipole logging data.
[0034] The verification module is used to determine whether the relative error between the second shear wave velocity and the first shear wave velocity is greater than the error threshold.
[0035] An initial data generation module is used to generate initial dispersion data based on the dipole logging data if the relative error is greater than an error threshold.
[0036] The feature extraction module is used to extract the feature dispersion curve from the initial dispersion data;
[0037] The forward modeling module is used to obtain the forward dispersion curves output by the dispersion forward modeling model;
[0038] The matching module is used to determine the target forward dispersion curve that best matches the characteristic dispersion curve from the forward dispersion curves output by the dispersion forward model;
[0039] The result generation module is used to generate dispersion correction results based on the target forward dispersion curve.
[0040] In one optional implementation, the feature extraction module is used to: perform clustering processing on the initial dispersion data to generate multiple clusters;
[0041] Identify the longest cluster in the frequency direction from the plurality of clusters;
[0042] The feature dispersion curve is obtained by fitting the longest cluster.
[0043] In one optional implementation, the feature extraction module is used to: identify the target frequency corresponding to the maximum velocity value in the longest cluster;
[0044] The dispersion data points in the longest cluster whose frequency is greater than or equal to the target frequency are taken as feature dispersion data points;
[0045] The characteristic dispersion curve is obtained by fitting the characteristic dispersion data points.
[0046] In one optional implementation, the forward modeling module is used to: train the dispersion forward modeling model in the following manner:
[0047] Training samples are constructed based on formation parameters and theoretical dispersion curves generated from dispersion equations.
[0048] A dispersion forward model is pre-constructed based on a deep neural network;
[0049] The dispersion forward model is trained using the training samples to obtain a trained dispersion forward model.
[0050] In one alternative implementation, the matching module is used to: identify the target frequency band of the characteristic dispersion curve;
[0051] The least squares result of the characteristic dispersion curve and the forward dispersion curve within the target frequency band is used as the iterative objective function;
[0052] The target forward dispersion curve is iteratively searched from the forward dispersion curves based on the iterative objective function.
[0053] In one alternative implementation, the matching module is used to: determine the formation parameters of the current iteration step.
[0054] Input the formation parameters of the current iteration step into the dispersion forward model and obtain the forward dispersion curve of the current iteration step output by the dispersion forward model.
[0055] Calculate the iterative objective function value for the current iteration step based on the forward dispersion curve of the current iteration step and the characteristic dispersion curve.
[0056] Based on the value of the iterative objective function, determine whether the iteration termination condition is met; if yes, end the iteration and take the forward dispersion curve of the current iteration step as the target forward dispersion curve; if no, determine the formation parameters for the next iteration step.
[0057] According to a third aspect of this application, a computing device is provided, comprising: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other through the communication bus;
[0058] The memory is used to store at least one executable instruction, which causes the processor to perform the operation corresponding to the above-described dipole acoustic logging dispersion correction method.
[0059] According to a fourth aspect of this application, a computer storage medium is provided, wherein at least one executable instruction is stored in the storage medium, the executable instruction causing a processor to perform the operation corresponding to the above-described diffusion correction method for dipole acoustic logging.
[0060] According to a fifth aspect of this application, a computer program product is provided, comprising at least one executable instruction that causes a processor to perform operations corresponding to the above-described diffusion correction method for dipole acoustic logging.
[0061] According to the dispersion correction method, apparatus, computing device, computer storage medium, and computer program product for dipole acoustic logging provided in this application, dispersion correction at the target depth is triggered only when the shear wave velocity error between monopole logging and dipole logging is large at the current target depth. This avoids the waste of computing resources caused by dispersion correction throughout the entire well section and improves the overall dispersion correction efficiency. Moreover, during the dispersion correction process, a high-quality characteristic dispersion curve that represents the true dispersion trend is extracted from the initial dispersion data. This characteristic dispersion curve is then compared with the forward dispersion curve output by the dispersion forward model to obtain the target forward dispersion curve that best matches the characteristic dispersion curve, thereby obtaining the dispersion correction result. Therefore, this scheme can be applied to different formation environments and is not limited by formation conditions, maintaining high correction accuracy even in weak signal formations. Thus, this scheme can significantly improve the dispersion correction accuracy and efficiency of dipole logging.
