Stratum velocity parameter prediction method and device for rock physical model frequency correction

Abstract

The invention provides a stratum velocity parameter prediction method and device for rock physical model frequency correction, and the method comprises the steps: correcting acoustic logging data according to the velocity frequency dispersion characteristics of a cross-frequency-band rock physical model, and enabling the acoustic velocity data of a logging frequency band (103-104 Hz) to be matched with the acoustic velocity data of a seismic frequency band (101-102 Hz). In actual seismic data, similar reservoirs generally have similar seismic waveform characteristics, through cross-band frequency dispersion correction, a high-frequency logging curve and low-frequency seismic wave characteristics are matched and associated, and a foundation is laid for subsequent multi-band velocity inversion. According to the method, the thought of'frequency division inversion and frequency-by-frequency progressive 'is adopted, the deep learning technology, the traditional constraint sparse pulse inversion technology, the waveform phase control inversion technology and the like are comprehensively utilized, low-frequency information, medium-frequency information and high-frequency information are accurately predicted in a targeted mode, the three technologies are organically combined and connected, and finally high-resolution speed inversion is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of oil and gas resource exploration and development, and particularly relates to a method and device for predicting formation velocity parameters with frequency correction of a rock physics model. Background Art

[0002] With the increasing complexity of geological problems, the accuracy requirements for velocity results are also increasing day by day. Accurate velocity is crucial for the results of seismic exploration. For example, in seismic interpretation, only by obtaining an accurate velocity field can the depth, dip, and trend of the target layer be accurately determined, the reliability of well location be determined, time-depth conversion be performed, accurate geological maps be provided for oil and gas exploration and development, and the properties of reservoirs and fluids be further determined.

[0003] According to the different stages of velocity inversion in the oil and gas exploration process and the different ultimate purposes of inversion, it can be mainly divided into: stacking velocity analysis, tomography inversion based on ray tracing, full waveform inversion, post-stack elastic parameter inversion, and pre-stack elastic parameter inversion. First is stacking velocity analysis, which is based on the assumption of a horizontally layered medium. The key steps of stacking velocity analysis include generating a convergent velocity spectrum and velocity picking. The convergence degree of the velocity spectrum determines the accuracy of velocity picking and is the key to the success or failure of velocity analysis. Then is tomography velocity inversion. A high-precision velocity model can be established from seismic observation data, and this method does not have the assumptions of a horizontally layered medium and lateral velocity invariance, and is suitable for complex geological conditions, but still needs to solve problems such as uneven ray coverage, observation data errors, and large sparse matrix inversion. Next is full waveform inversion. Full waveform inversion uses the full wavefield information of seismic waves and gradually approaches the true model, but it depends on an accurate low-frequency initial model and is computationally intensive and time-consuming. Then there is post-stack inversion, mainly referring to post-stack wave impedance inversion. Wave impedance inversion combines seismic, logging, and geological data, makes full use of the characteristics that logging data has a high vertical resolution and seismic data has a high horizontal resolution, and converts seismic data into wave impedance. Finally is pre-stack elastic parameter inversion. Compared with post-stack inversion, pre-stack inversion, in addition to using the rich information contained in pre-stack seismic data, in addition to inverting longitudinal wave information, can also estimate formation shear wave, rock modulus, fluid-sensitive parameters, physical property parameters, anisotropic parameters, absorption parameters, and even density and other information. Most of them are studied based on various approximate formulas. The simplicity of the approximate formulas is beyond doubt, but various assumptions and angular limitations will be faced when using the approximate formulas, which will make large-angle or large-offset data unable to be effectively utilized.

[0004] In the field of petrophysics, studying the velocity dispersion of seismic waves is an important topic. These phenomena not only provide a key theoretical basis for reservoir and fluid prediction in the frequency domain but also play an important role in solving the data matching difficulties among different geophysical measurement methods (such as seismic, logging, drilling, petrophysical core observation, etc.). The actual formation is viscoelastic, so the velocity dispersion phenomenon generated when seismic waves propagate in it is widespread, only the scale of velocity dispersion in different formations varies.

[0005] Rock matrix or dry layers usually do not exhibit obvious velocity dispersion and attenuation phenomena, but in reservoirs, due to the presence of fluids in the pores, this phenomenon becomes very obvious. When seismic waves encounter fluids, their propagation characteristics change significantly, which provides a key theoretical basis for detecting reservoirs and identifying fluids in the frequency domain. In addition, when using geophysical means to explore and predict the distribution of oil and gas reservoirs, the joint application of geophysical data in different frequency bands (ground seismic is 101 - 102 Hz, acoustic logging is 103 - 104 Hz, laboratory ultrasonic measurement exceeds 105 Hz) involves mutual matching problems, such as well-seismic calibration. This further emphasizes that we must attach importance to the study of velocity dispersion and attenuation phenomena to address the challenges of data matching. By analyzing and studying the velocities corresponding to different frequency bands, the velocity values corresponding to the target reservoir are further corrected. Summary of the Invention

[0006] Based on this, it is necessary to provide a method and device for predicting formation velocity parameters with frequency correction of petrophysical models in view of the above technical problems.

