167 results about "Seismic trace" patented technology

The invention discloses a pre-stack seismic elastic parameter inversion method based on a conditional diffusion model, and the method comprises the steps: obtaining a pre-stack CMP multi-angle seismic trace set and a well data low-frequency trend, mapping seismic data and an elastic parameter model to a potential space, training the conditional diffusion model in the potential space, carrying out the step-by-step denoising in order to recover the distribution of the elastic parameter, and carrying out the inversion of the elastic parameter. In the sampling process, Aki-Richards equation physical constraint is introduced, and uncertainty quantization is realized through transverse constraint and multiple sampling. According to the method, the spatial continuity of an inversion result can be ensured, uncertainty quantification can be realized, and reliable confidence evaluation is provided for geological interpretation. The method is wide in applicability and high in calculation efficiency, and the pre-stack elastic parameter inversion result with higher precision and higher credibility can be quickly obtained in actual three-dimensional seismic data processing.

The present application belongs to the technical field of geophysical well logging data processing, and particularly relates to a multi-scale dual-branch neural network well logging curve reconstruction method based on seismic constraint, which comprises reading in well-seismic and interpreted horizon data, target interval sublayering and facies belt division based on three-dimensional seismic interpretation, pyramid multi-scale training set sample construction by fusing multi-curve and well flanking seismic trace information, residual-Transformer dual-branch neural network training model establishment taking a composite loss function as the core, three-stage progressive training strategy and reconstruction of target well missing section and result evaluation. The method still maintains high stability and robustness under the condition of complex heterogeneous reservoirs; on the basis of seismic data constraint, geological sublayering and facies belt division and other interpretation information are fused, so that the reconstruction result is more in line with geological laws; and the obtained reconstructed curve provides reliable support for re-interpretation of old data and full use of incomplete well logging data.

The invention discloses a seismic wave velocity model inversion method based on time-space coupling deep learning, and the method achieves a time-space coupling mechanism through a Fourier neural operator framework of a wavelet transform-attention mechanism, and supports the reconstruction of a spatial two-dimensional velocity model from a time sequence track of multiple seismic sources. Firstly, the local time-frequency representation of each seismic time-domain trajectory is extracted by adopting continuous wavelet transform, and then a wavelet time-frequency spectrogram is coded through a convolution feature extractor. Then, information from multiple sources is fused at each receiver point location through a multi-head attention mechanism, enhancing correlation between multiple sources. And finally, projecting the fused features into a two-dimensional Fourier neural operator to efficiently learn space mapping and reconstruct a bottom-layer velocity model based on sound velocity. According to the method, the time sequence characteristic and the spatial distribution characteristic of the seismic channel are fully utilized, the precision and robustness of an inversion result are improved, and a more reliable speed model is provided for geological structure interpretation and acoustic imaging.

Examples of methods and systems are disclosed. The methods include, obtaining a seismic dataset regarding a subsurface region of interest, wherein the seismic dataset comprises a plurality of time-domain seismic traces. The methods also include determining a plurality of single-frequency volumes from the plurality of time-domain seismic traces. The methods further include determining a frequency ratio volume based, at least in part, on the plurality of single-frequency volumes. The methods still further include extracting a three-dimensional (3D) geological body based, at least in part, on the frequency ratio volume. The methods further include determining a drilling target in the subsurface region based on the 3D geological body.

The invention discloses an offshore field two-dimensional seismic data rapid imaging method, and relates to the technical field of seismic exploration data processing, and the method comprises the steps: obtaining SEGD format original seismic data of a target region from a seismic collection system, combining the data of each shot of each measuring line, decoding the data into internal available data, and obtaining a to-be-processed data body; and loading an observation system and assigning seismic trace header parameters according to actual acquisition parameters based on the decompiled to-be-processed data volume. According to the invention, through a multi-stage denoising and signal protection strategy, effective suppression of various sea interferences is realized, a frequency division multi-channel statistical abnormal amplitude suppression technology is adopted, zero-phase high-pass filtering is combined, low-frequency effective signals are protected while low-frequency strong-energy background noise is removed, and the noise suppression effect is improved. The direct wave cutting step further eliminates strong energy interference under a long cable condition, highlights an effective reflection signal of middle and far offset data, forms a progressive signal-to-noise ratio improvement mechanism from original data to final imaging, and improves the interpretability of seismic data.