[0062] The above description is merely an overview of the technical solutions of the embodiments of this application. In order to better understand the technical means of the embodiments of this application and to implement them in accordance with the contents of the specification, and to make the above and other objects, features and advantages of the embodiments of this application more obvious and understandable, specific implementation methods of the embodiments of this application are described below. Attached Figure Description
[0063] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the embodiments of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:
[0064] Figure 1 A schematic flowchart of a dispersion correction method for dipole acoustic logging provided in an embodiment of this application is shown.
[0065] Figure 2 This illustration shows a schematic diagram of an initial dispersion data provided in an embodiment of this application;
[0066] Figure 3 A flowchart illustrating a method for generating a feature dispersion curve according to an embodiment of this application is shown.
[0067] Figure 4 This illustration shows a schematic diagram of the longest cluster provided in an embodiment of this application;
[0068] Figure 5This illustration shows a schematic diagram of a characteristic dispersion data point provided in an embodiment of this application;
[0069] Figure 6 A flowchart illustrating a dissipative forward model training method provided in an embodiment of this application is shown.
[0070] Figure 7 This illustration shows a structural schematic diagram of a dispersion forward model provided in an embodiment of this application;
[0071] Figure 8 A flowchart illustrating a method for determining a target forward dispersion curve according to an embodiment of this application is shown.
[0072] Figure 9 This illustration shows a schematic diagram of a target forward dispersion curve provided in an embodiment of this application;
[0073] Figure 10 This illustration shows a schematic diagram of the dispersion correction device for dipole acoustic logging provided in an embodiment of this application;
[0074] Figure 11 A schematic diagram of the structure of a computing device provided in an embodiment of this application is shown. Detailed Implementation
[0075] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
[0076] Figure 1 The diagram shows a flow chart of a dispersion correction method for dipole acoustic logging provided in an embodiment of this application.
[0077] like Figure 1 As shown, the method specifically includes the following steps:
[0078] Step S101: Obtain monopole logging data and dipole logging data of the formation at the target depth, and calculate the first shear wave velocity based on the monopole logging data and the second shear wave velocity based on the dipole logging data.
[0079] Sonic logging instruments were used to perform both monopole and dipole acoustic logging to acquire monopole and dipole logging data at the target depth formation, respectively. Specifically, the monopole logging data for the target depth formation was obtained using monopole acoustic logging at the target depth, while the dipole logging data for the target depth formation was obtained using dipole acoustic logging at the target depth.
[0080] Shear wave velocities were extracted from both monopole and dipole logging data to extract the first shear wave velocity from the monopole logging data and the second shear wave velocity from the dipole logging data.
[0081] In this application, the method for extracting shear wave velocity is not limited, and the Slowness Time Coherence (STC) method can be used. For example, the energy-focused slowness identification formula in the STC method shown in Formula 1 can be used for shear wave velocity extraction.
[0082] (Formula 1)
[0083] In formula 1, Indicates the directness of the wave slowness to be searched; Indicates the time it takes for the first wave to reach the first receiver; Indicates the energy focusing value; N represents the total number of receivers; Indicates the half length of the time window; Indicates the time-adjusted index; Indicates the distance between receivers; Indicates the time sampling interval; This represents the original signal collected by the i-th receiver.
[0084] Specifically, each direct wave slowness can be achieved. Substituting the values into Formula 1 above yields the corresponding energy focusing values. Thus, a slow speed is obtained. With energy focus value The curve is used to determine the final transverse wave slowness, and the value corresponding to the peak value (i.e. the maximum energy focusing value) is taken as the transverse wave slowness, and thus the transverse wave velocity is obtained.
[0085] Step S102: Calculate the relative error between the second transverse wave velocity and the first transverse wave velocity.
[0086] Specifically, the absolute value of the difference between the second transverse wave velocity and the first transverse wave velocity is calculated, and the relative error is obtained based on the ratio of the absolute value of the difference to the second transverse wave velocity or the first transverse wave velocity.