[0007] A method for predicting formation velocity parameters with frequency correction of petrophysical models includes:

[0008] Obtain core data, logging data, and seismic data;

[0009] Conduct modeling analysis on the core data to obtain velocity dispersion characteristics;

[0010] Perform dispersion correction on the logging data through the velocity dispersion characteristics to obtain corrected logging data;

[0011] Perform velocity inversion on the seismic data and the corrected logging data to obtain low-frequency velocity, medium-frequency velocity, and high-frequency velocity;

[0012] Use the velocity dispersion characteristics to correct the low-frequency velocity, the medium-frequency velocity, and the high-frequency velocity respectively;

[0013] Fuse the corrected low-frequency velocity, medium-frequency velocity, and high-frequency velocity to obtain the velocity inversion result.

[0014] In one embodiment, the steps of obtaining core data, logging data, and seismic data include:

[0015] Obtain core data and seismic data;

[0016] Obtain logging and drilling data;

[0017] Preprocess the logging and drilling data to obtain the logging data.

[0018] In one embodiment, the steps of preprocessing the logging and drilling data to obtain the logging data include:

[0019] Perform data outlier rejection, data sensitive parameter crossplotting, and data parameter standardization on the logging and drilling data.

[0020] In one embodiment, the steps of performing modeling analysis on the core data to obtain velocity dispersion characteristics include:

[0021] Perform rock physics modeling on the core data, analyze and compare different models to obtain velocity dispersion characteristics.

[0022] In one embodiment, the steps of performing velocity inversion on the seismic data and the logging calibration data to obtain low-frequency velocity, medium-frequency velocity, and high-frequency velocity include:

[0023] Perform velocity inversion on the seismic data and the logging calibration data to obtain medium-frequency velocity and high-frequency velocity;

[0024] Use a deep learning network to perform low-frequency velocity inversion on the seismic data and the logging calibration data to obtain low-frequency velocity.

[0025] In one embodiment, the steps of performing velocity inversion on the seismic data and the logging calibration data to obtain low-frequency velocity, medium-frequency velocity, and high-frequency velocity include:

[0026] Perform velocity inversion on the seismic data and the logging calibration data to obtain low-frequency velocity and high-frequency velocity;

[0027] Use constrained sparse pulse inversion to perform velocity inversion in the seismic frequency band on the seismic data and the logging calibration data to obtain medium-frequency velocity.

[0028] In one embodiment, the steps of performing velocity inversion on the seismic data and the logging calibration data to obtain low-frequency velocity, medium-frequency velocity, and high-frequency velocity include:

[0029] Perform velocity inversion on the seismic data and the logging calibration data to obtain low-frequency velocity and medium-frequency velocity;

[0030] For the seismic data and well logging correction data, seismic waveform phase-controlled inversion is used to perform velocity inversion in the high-frequency band to obtain high-frequency velocities.

[0031] A device for predicting formation velocity parameters with rock physics model frequency correction, comprising:

[0032] A data acquisition module, configured to acquire core data, well logging data, and seismic data;

[0033] An analysis module, configured to perform modeling analysis on the core data to obtain velocity dispersion characteristics;

[0034] A data correction module, configured to perform dispersion correction on the well logging data through the velocity dispersion characteristics to obtain well logging correction data;

[0035] A frequency-divided velocity acquisition module, configured to perform velocity inversion on the seismic data and the well logging correction data to obtain low-frequency velocities, medium-frequency velocities, and high-frequency velocities;

[0036] A velocity correction module, configured to use the velocity dispersion characteristics to correct the low-frequency velocity, the medium-frequency velocity, and the high-frequency velocity respectively;

[0037] A result acquisition module, configured to fuse the corrected low-frequency velocity, medium-frequency velocity, and high-frequency velocity to obtain a velocity inversion result.

[0038] A computer device, comprising a memory and a processor, the memory storing a computer program, wherein when the processor executes the computer program, the following steps are implemented:

[0039] Acquire core data, well logging data, and seismic data;

[0040] Perform modeling analysis on the core data to obtain velocity dispersion characteristics;

[0041] Perform dispersion correction on the well logging data through the velocity dispersion characteristics to obtain well logging correction data;

[0042] Perform velocity inversion on the seismic data and the well logging correction data to obtain low-frequency velocities, medium-frequency velocities, and high-frequency velocities;

[0043] Use the velocity dispersion characteristics to correct the low-frequency velocity, the medium-frequency velocity, and the high-frequency velocity respectively;

[0044] Fuse the corrected low-frequency velocity, medium-frequency velocity, and high-frequency velocity to obtain a velocity inversion result.

[0045] A computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0046] Obtain core data, logging data, and seismic data;

[0047] Perform modeling analysis on the core data to obtain velocity dispersion characteristics;

[0048] Perform dispersion correction on the logging data through the velocity dispersion characteristics to obtain corrected logging data;

[0049] Perform velocity inversion on the seismic data and the corrected logging data to obtain low-frequency velocity, medium-frequency velocity, and high-frequency velocity;

[0050] Utilize the velocity dispersion characteristics to correct the low-frequency velocity, the medium-frequency velocity, and the high-frequency velocity respectively;

[0051] Fuse the corrected low-frequency velocity, medium-frequency velocity, and high-frequency velocity to obtain the velocity inversion result.