The invention belongs to the technical field of geophysical logging data processing, and particularly relates to a multi-scale double-branch neural network logging curve reconstruction method based on seismic constraints. Comprising the steps of reading in well seismic and interpreting horizon data, performing target layer section layering and phase belt division based on three-dimensional seismic interpretation, fusing multiple curves and well-side seismic channel information to construct a pyramid multi-scale training set sample, establishing a residual-Transformer double-branch neural network training model by taking a composite loss function as a core, and constructing a multi-scale training set sample based on the residual-Transformer double-branch neural network training model. And a three-stage progressive training strategy is adopted, and target well missing section reconstruction and result evaluation are carried out. According to the method, high stability and robustness are still kept under the condition of a complex heterogeneous reservoir; interpretation information such as geological layering and phase belt division is fused on the basis of seismic data constraints, so that a reconstruction result better conforms to geological laws; and the obtained reconstruction curve provides reliable support for re-explanation of old data and full utilization of incomplete logging data.

The invention provides a seismic fault identification method and system based on multi-attribute collaboration and spectral domain graph enhancement, and relates to the technical field of seismic exploration, and the method comprises the steps: carrying out the all-directional dip angle and azimuth angle scanning of a seismic data volume, extracting multi-scale guide field information through combining structure tensor decomposition, and executing anisotropic diffusion filtering; calculating characteristic value distribution through a characteristic value coherence algorithm based on the filtering data volume, and determining the structural consistency difference of adjacent seismic traces to obtain a fracture coherence attribute data volume; carrying out azimuth gather sorting and pre-stack time migration processing on the seismic data volume, extracting seismic wave dynamic response characteristics, carrying out Fourier series expansion on the azimuth change rate to obtain a crack indication information data volume, splicing the data volume and executing multi-scale three-dimensional convolution solution, and constructing a spatial dependency graph through a spectral clustering algorithm; spectral domain enhancement features are obtained through spectral domain graph transformation and frequency selective filtering reconstruction, and morphological connectivity analysis is executed to obtain a crack prediction result.

The invention provides a three-dimensional scattering imaging method, device and equipment and a storage medium, and relates to the technical field of seismic wave imaging, and the method comprises the steps: collecting seismic wave data through a three-component detector, carrying out the polarization analysis, obtaining the actual path data of scattered waves, constructing a common scattering point gather of each preset underground scattering point, and obtaining a common scattering point gather of each preset underground scattering point; when the common scattering point gather is constructed, the amplitude is adaptively weighted according to the consistency of theoretical path data and actual path data of a seismic trace, and then imaging is performed based on the common scattering point gather to obtain an imaging result including the underground scatterer. According to the method, by introducing path consistency weighting, the common scattering point gather quality and the imaging precision are remarkably improved.

The invention belongs to the technical field of seismic exploration, and discloses a seismic common reflection point intelligent superposition imaging method under three-dimensional structure feature constraint, which comprises the following steps of: 1, constructing a three-dimensional structure geometric model of a target area, and extracting structure features to generate a structure feature constraint field; 2, dividing the target area into a structure gentle area and a structure sudden change area according to the structure feature constraint field, and selecting a wide azimuth gather for the structure gentle area to carry out dynamic correction and superposition so as to obtain a first superposition result; selecting an optimal narrow azimuth gather subset in a narrow azimuth range along the structure trend for a structure mutation region; 3, superposing seismic traces in the subset of the optimal narrow azimuth trace gather according to a self-adaptive weight coefficient driven by a structural feature to obtain a second superposition result; and a fourth step of fusing the first superposition result and the second superposition result to obtain seismic common reflection point superposition imaging. According to the method, partition differentiation processing is constrained according to structural features, and imaging better conforms to real geologic features.