[0087] Step S103: Determine whether the relative error is greater than the error threshold; if so, proceed to step S104.
[0088] If the relative error between the second shear wave velocity and the first shear wave velocity is greater than the error threshold, it indicates that there is significant dispersion in the current dipole logging data. In this case, the subsequent steps S104-S107 are executed to perform dispersion correction. If the relative error between the second shear wave velocity and the first shear wave velocity is less than or equal to the error threshold, it indicates that there is no significant dispersion in the current dipole logging data. In this case, there is no need to perform dispersion correction, and the method ends.
[0089] Step S104: Generate initial dispersion data based on dipole logging data.
[0090] After determining that dispersion correction is needed, initial dispersion data is first generated based on dipole logging data. For example... Figure 2 As shown, the initial dispersion data contains a series of dispersion data points, each of which corresponds to a set of frequency-wave velocity.
[0091] This application does not limit the specific method for generating initial dispersion data based on dipole logging data in its embodiments. For example, the weighted spectral semblance (WSS) method can be used to generate initial dispersion data. The WSS method calculates coherence in the frequency domain, has a certain degree of noise suppression, and can still provide a relatively clear dispersion point distribution even under weak signal conditions, thereby ensuring the dispersion correction accuracy of subsequent weak signals.
[0092] Specifically, the frequency range and the frequency values of the dispersion data points are determined. For each frequency value, the optimal slowness corresponding to that frequency value is determined based on Formula 2, thereby obtaining the data pair of frequency value and wave velocity. When determining the optimal slowness for a frequency value, the slowness search range can be determined first. Within this slowness search range, the correlation coefficients corresponding to different slowness values at that frequency value are determined, and the slowness that maximizes the correlation coefficient is taken as the corresponding optimal slowness.
[0093] (Formula 2)
[0094] In formula 2, Represents the normalized correlation coefficient; Indicates angular frequency; The wavelength of the wavelet to be searched is represented by N; N represents the number of receivers. Indicates the first Each receiver at angular frequency The received spectral value; "*" indicates taking the complex conjugate; Indicates the distance between receivers; This represents the frequency response from the k-th signal source to the receiver array; This represents the angular frequency of the k-th signal source.
[0095] Step S105: Extract the characteristic dispersion curve from the initial dispersion data.
[0096] To avoid interference from a large amount of invalid data, this step extracts characteristic dispersion data points from the initial dispersion data, and then generates characteristic dispersion curves.
[0097] In one optional implementation, to improve the dispersion correction accuracy, the following can be specifically adopted: Figure 3 The steps shown generate the characteristic dispersion curve:
[0098] S1051 performs clustering processing on the initial dispersion data to generate multiple clusters.
[0099] The initial dispersion data contains a series of dispersion data points. Clustering these dispersion data points yields multiple clusters, each containing several dispersion data points. In this embodiment, a density-based clustering algorithm can be used to cluster the initial dispersion data, such as the density-based spatial clustering of applications with noise (DBSCAN).
[0100] S1052 identifies the longest cluster in the frequency direction from multiple clusters.
[0101] Specifically, for each cluster generated after clustering, the maximum and minimum frequency values of the dispersion data points in that cluster are determined. The length of the cluster in the frequency direction is obtained based on the difference between the maximum and minimum frequency values. The cluster corresponding to the maximum length in the frequency direction is then taken as the longest cluster, and the feature dispersion curve is subsequently fitted to this longest cluster.
[0102] like Figure 4 As shown, the cluster in the dashed box is the longest cluster identified in this step. This longest cluster is the cluster that spans the widest frequency range among the clusters obtained after clustering.
[0103] S1053 identifies the target frequency corresponding to the maximum velocity value in the longest cluster.
[0104] Although data fitting based on the longest cluster can yield the characteristic dispersion curve, some interfering dispersion data points still exist within this longest cluster. Therefore, to further improve the extraction accuracy of the characteristic dispersion curve and enhance the overall dispersion correction accuracy, after obtaining the longest cluster, step S1053-S1054 is used to further extract characteristic dispersion data points from the longest cluster to eliminate interfering dispersion data points.