[0052] The above-mentioned method and device for predicting formation velocity parameters with frequency correction of rock physics models, based on a reasonable cross-frequency rock physics model, perform dispersion correction on logging data, solve the problem of frequency scale mismatch between high-frequency logging data and low-frequency seismic data, and lay a foundation for subsequent velocity inversion. Compared with velocity inversion using a single method, the present invention improves the accuracy of velocity inversion, reduces the non-uniqueness, and provides more theoretical basis for reservoir prediction and fluid identification from the perspective of frequency division inversion.

[0053] The method and device for predicting formation velocity parameters with frequency correction of rock physics models provided by this application combine the applications of multiple disciplines such as seismology, logging, and rock physics. First, preprocess and analyze multi-source data. Based on the existing core data of reservoir sections, conduct rock physics experimental analysis in the laboratory to construct cross-frequency velocity dispersion characteristics, that is, the numerical change law of velocity corresponding to different frequencies. Match and correct the high-frequency logging data and the low-frequency seismic data to the same frequency scale, and use them as the basis for subsequent cross-frequency velocity inversion. Then, perform cross-frequency velocity inversion and correction. Taking the idea of "frequency division inversion and progressive frequency by frequency", use deep learning technology, constrained sparse pulse inversion technology, and seismic waveform phased inversion technology respectively to accurately invert and correct the low-frequency, medium-frequency, and high-frequency velocities. Finally, fuse the velocities of the three frequency bands by frequency division to obtain a high-resolution velocity result.

[0054] The above method overcomes the following main problems existing in the existing velocity inversion methods: 1. The logging data in the high-frequency band and the seismic data in the low-frequency band cannot be matched in the frequency scale, thus affecting the accuracy of velocity inversion; 2. The seismic data dominates, while information such as logging and geology is not organically incorporated into the inversion process, resulting in the problem of "two separate layers" between the seismic and other source information. Based on the high-resolution velocity inversion results obtained by final frequency fusion, other elastic parameters and physical properties parameters can be further converted. To judge the reservoir distribution and fluid properties, and provide a basis for the design of subsequent drilling construction plans to ensure efficient, safe and economic drilling. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is a schematic flow chart of a method for predicting formation velocity parameters with frequency correction of a rock physics model in an embodiment;

[0056] Figure 2 It is a schematic diagram of the implementation process of a method for predicting formation velocity parameters with frequency correction of a rock physics model in an embodiment;

[0057] Figure 3 It is a cross-band velocity inversion and frequency division diagram in an embodiment;

[0058] Figure 4 It is a velocity model data diagram in an embodiment;

[0059] Figure 5 It is a synthetic seismic record data diagram in an embodiment;

[0060] Figure 6 It is a low-frequency velocity inversion result diagram in an embodiment;

[0061] Figure 7 It is a mid-frequency velocity inversion result diagram in an embodiment;

[0062] Figure 8 It is a high-frequency velocity inversion result diagram in an embodiment;

[0063] Figure 9 It is a frequency division fusion high-resolution velocity inversion result diagram in an embodiment;

[0064] Figure 10 It is an amplitude spectrum diagram of the normalized multi-channel reflection coefficients in the 0-120 Hz range derived from a velocity model in an embodiment;

[0065] Figure 11 It is an amplitude spectrum diagram of the normalized multi-channel reflection coefficients in the 120-300 Hz range derived from a velocity model in an embodiment;

[0066] Figure 12 It is a schematic flow chart of a device for predicting formation velocity parameters with frequency correction of a rock physics model in an embodiment;

[0067] Figure 13 It is the internal structure diagram of a computer device in an embodiment. Detailed implementation manners

[0068] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0069] Embodiment 1

[0070] In this embodiment, as Figure 1 shown, a method for predicting formation velocity parameters by frequency correction of a rock physics model includes:

[0071] Step 110, obtaining core data, logging data and seismic data.

[0072] In one embodiment, the step of obtaining core data, logging data and seismic data includes: obtaining core data and seismic data; obtaining logging and drilling data; preprocessing the logging and drilling data to obtain the logging data.

[0073] In one embodiment, the step of preprocessing the logging and drilling data includes: removing data outliers, performing crossplot of data sensitive parameters, and standardizing data parameters on the logging and drilling data. Specifically, the logging and drilling data are multi-source data, that is, the logging and drilling data are various data measured during logging and drilling, including acoustic logging data.

[0074] Step 120, performing modeling analysis on the core data to obtain velocity dispersion characteristics.

[0075] In one embodiment, performing rock physics modeling on the core data, and establishing different velocity dispersion characteristics across frequency bands by analyzing and comparing different models. The models include Biot model, squirt flow model, BISQ model, patchy saturation model, double-porosity model, fracture-porosity microstructure model, and partial saturation-multiple pore type fluid pressure change induced velocity dispersion model.

[0076] In this embodiment, the velocity dispersion characteristics are the velocity dispersion characteristics of a cross-frequency band rock physics model.

[0077] Step 130, performing dispersion correction on the logging data through the velocity dispersion characteristics to obtain corrected logging data.

[0078] In one embodiment, the acoustic logging data in the logging data is corrected according to the velocity dispersion characteristics of the cross-frequency band rock physics model, so that the acoustic wave logging frequency band (10 3 -10 4 Hz) matches the acoustic wave velocity data in the seismic frequency band (10 1 -10 2 Hz).