This invention provides a method for calculating the minimum delay time between two seismic traces: Calculating a first error value between the two seismic traces under at least three first delay times (either preset or obtained after the last iteration update); fitting an error function between the delay time and the error value; calculating a second delay time and a second error value at the extreme point of the error function; if the first difference between the second delay time and a threshold delay time of any of the first delay times is greater than or equal to a first threshold, or the second difference between the second error value and the first error value is greater than or equal to a second threshold, then the second delay time is used to replace one of the edge delay times in the current first delay time. All the above steps are repeated until the first difference is less than the first threshold and the second difference is less than the second threshold, and this second delay time is determined as the minimum delay time between the two seismic traces. This invention improves the calculation speed for calculating the minimum delay time between two seismic traces.

The present invention belongs to the technical field of petroleum and natural gas exploration. Provided are an intelligent seismic high-frequency sequence interpretation method and apparatus using logging information. The method comprises: inputting, into a sequence stratigraphy prediction model in a current state, seismic attribute data of determined point locations, which are determined on the basis of a sequential path, along a well-tie seismic trace of a known conditioning well in a target block, so as to obtain high-frequency sequence stratigraphy data of the determined point locations; and determining whether any high-frequency sequence stratigraphy data of a point location in the target block has not been obtained, and if any high-frequency sequence stratigraphy data of a point location in the target block has not been obtained, adding the high-frequency sequence stratigraphy data of the determined point locations to sequential simulation conditioning data, using updated sequential simulation conditioning data to construct a learning sample to train the sequence stratigraphy prediction model in the current state, and then jumping to the previous step to perform high-frequency sequence stratigraphy prediction on the next determined point location determined on the basis of the sequential path. The method can be applied to onshore early-exploration blocks and offshore sparse-well blocks, etc., so as to realize high-frequency sequence interpretation of these blocks with high efficiency, low costs and high accuracy.

The application discloses a method for extracting and correcting the time difference of synchronous seismic data collection while digging, which comprises the following steps: 1, establishing a seismic data collection system while digging, reading all seismic wave data in the collection system and filtering; 2, performing velocity analysis on the filtered seismic wave to obtain the propagation velocity of the seismic direct wave while digging; 3, using the seismic wave propagation velocity original data to correct the direct wave time difference of each seismic trace, and eliminating the influence of the seismic wave propagation time difference; 4, calculating the model trace data of each collection substation through the data after the time difference correction; 5, determining the reference substation according to the model trace data of each collection substation and the data of each seismic trace in the corresponding collection substation; and 6, calculating the time difference of the corresponding collection substation according to the cross-correlation time difference between the model trace of the reference substation and the model trace of each collection substation, and correcting the original data.

This invention belongs to the field of seismic exploration, specifically relating to a method for calculating minimum offset and delay based on direct waves and seafloor reflected waves. The method includes: selecting single-shot seismic data from a flat seafloor location, picking the direct wave travel time, and picking the seafloor reflected travel time of the first data trace at the single-shot location; performing regression fitting on the picked direct wave travel time to fit the direct wave equation, and obtaining the seawater velocity and initial minimum offset based on the direct wave equation; setting an initial delay time range, and calculating the seafloor depth h for each seismic trace at different delay times based on the seawater velocity, initial minimum offset, and seafloor reflected travel time; calculating the 2h variance based on the seafloor depth of the seismic trace, and obtaining the minimum offset and delay time. This invention can quickly obtain the minimum offset and delay time, achieving accurate positioning of seismic data, and the method is simple and easy to implement.