[0105] Specifically, the maximum velocity value is identified from the longest cluster, the dispersion data point corresponding to the maximum velocity value is determined, and the frequency value of the corresponding dispersion data point is used as the target frequency.
[0106] S1054, the dispersion data points in the longest cluster with a frequency greater than or equal to the target frequency are taken as feature dispersion data points.
[0107] Dispersion data points with frequencies greater than or equal to the target frequency in the longest cluster are selected as feature dispersion data points, while dispersion data points with frequencies less than the target frequency in the longest cluster are removed. The extracted feature dispersion data points thus reflect the true dispersion trend and avoid interference from low-frequency data points.
[0108] like Figure 5 As shown, P is the dispersion data point corresponding to the maximum velocity in the longest similarity, and the frequency of point P is the target frequency. Therefore, the dispersion data points whose longest cluster frequency is greater than or equal to the target frequency are taken as the feature dispersion data points, i.e. Figure 5 The dispersive data points are shown within the dashed box.
[0109] S1055, the characteristic dispersion curve is obtained by fitting the characteristic dispersion data points.
[0110] The characteristic dispersion data points are fitted, and the resulting curve is the characteristic dispersion curve. This characteristic dispersion curve can reflect the true dispersion trend and significantly suppress interfering dispersion data points.
[0111] This embodiment does not limit the specific fitting algorithm. For example, a multivariate nonlinear regression algorithm can be preferred to fit the characteristic dispersion data points to obtain the characteristic dispersion curve.
[0112] Step S106: From the forward dispersion curves output by the dispersion forward model, determine the target forward dispersion curve that best matches the characteristic dispersion curve.
[0113] A dispersion forward model is pre-constructed and trained based on a deep neural network algorithm. This dispersion forward model can quickly generate corresponding dispersion curves under given formation parameters. These dispersion curves are called forward dispersion curves. Therefore, the dispersion forward model can be used to quickly obtain the forward dispersion curves and the formation parameters corresponding to each forward dispersion curve.
[0114] In one alternative implementation, the following can be used: Figure 6 The training steps shown yield the dispersive forward model:
[0115] S1061, training samples are constructed based on formation parameters and theoretical dispersion curves generated based on dispersion equations.
[0116] Different formation parameters are set, and theoretical dispersion curves corresponding to each parameter are obtained. These theoretical dispersion curves can be obtained by solving dispersion equations. Training samples are then constructed based on the formation parameters and theoretical dispersion curves, and divided into training and testing sets. Optionally, after obtaining the theoretical dispersion curves corresponding to the formation parameters, mean normalization (subtracting the overall average value from the original data value) can be performed on the data points of the formation parameters and / or theoretical dispersion curves to generate sample data.
[0117] S1062, a dispersion forward model is pre-built based on a deep neural network.
[0118] The dispersive forward model is built upon a deep neural network, with the appropriate number of layers and neurons set. For example... Figure 7 As shown, the dispersive forward model can include an input layer, a hidden layer, and an output layer.
[0119] S1063, the dispersion forward model is trained using training samples to obtain a trained dispersion forward model.
[0120] Select appropriate activation functions, loss functions, and optimization functions, and set the learning rate and number of iterations. Then, train the dispersion forward model using training samples according to the set training parameters. The iteration ends when the termination condition is met, yielding the trained dispersion forward model.
[0121] Furthermore, from the forward dispersion curves output by the dispersion forward model, the target forward dispersion curve that best matches the characteristic dispersion curve is determined.
[0122] In one optional implementation, a target frequency band of the characteristic dispersion curve is identified; the least squares result of the characteristic dispersion curve and the forward dispersion curve within the target frequency band is used as the iterative objective function; and a target forward dispersion curve is iteratively searched from the forward dispersion curves based on the iterative objective function. Specifically, the frequency range covered by the characteristic dispersion curve is called the target frequency band. Through continuous iteration, a target forward dispersion curve that minimizes the iterative objective function is searched.
[0123] Specifically, it can be adopted Figure 8 The steps shown are for determining the target forward dispersion curve:
[0124] S1064, determine the formation parameters for the current iteration step.
[0125] In the initial state, the search range of formation parameters can be determined first, and the initial values of formation parameters can be determined based on the search range.