[0079] Step 140, perform velocity inversion on the seismic data and the logging correction data to obtain low-frequency velocity, medium-frequency velocity, and high-frequency velocity.

[0080] In one embodiment, the step of performing velocity inversion on the seismic data and the logging correction data to obtain low-frequency velocity, medium-frequency velocity, and high-frequency velocity includes:

[0081] As Figure 6 shown, perform low-frequency velocity inversion on the seismic data and the logging correction data using a deep learning network to obtain low-frequency velocity.

[0082] In this embodiment, according to the neural network training rule, during the process of training and fitting data, following the frequency principle, usually the low-frequency trend is learned first, and then gradually extended to high-frequency details. For most signals, such as seismic wave propagation, the amplitude intensity has the characteristic of gradually decaying with the increase of frequency. Due to the advantages of the above neural network and the ability of deep learning to learn and represent complex non-linear relationships, a data-driven artificial intelligence method is used for low-frequency velocity inversion. Use a deep learning network to directly establish a non-linear relationship between seismic data and low-frequency velocity, and discard the interference brought by medium and high-frequency information when mapping full-band information. Its objective function can be described as the following calculation formula:

[0083] G * =m(M, G θ (D))

[0084] In the above calculation formula: M is the velocity, marked as the output parameter of the deep learning network, that is, the training set label, and M is the logging velocity that has been corrected by the cross-frequency band rock physics model velocity dispersion; D is the seismic data, marked as the input parameter of the deep learning network, that is, the training set data; G* is the low-frequency velocity inversion model updated by the deep learning network training; θ is the network parameter to be updated during the learning process; m(·) is the distance metric, used to minimize the difference between G θ (D) and M. After a certain number of iterative trainings, the learned optimal low-frequency velocity inversion model can be directly promoted to map the unpredicted seismic data to low-frequency velocity. The calculation formula is as follows:

[0085] M * =G * (D* )

[0086] In the above calculation formula: D* is the seismic data to be predicted that has not participated in training; M* is the inversion result of low-frequency velocity by artificial intelligence.

[0087] In one embodiment, the step of performing velocity inversion on the seismic data and the well logging correction data to obtain low-frequency velocity, medium-frequency velocity, and high-frequency velocity includes:

[0088] As Figure 7 shown, for the seismic data and well logging correction data, use constrained sparse pulse inversion to perform velocity inversion in the seismic frequency band to obtain medium-frequency velocity.

[0089] In this embodiment, constrained sparse pulse inversion is selected to predict the medium-frequency component. Constrained sparse pulse inversion is a recursive inversion method based on sparse pulse deconvolution. It is based on the assumption that the reflection coefficients of the formation are sparsely distributed. Then, the reflection coefficients are extracted from the original seismic records, convolved with the given seismic wavelet to generate synthetic seismic data, and the residual between it and the original seismic data is continuously compared. Based on this, the reflection coefficients are continuously modified, and the most approximate solution is obtained by iteration. Constrained sparse pulse inversion can use existing seismic interpretation horizons, well logging data, etc. as constraints, so as to further control the trend and amplitude range of the inversion results. Here, the low-frequency velocity obtained by generalization through the deep learning network model in the previous step of low-frequency velocity inversion is used as the initial model in this step and used as the constraint term of constrained sparse pulse inversion. The calculation formula of the objective function of constrained sparse pulse inversion is as follows:

[0090] F = Σ|R i | p + λ q Σ|(D i - S i )| q + α 2 Σ(T i - Z i ) 2

[0091] In the above calculation formula: F is the objective function; R i is the reflection coefficient; p is the reflection coefficient factor, with a value of 2; D i is the seismic record; S i is the synthetic seismic record; (D i - S i ) is the seismic record residual; q is the seismic residual factor, with a value of 2; T i is the initial model of low-frequency wave impedance; Z iis the wave impedance sampling between the maximum and minimum wave impedances of the well constraint; where the wave impedance is the product of velocity and density, and according to the corresponding relationship between velocity and density in the known logging data, the velocity data can be obtained from the wave impedance data of the constrained sparse pulse inversion. λ and α are weighting factors, and λ represents the matching degree between the actual seismic record and the synthetic seismic record; when λ is large and α is small, the goal is to pursue a smaller residual, resulting in a low signal-to-noise ratio of the final inversion result and a relatively small proportion of low-frequency information; when λ is small and α is large, it is sparser, resulting in a low-resolution final inversion result and obvious lack of high-frequency details.

[0092] The correct selection of the seismic wavelet, the correct selection of the number of constrained sparse pulses, the accurate establishment of the low-frequency initial model, etc. are all the keys to the accuracy of the sparse pulse inversion. The advantage of the constrained sparse pulse inversion is that it retains the basic characteristics of seismic reflection.

[0093] In one embodiment, the step of performing velocity inversion on the seismic data and the well logging correction data to obtain low-frequency velocity, medium-frequency velocity, and high-frequency velocity includes:

[0094] As Figure 8 shown, for the seismic data and the well logging correction data, seismic waveform phase-controlled inversion is used to perform velocity inversion in the high-frequency band to obtain high-frequency velocity.