The application discloses a multiple wave prediction method and device, a computing device and a storage medium. The method comprises the following steps: acquiring frequency domain data of any shot point at any seismic trace; generating a seismic data matrix according to the frequency domain data, the seismic data matrix being an upper triangular matrix with a main diagonal line of 0, and any non-first column of the seismic data matrix containing frequency domain data of at least one seismic trace of the same shot point; constructing a multiple wave matrix, the multiple wave matrix being an upper triangular matrix with a main diagonal line of 0, and any non-first column of the multiple wave matrix containing a multiple wave model of at least one seismic trace of the same shot point; establishing a functional relationship between the multiple wave matrix and the seismic data matrix; and obtaining model frequency domain data corresponding to any multiple wave model in the multiple wave matrix according to the functional relationship. According to the scheme, the prediction efficiency of the multiple wave can be improved under the premise of ensuring the prediction accuracy of the multiple wave.

The invention belongs to the technical field of petroleum seismic exploration, and provides an exogenous interference suppression method based on background noise recognition and a related device, and the method comprises the steps: carrying out the first arrival time pickup of original single-shot data obtained through seismic collection, and dividing the single-shot data into a background noise region and a signal + noise mixed region according to the picked first arrival time; identifying seismic traces containing external interference in the background noise area according to the energy difference; in a signal + noise mixing area, for the identified seismic trace containing the external interference, sparse representation is carried out on seismic data through a morphological component analysis method (Q wavelet transform and cosine transform); the sparse distribution coefficient is screened by using the characteristic difference between the effective signal and the external interference, and gradual decomposition of the effective signal and the external interference is realized through an iteration mode; and reconstructing the screened Q wavelet coefficient to obtain a shot gather record after exogenous interference suppression. Compared with a conventional denoising method which only utilizes apparent velocity, frequency and energy differences, the method can more accurately realize separation of external interference and effective reflected waves, and the external interference suppression effect is better.

The application provides a near-trace residual multiple wave suppression method and system based on deep learning, and belongs to the field of oil and gas exploration and development. The near-trace residual multiple wave suppression method based on deep learning comprises steps 1) to 19). The application obtains a large amount of label data which is consistent with the near-trace data characteristics of the actual seismic trace set and does not contain multiple wave signals by performing AVO background trend correction, spectrum equalization processing, time-varying wavelet inversion and time-varying wavelet convolution processing on far-trace data in actual seismic data. Deep learning training is only performed in a water-containing background range, the influence of oil and gas on the AVO characteristics of the seismic trace set is avoided, and then a deep neural network for suppressing multiple waves can be more accurately constructed, the near-trace residual multiple wave suppression effect can be improved, and reliable trace set data can be provided for subsequent AVO analysis and pre-stack reservoir prediction.

This invention discloses a thin-layer weak signal enhancement method based on cyclic spectrum enhancement technology. The method first preprocesses the original seismic data using a Gaussian smoothing operator to suppress random noise while preserving the main signal features. Then, even-order derivative operations are introduced to enhance the high-frequency components and detailed information of the seismic signal, highlighting the reflection characteristics of the thin-layer interface. To address the issue of high-order derivatives easily introducing high-frequency noise, a Butterworth low-pass filter is used for directional noise reduction. Finally, an adaptive cyclic termination condition is used to determine the optimal derivative order, and the derivative seismic traces are superimposed onto the original seismic traces to achieve thin-layer weak signal enhancement. Numerical simulations and tests with actual seismic data show that this method can effectively enhance the weak signals of 10m (λ / 4) and 5m (λ / 8) thin layers, improve the continuity and clarity of the phase axis of thin-layer reflected waves, and thus enhance the identification capability of thin-layer reservoirs under complex geological conditions.