[0126] If, in step S1067, the iteration termination condition is not met, the next iteration step is performed to update the formation parameter values. In the specific process of updating formation parameters, a fixed step size update or an algorithm such as gradient descent can be used. This embodiment does not limit the specific update method.
[0127] S1065, input the formation parameters into the dispersion forward model and obtain the forward dispersion curve of the current iteration step output by the dispersion forward model.
[0128] Input the formation parameters of the current iteration step into the dispersion forward model, and the dispersion forward model can output the corresponding forward dispersion curve. This forward dispersion curve is the theoretical dispersion curve predicted by the dispersion forward model under the formation parameters of the current iteration step.
[0129] S1066, Calculate the iterative objective function value for the current iteration step based on the forward dispersion curve and the characteristic dispersion curve of the current iteration step.
[0130] The objective function can be expressed as shown in Formula 3:
[0131] (Formula 3)
[0132] In Formula 3, L represents the objective function; n represents the number of sampling points within the target frequency band; This represents the frequency value of the i-th sampling point within the target frequency band; Represents the frequency values in the characteristic dispersion curve The corresponding speed value; Represents the frequency values in the forward dispersion curve The corresponding speed value.
[0133] Formula 3 can be used to calculate the value of the iterative objective function for the current iteration step.
[0134] S1067, determine whether the iteration termination condition is met based on the value of the iterative objective function; if yes, proceed to step S1068; if no, proceed to step S1064.
[0135] The iteration termination condition can be: the value of the iteration objective function is less than a preset threshold and / or the difference between the values of the iteration objective function of adjacent M iteration steps is less than a preset difference; or, the current iteration number reaches the maximum number threshold.
[0136] If the iteration termination condition is met, the iteration ends and step S1068 is executed; if the iteration termination condition is not met, the next iteration step is initiated and step S1064 is executed to update the formation parameters.
[0137] S1068, take the forward dispersion curve of the current iteration step as the target forward dispersion curve.
[0138] The forward dispersion curve at the end of the iteration is taken as the target forward dispersion curve, which is the forward dispersion curve that best matches the characteristic dispersion curve.
[0139] like Figure 9 As shown, L1 is the characteristic dispersion curve, and L2 is the target forward dispersion curve. L1 and L2 highly overlap in the target frequency band. Additionally, the wave velocity corresponding to the dashed line is the corrected shear wave velocity.
[0140] Step S107: Generate dispersion correction results based on the target forward dispersion curve.
[0141] Specifically, obtain the formation parameters corresponding to the target forward dispersion curve, and use these formation parameters as the dispersion correction result.
[0142] Therefore, the dispersion correction method for dipole acoustic logging provided in this application only triggers dispersion correction at the target depth when the shear wave velocity error between monopole and dipole logging is large at the current target depth. This avoids wasting computational resources caused by dispersion correction throughout the entire well section and improves the overall dispersion correction efficiency. Furthermore, during the dispersion correction process, a high-quality characteristic dispersion curve representing the true dispersion trend is extracted from the initial dispersion data. This characteristic dispersion curve is then compared with the forward dispersion curve output by the dispersion forward model to obtain the target forward dispersion curve that best matches the characteristic dispersion curve, thus yielding the dispersion correction result. Therefore, this solution is applicable to different formation environments and is not limited by formation conditions, maintaining high correction accuracy even in weak signal formations. Consequently, this solution can significantly improve the dispersion correction accuracy and efficiency of dipole logging.
[0143] Figure 10 A schematic diagram of the structure of a dispersion correction device for dipole acoustic logging provided in an embodiment of this application is shown.
[0144] like Figure 10 As shown, the device 1000 includes: an acquisition module 1010, a verification module 1020, an initial data generation module 1030, a feature extraction module 1040, a forward modeling module 1050, a matching module 1060, and a result generation module 1070.
[0145] The acquisition module 1010 is used to acquire monopole logging data and dipole logging data of the formation at the target depth, and to calculate a first shear wave velocity based on the monopole logging data and a second shear wave velocity based on the dipole logging data.