[0095] In this embodiment, the conventional inversion only uses well logging data interpolation to obtain a low-frequency initial model, and combines the low-frequency and medium-frequency through inversion only in the seismic frequency band, and the high-frequency components cannot achieve the desired effect. The seismic waveform phase-controlled inversion adopted in this application is a new method developed on the basis of traditional geostatistics. Traditional geostatistical inversion is based on the variogram in the spatial domain and cannot reflect the phase change characteristics. The basic principle of the seismic waveform phase-controlled inversion method is that the change of the seismic waveform reflects the spatial change of the sedimentary environment and lithologic combination. Using the basic principles of sedimentology, the lateral change of the seismic waveform is fully utilized to reflect the phase change characteristics of the reservoir space, and then the thin-layer high-frequency characteristics of the longitudinal lithology of the reservoir are analyzed, reflecting the phase control, that is, the idea of seismic phase control. First, the seismic data is classified and analyzed according to sedimentary characteristics, then within the range of each type of waveform, the sand ratio of sedimentary characteristics and the variogram in the longitudinal and lateral directions are statistically analyzed, and finally the Markov chain Monte Carlo stochastic simulation algorithm is used to obtain multiple equally probable realizations. This method has low requirements for the distribution of wells, selects different statistical parameters for different waveform phase belts, better reflects the constraints of the sedimentary environment, can improve the longitudinal and lateral resolution of the inversion at the same time, and makes the distribution boundary of the sandstone clear and the shape conform to geological laws.

[0096] The method of singular value decomposition is used to establish the mapping relationship between the seismic waveform and the well curve, and the matrix singular decomposition calculation formula is as follows:

[0097]

[0098] In the above calculation formula: C is an n×m order matrix of seismic waveform parameters and well curves; D is seismic data; W is well curve attributes; ∑i is a non - negative real - number diagonal matrix, and the elements on the diagonal are singular values. The main characteristics of matrix C can be completely represented by the singular vectors corresponding to the first r non - zero singular values.

[0099] Phased variograms mainly include horizontal variograms and vertical variograms, which affect the horizontal continuity and vertical thickness of reservoirs respectively. To adapt to complex subsurface structures and geological conditions, the variogram with the following calculation formula is adopted here:

[0100] F i (h) = α(e)F i (h) e +α(g)F i (h) g

[0101] In the formula: i is the number of different regions; h is the lag distance; F i (h) e is an exponential variogram; F i (h) g is a Gaussian variogram; α(e) is the weight coefficient of the exponential variogram; α(g) is the weight coefficient of the Gaussian variogram, and the sum of the two weight coefficients is 1.

[0102] Step 150: Use the velocity dispersion characteristics to correct the low - frequency velocity, the medium - frequency velocity, and the high - frequency velocity respectively.

[0103] In one embodiment, by performing cross - frequency - band extraction on the velocity dispersion characteristics, the low - frequency velocity dispersion characteristics, the medium - frequency velocity dispersion characteristics, and the high - frequency velocity dispersion characteristics are extracted respectively. Use the low - frequency velocity dispersion characteristics to correct the low - frequency velocity; use the medium - frequency velocity dispersion characteristics to correct the medium - frequency velocity; use the high - frequency velocity dispersion characteristics to correct the high - frequency velocity.

[0104] Step 160: Fuse the corrected low - frequency velocity, the medium - frequency velocity, and the high - frequency velocity to obtain the velocity inversion result.

[0105] In one embodiment, as Figure 9 shown, by fusing the low - frequency velocity, the medium - frequency velocity, and the high - frequency velocity, a high - resolution velocity inversion result can be obtained. The high - resolution velocity inversion result can be further converted to obtain other elastic parameters and physical property parameters, and provide a basis for judging reservoir distribution, fluid properties, and for the design of subsequent drilling construction plans, ensuring efficient, safe, and economic drilling.

[0106] It should be understood that although Figure 1 each step in the flowchart is shown in sequence according to the indication of the arrow, these steps are not necessarily executed in sequence according to the order indicated by the arrow. Unless there is a clear description in this article, there is no strict order limit for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 1 at least a part of the steps in

[0107] Example Two

[0108] In this example, as Figure 2 shown, a method for predicting formation velocity parameters for frequency correction of a rock physics model is provided, including:

[0109] Step 1, obtaining core data, logging data, and seismic data.

[0110] In one embodiment, the step of obtaining core data, logging data, and seismic data includes: obtaining core data and seismic data; obtaining logging and drilling data; preprocessing the logging and drilling data to obtain the logging data.

[0111] In one embodiment, the step of preprocessing the logging and drilling data includes: removing data outliers, performing crossplot of data sensitive parameters, and standardizing data parameters on the logging and drilling data.

[0112] In one embodiment, the logging data includes acoustic logging data.

[0113] Step 2, performing modeling analysis on the core data to obtain velocity dispersion characteristics.

[0114] In one embodiment, performing rock physics modeling in the laboratory on the core data, analyzing and comparing different models, and establishing velocity dispersion characteristics across different frequency bands.

[0115] Step 3, performing dispersion correction on the logging data through the velocity dispersion characteristics to obtain logging corrected data.