The application discloses a free interface multiple wave prediction method and device, a computing device and a storage medium, and the method comprises the following steps: performing prestack time migration processing on a streamer seismic data volume to obtain a prestack time migration data volume; performing reverse migration processing on the prestack time migration data volume to construct a multiple wave prediction seismic trace set; and performing free interface multiple wave prediction on the basis of seabed acquisition data and the multiple wave prediction seismic trace set to obtain a free interface multiple wave record. The scheme provided by the application realizes free interface multiple wave prediction of seabed acquisition data, can predict all free interface multiple wave components in the data, effectively solves the problem that a conventional streamer SRME method cannot be applied to such data, and can significantly improve the multiple wave suppression effect of the P-wave component and the converted wave component data of seabed acquisition data.

The invention discloses a seismic data removal processing method, system and device and a medium, and the method comprises the following steps: S1, analyzing and processing seismic data, and generating a minimum offset seismic attribute; s2, determining a'full coverage 'boundary and a'primary coverage' boundary by using the minimum offset seismic attribute, giving a corresponding water bottom removal time, and calculating an unknown water bottom removal time by using an isoline interpolation method; s3, smoothing the calculated water bottom removal time; and S4, applying a removal time table: writing all the water bottom removal time into a seismic trace header in a trace-by-trace matching mode, and carrying out zero filling processing on the amplitude value of the part above the water bottom removal time. The method is high in adaptability to irregular boundary three-dimensional seismic data, and the purposes of shortening the data processing period, reducing the labor intensity of processing personnel and improving the project operation efficiency can be achieved.

A method for training travel time-based networks and building an image of a velocity model includes obtaining a seismic dataset of seismic traces and determining an observed travel time for each seismic trace. The method further includes obtaining a velocity network, that depends on one or more velocity parameters, and a travel time network, that depends on one or more travel time parameters. The method further includes training the velocity network and the travel time network using a cost function and an optimizer. The cost function is based on the travel times parameters, the velocity parameters, a travel times equation, a derivative of the travel times equation, and a travel time mismatch between a first observed travel time and a first travel time value output by the travel time network. The method further includes building the image of a velocity model using the trained velocity network.

The application discloses a seismic data interpolation denoising integrated method and system, and relates to the technical field of seismic data processing.The method comprises the following steps: firstly, obtaining noisy missing seismic data, which can be obtained by simulating a field acquisition scene or directly using actual field acquisition data; based on the compressed sensing theory, an interpolation denoising objective function is established, which comprises a data error term and a sparse promotion constraint term; based on the idea of shape regularization, an iterative solution framework is constructed, and a local signal-noise orthogonalization process is integrated, through which the useful signals leaked in the preliminary denoising can be reconstructed; after the iterative solution of the objective function, high-fidelity reconstructed seismic data are output.The application combines the local signal-noise orthogonalization and the shape regularization, effectively restores the missing seismic traces and suppresses the noise, realizes the protection and compensation of the useful signals, and significantly improves the signal-to-noise ratio and fidelity of the reconstructed data.

The invention discloses a seismic signal adaptive amplitude compensation method based on multi-domain joint parameter calculation. The method relates to the technical field of amplitude compensation, and comprises the following steps: preprocessing an original seismic trace set and extracting multi-domain basic information, cooperatively acquiring compensation core parameters, and performing staged adaptive amplitude compensation and verifying and iteratively optimizing a compensation effect. The method comprises the steps of preprocessing an original seismic trace set, extracting multi-domain basic information, cooperatively obtaining multi-domain compensation core parameters, compensating seismic signal attenuation and energy difference in stages based on the parameters, and finally ensuring that data is adaptive to subsequent geological interpretation and reservoir prediction through compensation effect verification and iterative optimization. The accuracy of seismic signal self-adaptive amplitude compensation is improved, and the problem that in the prior art, the accuracy of seismic signal self-adaptive amplitude compensation is low is solved.