[0146] The verification module 1020 is used to determine whether the relative error between the second shear wave velocity and the first shear wave velocity is greater than the error threshold.
[0147] The initial data generation module 1030 is used to generate initial dispersion data based on the dipole logging data if the relative error is greater than the error threshold.
[0148] Feature extraction module 1040 is used to extract feature dispersion curves from the initial dispersion data;
[0149] The forward modeling module 1050 is used to obtain the forward dispersion curve output by the dispersion forward modeling model;
[0150] The matching module 1060 is used to determine the target forward dispersion curve that best matches the characteristic dispersion curve from the forward dispersion curves output by the dispersion forward model;
[0151] The result generation module 1070 is used to generate dispersion correction results based on the target forward dispersion curve.
[0152] In one optional implementation, the feature extraction module 1040 is used to: perform clustering processing on the initial dispersion data to generate multiple clusters;
[0153] Identify the longest cluster in the frequency direction from the plurality of clusters;
[0154] The feature dispersion curve is obtained by fitting the longest cluster.
[0155] In one optional implementation, the feature extraction module 1040 is used to: identify the target frequency corresponding to the maximum velocity value in the longest cluster;
[0156] The dispersion data points in the longest cluster whose frequency is greater than or equal to the target frequency are taken as feature dispersion data points;
[0157] The characteristic dispersion curve is obtained by fitting the characteristic dispersion data points.
[0158] In one optional implementation, the forward modeling module 1050 is used to: train the dispersion forward modeling model in the following manner:
[0159] Training samples are constructed based on formation parameters and theoretical dispersion curves generated from dispersion equations.
[0160] A dispersion forward model is pre-constructed based on a deep neural network;
[0161] The dispersion forward model is trained using the training samples to obtain a trained dispersion forward model.
[0162] In one alternative implementation, the matching module 1060 is used to: identify the target frequency band of the characteristic dispersion curve;
[0163] The least squares result of the characteristic dispersion curve and the forward dispersion curve within the target frequency band is used as the iterative objective function;
[0164] The target forward dispersion curve is iteratively searched from the forward dispersion curves based on the iterative objective function.
[0165] In one alternative implementation, the matching module 1060 is used to: determine the formation parameters of the current iteration step.
[0166] Input the formation parameters of the current iteration step into the dispersion forward model and obtain the forward dispersion curve of the current iteration step output by the dispersion forward model.
[0167] Calculate the iterative objective function value for the current iteration step based on the forward dispersion curve of the current iteration step and the characteristic dispersion curve.
[0168] Based on the value of the iterative objective function, determine whether the iteration termination condition is met; if yes, end the iteration and take the forward dispersion curve of the current iteration step as the target forward dispersion curve; if no, determine the formation parameters for the next iteration step.
[0169] Therefore, the dispersion correction device for dipole acoustic logging provided in this application only triggers dispersion correction at the target depth when the shear wave velocity error between monopole logging and dipole logging is large at the current target depth. This avoids wasting computational resources caused by dispersion correction throughout the entire well section and improves the overall dispersion correction efficiency. Furthermore, during the dispersion correction process, a high-quality characteristic dispersion curve representing the true dispersion trend is extracted from the initial dispersion data. This characteristic dispersion curve is then compared with the forward dispersion curve output by the dispersion forward model to obtain the target forward dispersion curve that best matches the characteristic dispersion curve, thus yielding the dispersion correction result. Therefore, this solution is applicable to different formation environments and is not limited by formation conditions, maintaining high correction accuracy even in weak signal formations. Consequently, this solution can significantly improve the dispersion correction accuracy and efficiency of dipole logging.
[0170] This application provides a non-volatile computer storage medium storing at least one executable instruction or computer program that enables a processor to perform the operation corresponding to the dispersion correction method for dipole acoustic logging in any of the above method embodiments.
[0171] This application provides a computer program product, which includes at least one executable instruction or computer program that enables a processor to perform the operation corresponding to the dispersion correction method for dipole acoustic logging in any of the above method embodiments.
[0172] Figure 11The diagram shows a structural schematic of a computing device embodiment provided in this application. The specific embodiments of this application do not limit the specific implementation of the computing device.