[0116] In one embodiment, as Figure 4 and Figure 5As shown, the high-frequency well logging data is corrected to the seismic frequency band components using the velocity dispersion characteristics to obtain well logging correction data. The well logging correction data is used for subsequent data processing steps such as well-seismic calibration and dataset production. Specifically, the acoustic well logging data in the well logging data is corrected according to the velocity dispersion characteristics of the cross-frequency band rock physics model, so that the acoustic wave velocity data in the acoustic well logging frequency band (10 3 -10 4 Hz) matches the acoustic wave velocity data in the seismic frequency band (10 1 -10 2 Hz).

[0117] Step Four: Perform velocity inversion on the seismic data and the well logging correction data to obtain low-frequency velocity, medium-frequency velocity, and high-frequency velocity.

[0118] In one embodiment, the step of performing velocity inversion on the seismic data and the well logging correction data to obtain low-frequency velocity, medium-frequency velocity, and high-frequency velocity includes:

[0119] Perform low-frequency velocity inversion on the seismic data and well logging correction data using a deep learning network to obtain low-frequency velocity.

[0120] Perform velocity inversion in the seismic frequency band on the seismic data and well logging correction data using constrained sparse pulse inversion to obtain medium-frequency velocity.

[0121] Perform high-frequency band velocity inversion on the seismic data and well logging correction data using seismic waveform phase-controlled inversion to obtain high-frequency velocity.

[0122] In one embodiment, as Figure 3 shown, the frequency range of the low-frequency velocity is 0 - 10 Hz, and it is obtained by means of deep learning inversion. The frequency range of the medium-frequency velocity is 10 - 100 Hz, and it is obtained by means of constrained sparse pulse inversion. The frequency range of the high-frequency velocity is 100 - 300 Hz, and it is obtained by means of seismic waveform phase-controlled inversion.

[0123] Step Five: Use the velocity dispersion characteristics to correct the low-frequency velocity, the medium-frequency velocity, and the high-frequency velocity respectively.

[0124] In one embodiment, when studying velocity dispersion in an existing model or a constructed model, the high and low frequency limits and the characteristic frequency are used as three parameters. The approximate calculation formula for the full frequency band is as follows:

[0125]

[0126] In the above calculation formula, V0 is the lower limit velocity of the low frequency, indicating that the pressure in the pores is in a relatively balanced state; V ∞is the upper limit speed of high frequency, indicating that the pore pressure is still in a relatively unbalanced state under the influence of wave propagation; f c is the characteristic frequency, indicating the frequency at which the velocity change is the strongest, serving as the dividing line between the relatively balanced state and the relatively unbalanced state of the pore pressure; f is any frequency. The above calculation formula can be used to establish a seismic-logging velocity dispersion calculation model. At the same time, the complex modulus at any frequency can be obtained from the following calculation formula:

[0127]

[0128] By comparing and analyzing this calculation formula with the above calculation formula for velocity dispersion, a velocity dispersion correction model can be obtained, and the calculation formula is as follows:

[0129]

[0130] In the formula: M*(f) is the corrected equivalent modulus; M W is the square of the logging velocity multiplied by the density; M d is the square of the seismic velocity multiplied by the density; c is a coefficient related to the fluid properties; f is the correction frequency.

[0131] By using the above velocity dispersion correction model, the low-frequency velocity, medium-frequency velocity, and high-frequency velocity can be corrected respectively.

[0132] Step six, fuse the corrected low-frequency velocity, medium-frequency velocity, and high-frequency velocity to obtain a velocity inversion result.

[0133] In one embodiment, as Figure 10 and Figure 11 shown, fusing the corrected low-frequency velocity, medium-frequency velocity, and high-frequency velocity obtains a high-resolution and wide-frequency velocity inversion result. That is, merging in the frequency domain to obtain the final high-resolution wide-frequency velocity inversion result.

[0134] Embodiment Three

[0135] In this embodiment, as Figure 12 shown, a formation velocity parameter prediction device for frequency correction of a rock physics model is provided, including:

[0136] A data acquisition module 210, configured to acquire core data, logging data, and seismic data;

[0137] An analysis module 220, configured to perform modeling analysis on the core data to obtain velocity dispersion characteristics;

[0138] A data correction module 230, configured to perform dispersion correction on the logging data through the velocity dispersion characteristics to obtain logging correction data;

[0139] A frequency division velocity acquisition module 240 is used to perform velocity inversion on the seismic data and the well logging correction data to obtain low-frequency velocity, medium-frequency velocity and high-frequency velocity;

[0140] A speed correction module 250, used to correct the low-frequency speed, the medium-frequency speed and the high-frequency speed respectively by using the speed dispersion characteristics;

[0141] The result acquisition module 260 is used to fuse the corrected low-frequency velocity, the medium-frequency velocity and the high-frequency velocity to obtain a velocity inversion result.

[0142] In one embodiment, the data acquisition module includes a data preprocessing unit, and the data preprocessing unit is used to acquire well logging and drilling data; and preprocess the well logging and drilling data to obtain the well logging data.

[0143] In this embodiment, the data preprocessing unit is used to preprocess the multi-source data of logging and drilling, including: removing data outliers, performing data sensitive parameter intersection and data parameter standardization on the logging and drilling data.

[0144] In one embodiment, the analysis module includes a modeling unit and an analysis and comparison unit;

[0145] The modeling unit is used to perform rock physics modeling on the core data.