The application discloses a method and system for constructing a pre-stack inversion model based on synchronous extrusion wavelet transform fusion, which comprises the following steps: obtaining a low-frequency velocity model in a time range of seismic data and a high-frequency velocity model above the top of a fracture-vug body; performing time-frequency analysis on the high-frequency model and the low-frequency model by using synchronous extrusion wavelet transform to obtain synchronous extrusion wavelet transform coefficients; analyzing the overlap based on the synchronous extrusion wavelet transform coefficients to obtain a spectrum at a certain moment; performing frequency fusion processing on all moments of a certain trace of data based on the spectrum at the certain moment to obtain a relatively wide time-frequency spectrum of the trace of data; obtaining the wide time-frequency spectrum of all seismic traces to obtain relatively wide frequency model data. The application performs time-frequency domain fusion based on synchronous extrusion wavelet transform on two kinds of background velocity fields, thereby constructing an inversion initial model for revealing deep information, and effectively solving the problems of insufficient drilling data of the fracture-vug body and the lower part and incomplete initial model information.

This invention discloses a method for establishing time-spectral sedimentary facies samples based on seismic forward modeling. The method includes: Step S1: Constructing typical sedimentary models and various sedimentary facies property templates for the actual work area based on core sedimentary characteristics; Step S2: Obtaining sedimentary facies geological profiles based on typical sedimentary models and sedimentary facies property templates; Step S3: Extracting actual seismic forward modeling parameters from actual seismic data; Step S4: Conducting forward modeling of the sedimentary facies geological model based on the actual seismic parameters; Step S5: Converting the forward modeled seismic data into time-spectral data through continuous wavelet transform; Step S6: Extracting typical seismic traces from the forward modeled seismic profiles, using the sedimentary facies in the geological model as labels and the time-spectral data as feature values ​​to establish sedimentary facies seismic samples. A large number of high-precision samples are generated under the guidance of typical models, and the sample labels are completely accurate, eliminating errors caused by time-depth calibration, thus meeting the needs of machine learning for sedimentary facies prediction.

The application provides a seismic trace stacking method with trace header information protection, comprising the following steps: step 1, data preparation; step 2, resetting the receiver point stake number; step 3, calculating the average value of all pre-stack trace header information of the same receiver point stake number; step 4, stacking the seismic traces of the same receiver point stake number; step 5, assigning the pre-stack trace header information value to the post-stack seismic trace corresponding to the receiver point stake number; step 6, resetting the shot point stake number; step 7, calculating the average value of all pre-stack trace header information of the same shot point stake number; step 8, stacking the seismic traces of the same shot point stake number; step 9, assigning the pre-stack trace header information average value to the post-stack seismic trace corresponding to the shot point stake number; and step 10, sorting the shot point receiver point stake number and outputting data. The seismic trace stacking method with trace header information protection is reliable in effect, simple in operation and easy to realize, and can maximize the reservation of pre-stack trace header information while keeping the basic properties unchanged.

The invention relates to the technical field of oil and gas reservoir exploration and development, in particular to a wrinkle strength calculation method. The wrinkle strength calculation method comprises the following steps: firstly, intercepting a seismic trace coordinate point set based on a target layer wrinkle zone plane range, and obtaining the wrinkle width of a corresponding wrinkle zone on each seismic survey line; secondly, calculating the relative amplitude of a target layer fold zone by utilizing a structural trend surface method; assigning the relative amplitude to a seismic trace coordinate point set according to a coordinate corresponding relation to obtain a wrinkle amplitude of each seismic survey line in a corresponding wrinkle zone; and finally, dividing the wrinkle amplitude by the wrinkle width to obtain the wrinkle strength of the corresponding wrinkle zone on each seismic survey line. According to the method, seismic attributes, horizon interpretation and three-dimensional seismic survey network information are comprehensively utilized, the wrinkle range is quantitatively delineated through curvature attributes, horizon is interpreted, the wrinkle amplitude is calculated, and an objective wrinkle strength evaluation result can be obtained.