[0173] like Figure 11 As shown, the computing device may include: a processor 1102, a communication interface 1104, a memory 1106, and a communication bus 1108.
[0174] The processor 1102, communication interface 1104, and memory 1106 communicate with each other via communication bus 1108. Communication interface 1104 is used to communicate with other network elements, such as clients or other servers. The processor 1102 executes program 1110, specifically performing the relevant steps in the above-described embodiment of the dispersion correction method for dipole acoustic logging for computing devices.
[0175] Specifically, program 1110 may include program code that includes computer operation instructions.
[0176] The processor 1102 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application. The computing device includes one or more processors, which may be processors of the same type, such as one or more CPUs; or processors of different types, such as one or more CPUs and one or more ASICs.
[0177] Memory 1106 is used to store program 1110. Memory 1106 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage.
[0178] Specifically, program 1110 can be used to cause processor 1102 to execute the dispersion correction method for dipole acoustic logging in any of the above method embodiments. The specific implementation of each step in program 1110 can be found in the corresponding descriptions of the steps and units in the above embodiments of the dispersion correction method for dipole acoustic logging, and will not be repeated here. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the described equipment and modules can be referred to the corresponding process descriptions in the foregoing method embodiments, and will not be repeated here.
[0179] The algorithms and displays provided herein are not inherently related to any particular computer, virtual system, or other device. Various general-purpose systems can also be used in conjunction with the teachings herein. The required structure for constructing such systems is apparent from the above description. Furthermore, the embodiments of this application are not directed to any particular programming language. It should be understood that the contents of the embodiments of this application described herein can be implemented using various programming languages, and the above description of specific languages is for the purpose of disclosing the best implementation of the embodiments of this application.
[0180] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of this application may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.
[0181] Similarly, it should be understood that, in order to streamline this disclosure and aid in understanding one or more of the various inventive aspects, features of the embodiments of this application are sometimes grouped together in a single embodiment, figure, or description thereof in the foregoing description of exemplary embodiments of the present application. However, this approach to disclosure should not be construed as reflecting an intention that the claimed embodiments of the present application require more features than expressly recited in each claim. Rather, as reflected in the claims, inventive aspects lie in fewer than all features of a single foregoing disclosed embodiment. Therefore, the claims following the detailed description are hereby expressly incorporated into that detailed description, wherein each claim itself is a separate embodiment of the present application.
[0182] Those skilled in the art will understand that modules in the device of the embodiments can be adaptively changed and placed in one or more devices different from that embodiment. Modules, units, or components in the embodiments can be combined into a single module, unit, or component, and further, they can be divided into multiple sub-modules, sub-units, or sub-components. Except where at least some of such features and / or processes or units are mutually exclusive, any combination can be used to combine all features disclosed in this specification (including the accompanying claims, abstract, and drawings) and all processes or units of any method or device so disclosed. Unless expressly stated otherwise, each feature disclosed in this specification (including the accompanying claims, abstract, and drawings) may be replaced by an alternative feature that serves the same, equivalent, or similar purpose.
[0183] Furthermore, those skilled in the art will understand that although some embodiments described herein include certain features but not others included in other embodiments, combinations of features from different embodiments are meant to be within the scope of the embodiments of this application and form different embodiments. For example, in the following claims, any one of the claimed embodiments can be used in any combination.
[0184] The various component embodiments of this application can be implemented in hardware, or as software modules running on one or more processors, or a combination thereof. Those skilled in the art will understand that microprocessors or digital signal processors (DSPs) 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 this application. The embodiments of this application can also be implemented as device or apparatus programs (e.g., computer programs and computer program products) for performing part or all of the methods described herein. Such programs implementing the embodiments of this 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, provided on a carrier signal, or provided in any other form.
[0185] It should be noted that the above embodiments are illustrative of the embodiments of this application and not limiting of the embodiments of this application, and those skilled in the art can devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. Embodiments of this application can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names.