[0146] The analysis and comparison unit is used to analyze and compare the constructed rock physics model with different models to obtain velocity dispersion characteristics.

[0147] In one embodiment, the frequency division speed acquisition module includes a low frequency speed acquisition unit, a medium frequency speed acquisition unit and a high frequency speed acquisition unit;

[0148] The low-frequency velocity acquisition unit is used to perform low-frequency velocity inversion on the seismic data and the well logging correction data using a deep learning network to acquire low-frequency velocity.

[0149] The intermediate frequency velocity acquisition unit is used to perform velocity inversion in the seismic frequency band on the seismic data and the well logging correction data using constrained sparse pulse inversion to acquire the intermediate frequency velocity.

[0150] The high-frequency velocity acquisition unit is used to perform high-frequency velocity inversion on the seismic data and the well logging correction data using seismic waveform phase-controlled inversion to acquire high-frequency velocity.

[0151] For the specific limitations of the formation velocity parameter prediction device for rock physics model frequency correction, reference can be made to the limitations of the formation velocity parameter prediction method for rock physics model frequency correction in the above text, which will not be elaborated here. Each unit in the above-mentioned formation velocity parameter prediction device for rock physics model frequency correction can be implemented in whole or in part by software, hardware, and their combination. The above-mentioned units can be embedded in the processor of the computer device in hardware form or be independent of it, or be stored in the memory of the computer device in software form, so as to facilitate the processor to call and execute the operations corresponding to each of the above units.

[0152] Embodiment 4

[0153] In this embodiment, a computer device is provided. Its internal structural diagram can be as Figure 13 shown. The computer device includes a processor, a memory, a network interface, a display screen, and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program, and a database is deployed on the non-volatile storage medium, and the database is used to store core data, logging data, seismic data, velocity dispersion characteristics, low-frequency velocity, medium-frequency velocity, and high-frequency velocity. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with other computer devices on which application software is deployed. When the computer program is executed by the processor, it implements a method for processing goods in and out data. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a button, a trackball, or a touchpad provided on the housing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0154] Those skilled in the art can understand that Figure 13 the structure shown in

[0155] is only a block diagram of some structures related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0156] Obtain core data, logging data, and seismic data;

[0157] Perform modeling analysis on the core data to obtain velocity dispersion characteristics;

[0158] Perform dispersion correction on the well logging data based on the velocity dispersion characteristics to obtain corrected well logging data;

[0159] Perform velocity inversion on the seismic data and the corrected well logging data to obtain low-frequency velocity, medium-frequency velocity, and high-frequency velocity;

[0160] Utilize the velocity dispersion characteristics to correct the low-frequency velocity, the medium-frequency velocity, and the high-frequency velocity respectively;

[0161] Fuse the corrected low-frequency velocity, medium-frequency velocity, and high-frequency velocity to obtain the velocity inversion result.

[0162] In one embodiment, when the processor executes the computer program, the following steps are further implemented:

[0163] Obtain core data and seismic data;

[0164] Obtain well logging and drilling data;

[0165] Preprocess the well logging and drilling data to obtain the well logging data.

[0166] In one embodiment, when the processor executes the computer program, the following steps are further implemented:

[0167] Perform data outlier rejection, data sensitive parameter crossplotting, and data parameter standardization processing on the well logging and drilling data.

[0168] In one embodiment, when the processor executes the computer program, the following steps are further implemented:

[0169] Perform rock physics modeling on the core data, analyze and compare different models to obtain velocity dispersion characteristics.

[0170] In one embodiment, when the processor executes the computer program, the following steps are further implemented:

[0171] Perform velocity inversion on the seismic data and the corrected well logging data to obtain medium-frequency velocity and high-frequency velocity;

[0172] Use a deep learning network to perform low-frequency velocity inversion on the seismic data and the corrected well logging data to obtain low-frequency velocity.

[0173] In one embodiment, when the processor executes the computer program, the following steps are further implemented:

[0174] Perform velocity inversion on the seismic data and the corrected well logging data to obtain low-frequency velocity and high-frequency velocity;

[0175] For the seismic data and well logging correction data, use constrained sparse pulse inversion to perform velocity inversion in the seismic frequency band to obtain the intermediate frequency velocity.

[0176] In one embodiment, when the processor executes the computer program, the following steps are further implemented:

[0177] Perform velocity inversion on the seismic data and the well logging correction data to obtain the low-frequency velocity and the intermediate frequency velocity;

[0178] For the seismic data and well logging correction data, use seismic waveform phased inversion to perform velocity inversion in the high-frequency band to obtain the high-frequency velocity.

[0179] Embodiment Five

[0180] In this embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0181] Obtain core data, well logging data, and seismic data;

[0182] Perform modeling analysis on the core data to obtain the velocity dispersion characteristics;

[0183] Perform dispersion correction on the well logging data through the velocity dispersion characteristics to obtain the well logging correction data;

[0184] Perform velocity inversion on the seismic data and the well logging correction data to obtain the low-frequency velocity, the intermediate frequency velocity, and the high-frequency velocity;

[0185] Use the velocity dispersion characteristics to correct the low-frequency velocity, the intermediate frequency velocity, and the high-frequency velocity respectively;

[0186] Fuse the corrected low-frequency velocity, intermediate frequency velocity, and high-frequency velocity to obtain the velocity inversion result.