The present application belongs to the field of seismic data processing, and relates to a thin sandstone reservoir prediction method and system based on prestack phase gradient attributes, comprising: S1: obtaining original seismic data, and preprocessing the original seismic data to obtain preprocessed seismic data; S2: processing the preprocessed seismic data using complex seismic trace analysis to calculate the total phase sequence of each seismic trace; S3: constructing a sandstone-mudstone thin interbedded geological model to simulate multiple configurations and obtain the characteristic law of prestack phase gradient attributes; S4: identifying the thin sandstone reservoir based on the total phase sequence and the characteristic law of prestack phase gradient attributes, realizing fluid prediction, and obtaining a prediction result; and S5: predicting the thickness of the thin sandstone reservoir through the characteristics of the prestack phase gradient attributes in the high-velocity-difference area; avoiding the thin-layer phase interference effect that cannot be solved by traditional AVO, so as to realize the purpose of super-resolution prediction of gas-containing thin sandstone.

The application provides a multiple wave suppression method, device, equipment and storage medium, and belongs to the technical field of seismic data processing. The method comprises the following steps: determining first seismic data of a plurality of first seismic trace sets; determining the curvature and frequency of each seismic trace in the first seismic trace set; determining the maximum curvature and minimum curvature in the curvatures of a plurality of seismic traces, and determining the maximum frequency in the frequencies of a plurality of seismic traces; generating a three-dimensional conical body filter based on the maximum curvature, the minimum curvature and the maximum frequency; performing three-dimensional Radon transform on the first seismic data of the first seismic trace set to obtain three-dimensional Radon transform data; filtering the three-dimensional Radon transform data based on the three-dimensional conical body filter to obtain three-dimensional Radon multiple wave data; and determining target seismic data after suppressing multiple waves based on the difference between the first seismic data of a plurality of first seismic traces and the three-dimensional Radon multiple wave data of a plurality of first seismic trace sets. The method improves the efficiency of suppressing multiple waves.

The invention discloses an organic matter identification method based on seismic wave dispersion, and relates to the technical field of petroleum geophysical exploration, and the method comprises the steps: obtaining a target pre-stack seismic trace set and matched logging data, building an underground medium initial model based on the kerogen content and elastic response characteristics in the logging data, and obtaining an underground medium initial model; a to-be-inverted parameter matrix and a reflection coefficient weight matrix of an incident angle are calculated through a Shaw reflection coefficient calculation equation, frequency-division seismic wavelets are generated through the high sensitivity of organic matter to transverse wave dispersion, and a positive operator is constructed in combination with the wavelets and the reflection weight matrix. A shear wave velocity dispersion gradient which is highly sensitive to organic matters and weak in response to pore fluid is used as a main control inversion parameter, a combined recognition constraint term is constructed by combining a longitudinal wave velocity dispersion gradient and attenuation spectrum morphological characteristics, and a spatial optimal solution of a forward operator and a seismic trace set is calculated through a minimization objective function. And quantitatively associating the spatial optimal solution with the geophysical response characteristics of the organic matters to generate an organic matter spatial distribution diagram of the target reservoir.

The application discloses a pumping unit noise positioning method based on deep learning, and steps are as follows: S1, local seismic trace sets of several fixed sizes are obtained by using sliding window single-step sliding to block seismic data by columns, if the number of seismic traces containing pumping unit noise reaches threshold Th1, it is marked as 1, and if the number of seismic traces containing pumping unit noise is lower than threshold Th2, it is marked as 0; S2, the local seismic trace sets are pretreated, the energy spectrum is calculated, mean filtering is carried out, then down-sampling is carried out, and part of data is randomly selected as a training set, and the remaining part is used as a test set; S3, a deep CNN network is built, and the training set obtained in step S2 is used to train the CNN network, and in the training process, the misclassified data is used to supplement the training set for repeated training; S4, the test sample is used to test the CNN network trained in step S3, and the positioning function is quantitatively evaluated; S5, the width of the positioned pumping unit noise is estimated according to the positioning result in step S4.