Claims
1. A dispersion correction method for dipole acoustic logging, characterized in that, include: Acquire monopole logging data and dipole logging data of the formation at the target depth, and calculate the first shear wave velocity based on the monopole logging data and the second shear wave velocity based on the dipole logging data; If the relative error between the second shear wave velocity and the first shear wave velocity is greater than the error threshold, then initial dispersion data is generated based on the dipole logging data. Feature dispersion curves are extracted from the initial dispersion data; wherein, the initial dispersion data is clustered to generate multiple clusters; the longest cluster in the frequency direction is identified from the multiple clusters; the target frequency corresponding to the maximum velocity value in the longest cluster is identified; dispersion data points in the longest cluster with frequencies greater than or equal to the target frequency are taken as feature dispersion data points; and the feature dispersion data points are fitted to obtain the feature dispersion curves. From the forward dispersion curves output by the dispersion forward model, determine the target forward dispersion curve that best matches the characteristic dispersion curve; Dispersion correction results are generated based on the target forward dispersion curve.
2. The method according to claim 1, characterized in that, The dispersion forward model is obtained by training in the following manner: Training samples are constructed based on formation parameters and theoretical dispersion curves generated from dispersion equations. A dispersion forward model is pre-constructed based on a deep neural network; The dispersion forward model is trained using the training samples to obtain a trained dispersion forward model.
3. The method according to claim 1, characterized in that, The process of determining the target forward dispersion curve that best matches the characteristic dispersion curve from the forward dispersion curves output from the dispersion forward model includes: Identify the target frequency band of the characteristic dispersion curve; The least squares result of the characteristic dispersion curve and the forward dispersion curve within the target frequency band is used as the iterative objective function; The target forward dispersion curve is iteratively searched from the forward dispersion curves based on the iterative objective function.
4. The method according to claim 3, characterized in that, The iterative search for the target forward dispersion curve based on the objective function includes: Determine the formation parameters for the current iteration step; Input the formation parameters of the current iteration step into the dispersion forward model and obtain the forward dispersion curve of the current iteration step output by the dispersion forward model; Calculate the iterative objective function value for the current iteration step based on the forward dispersion curve of the current iteration step and the characteristic dispersion curve. Based on the value of the iterative objective function, determine whether the iteration termination condition is met; if yes, end the iteration and take the forward dispersion curve of the current iteration step as the target forward dispersion curve; if no, determine the formation parameters for the next iteration step.
5. A dispersion correction device for dipole acoustic logging, characterized in that, include: The acquisition module is used to acquire monopole logging data and dipole logging data of the formation at the target depth, and to calculate a first shear wave velocity based on the monopole logging data and a second shear wave velocity based on the dipole logging data. The verification module is used to determine whether the relative error between the second shear wave velocity and the first shear wave velocity is greater than the error threshold. An initial data generation module is used to generate initial dispersion data based on the dipole logging data if the relative error is greater than an error threshold. A feature extraction module is used to extract a feature dispersion curve from the initial dispersion data; wherein, the initial dispersion data is clustered to generate multiple clusters; the longest cluster in the frequency direction is identified from the multiple clusters; the target frequency corresponding to the maximum velocity value in the longest cluster is identified; dispersion data points in the longest cluster with frequencies greater than or equal to the target frequency are used as feature dispersion data points; and the feature dispersion curve is obtained by fitting the feature dispersion data points. The forward modeling module is used to obtain the forward dispersion curves output by the dispersion forward modeling model; The matching module is used to determine the target forward dispersion curve that best matches the characteristic dispersion curve from the forward dispersion curves output by the dispersion forward model; The result generation module is used to generate dispersion correction results based on the target forward dispersion curve.
6. A computing device, characterized in that, include: The processor, memory, communication interface, and communication bus are provided, wherein the processor, memory, and communication interface communicate with each other via the communication bus. The memory is used to store at least one executable instruction that causes the processor to perform the operation corresponding to the dispersion correction method of dipole acoustic logging as described in any one of claims 1-4.
7. A computer storage medium, characterized in that, The storage medium stores at least one executable instruction that causes the processor to perform the operation corresponding to the dispersion correction method for dipole acoustic logging as described in any one of claims 1-4.
8. A computer program product, characterized in that, It includes at least one executable instruction that causes the processor to perform the operation corresponding to the dispersion correction method for dipole acoustic logging as described in any one of claims 1-4.