[0187] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0188] Obtain core data and seismic data;

[0189] Obtain well logging and drilling data;

[0190] Preprocess the well logging and drilling data to obtain the well logging data.

[0191] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0192] Perform data outlier rejection, data sensitive parameter crossplotting, and data parameter standardization processing on the well logging and drilling data.

[0193] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0194] Perform petrophysical modeling on the core data, analyze and compare different models, and obtain velocity dispersion characteristics.

[0195] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0196] Perform velocity inversion on the seismic data and the well log calibration data to obtain intermediate-frequency velocity and high-frequency velocity;

[0197] Perform low-frequency velocity inversion on the seismic data and the well log calibration data using a deep learning network to obtain low-frequency velocity.

[0198] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0199] Perform velocity inversion on the seismic data and the well log calibration data to obtain low-frequency velocity and high-frequency velocity;

[0200] Perform velocity inversion in the seismic frequency band on the seismic data and the well log calibration data using constrained sparse pulse inversion to obtain intermediate-frequency velocity.

[0201] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0202] Perform velocity inversion on the seismic data and the well log calibration data to obtain low-frequency velocity and intermediate-frequency velocity;

[0203] Perform high-frequency band velocity inversion on the seismic data and the well log calibration data using seismic waveform phase-controlled inversion to obtain high-frequency velocity.

[0204] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the embodiments provided in the present application can include non-volatile and / or volatile memories. Non-volatile memories can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memories can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0205] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0206] The above-described embodiments merely represent several implementation manners of the present application. The description thereof is relatively specific and detailed, but it should not be construed as a limitation on the scope of the invention patent. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the appended claims.

Claims

1. A method for predicting formation velocity parameters with frequency correction of a rock physics model, characterized in that, Including: Obtain core data, logging data, and seismic data; Perform modeling analysis on the core data to obtain velocity dispersion characteristics; Perform dispersion correction on the logging data through the velocity dispersion characteristics to obtain corrected logging data; Perform velocity inversion on the seismic data and the corrected logging data to obtain low-frequency velocity, medium-frequency velocity, and high-frequency velocity; Use the velocity dispersion characteristics to correct the low-frequency velocity, the medium-frequency velocity, and the high-frequency velocity respectively; Fuse the corrected low-frequency velocity, medium-frequency velocity, and high-frequency velocity to obtain the velocity inversion result.

2. The method according to claim 1, characterized in that The steps of obtaining core data, logging data, and seismic data include: Obtain core data and seismic data; Obtain logging and drilling data; Preprocess the logging and drilling data to obtain the logging data.

3. The method according to claim 2, wherein The steps of preprocessing the logging and drilling data to obtain the logging data include: Perform data outlier removal, data sensitive parameter crossplotting, and data parameter standardization on the logging and drilling data.

4. The method according to claim 1, wherein The steps of performing modeling analysis on the core data to obtain velocity dispersion characteristics include: Perform rock physics modeling on the core data, analyze and compare different models to obtain velocity dispersion characteristics.

5. The method according to claim 1, wherein The steps of performing velocity inversion on the seismic data and the corrected logging data to obtain low-frequency velocity, medium-frequency velocity, and high-frequency velocity include: Perform velocity inversion on the seismic data and the corrected logging data to obtain medium-frequency velocity and high-frequency velocity; Use a deep learning network to perform low-frequency velocity inversion on the seismic data and the corrected logging data to obtain low-frequency velocity.

6. The method according to claim 1, wherein The steps of performing velocity inversion on the seismic data and the corrected logging data to obtain low-frequency velocity, medium-frequency velocity, and high-frequency velocity include: Perform velocity inversion on the seismic data and the corrected logging data to obtain low-frequency velocity and high-frequency velocity; Use constrained sparse pulse inversion to perform velocity inversion in the seismic frequency band on the seismic data and the corrected logging data to obtain medium-frequency velocity.

7. The method according to claim 1, characterized in that, The steps of performing velocity inversion on the seismic data and the corrected logging data to obtain low-frequency velocity, medium-frequency velocity, and high-frequency velocity include: Perform velocity inversion on the seismic data and the corrected logging data to obtain low-frequency velocity and medium-frequency velocity; Use seismic waveform phase-controlled inversion to perform high-frequency band velocity inversion on the seismic data and the corrected logging data to obtain high-frequency velocity.

8. A formation velocity parameter prediction device for frequency correction of a rock physics model, characterized in that, Including: A data acquisition module for obtaining core data, logging data, and seismic data; An analysis module for performing modeling analysis on the core data to obtain velocity dispersion characteristics; A data correction module for performing dispersion correction on the logging data through the velocity dispersion characteristics to obtain corrected logging data; A frequency-divided velocity acquisition module for performing velocity inversion on the seismic data and the corrected logging data to obtain low-frequency velocity, medium-frequency velocity, and high-frequency velocity; A velocity correction module for using the velocity dispersion characteristics to correct the low-frequency velocity, the medium-frequency velocity, and the high-frequency velocity respectively; A result acquisition module, configured to fuse the corrected low-frequency velocity, medium-frequency velocity, and high-frequency velocity to obtain a velocity inversion result.

9. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images