180 results about "Seismic interpretation" patented technology

The invention discloses a fault identification method combining enhanced convolution and multi-scale attention, and relates to the technical field of seismic interpretation and reservoir prediction. The method comprises the following steps: firstly, on the basis of seismic data high-resolution enhancement processing, generating a contrast-invariant four-channel image representation by using a local intensity sequence transformation filter; then, constructing an enhanced convolution encoder in combination with differential convolution to extract fault edge information, and meanwhile, constructing a convolution Mama attention module by fusing multi-scale separable convolution and a Mama framework to extract space and global information; and finally, constructing a feature fusion module to fuse the shallow features of the jump connection with the deep features of the decoder, and outputting a fault recognition result through full connection layer mapping. Compared with the prior art, the method combines the advantages of the local intensity sequence transformation, the attention mechanism, the convolutional network and the Mama architecture, can more effectively learn the fault feature information in the seismic data, and improves the fault recognition precision.

The invention relates to a seismic frequency broadening method, system, equipment and medium for seismic frequency division and inverse Q filtering, and the method comprises the steps: carrying out the spectrum analysis of post-stack seismic data, and decomposing the original post-stack seismic data of a target interval into frequency division data bodies of different frequency bands through generalized S transformation; based on the deep logging curve, a Q value curve is obtained through estimation by means of a preset Q value fitting formula; establishing a three-dimensional Q-value data volume by using the Q-value curve obtained by fitting and combining a seismic interpretation result; on the basis of the three-dimensional Q-value data volume, performing inverse Q filtering processing on the obtained frequency division data volumes of the medium and high frequency bands, and combining the frequency division data volumes of the medium and high frequency bands after inverse Q filtering processing with the frequency division data volumes of the low frequency bands; and based on the combined post-stack seismic data volume, carrying out seismic frequency expansion by taking an artificially synthesized seismic record as quality control, and carrying out seismic interpretation and sedimentary unit fine description. The method can be widely applied to the technical field of oil-gas exploration and development.

A method is described for 3D seismic interpretation using a pre-trained 3D transformer that is fine-tuned for desired types of interpretation. It may be used with field seismic or synthetic seismic. The method is executed by a computer system.

The invention relates to a fault stability evaluation method considering a pressure-stress coupling effect. The method comprises the following steps: calculating a pressure-stress coupling ratio according to target layer rock mechanical parameters or field actual measurement pressure and ground stress data; fault three-dimensional structure modeling is carried out based on the seismic interpretation result, and fault occurrence distribution is simulated and calculated; analyzing and calculating the initial stress of the section and the stress after the coupling action by combining the three-way principal stress, wherein the stress comprises shear stress and effective normal stress; a quadratic equation of unary about fault instability critical pressure is constructed, and fault instability risk types are judged in combination with the actual pressure limit of oil field production; and calculating the critical pressure of fault instability under different risk types. According to the method, the influence of the pressure-stress coupling effect on the fault stability can be considered, so that the fault stability evaluation result is more practical.

Storing and retrieving subsurface rock physical property models for well drilling by generating estimated subsurface rock physical property models based on a respective set of seismic data, determining a stratigraphic classification for each model based on seismic interpretation data; storing the models associated classification in a database, determining characteristics of the seismic data used to generate the models, selecting from the database a subset of the models based on a similarity of the stratigraphic classification, determining measures of similarity between the characteristics of the seismic data used to generate each of subset of the models and a new set of seismic data and retrieving an estimated subsurface rock physical property model from the subset models based on the similarity measures.

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 image super-resolution reconstruction method based on a double-branch multiple attention network, and belongs to the technical field of seismic exploration, and the method comprises the steps: constructing a data sample library for model training; adopting a super-resolution backbone network and a gradient branch auxiliary network to construct a double-branch multiple attention network model; and training the double-branch multiple attention network model by adopting the data sample library and the mixed loss function to obtain a network model for realizing seismic image super-resolution reconstruction and high-frequency texture information reconstruction, and performing reconstruction based on the trained network model to obtain a seismic image subjected to super-resolution reconstruction. According to the method, a gradient branch is introduced as an auxiliary network, so that the super-resolution reconstruction effect is enhanced. Meanwhile, a mixed loss function is adopted to train the network, and a loss function for constraining high-frequency information is particularly introduced, so that the common structural deformation problem in a super-resolution task is effectively solved, the generation of artifacts is prevented, and a new technical approach is provided for high-precision seismic interpretation.

A method for the time-depth conversion of a concealed structure in a foreland thrust belt with a low exploration degree. The method comprises: obtaining velocity variation characteristics of a study area; using seismically interpreted horizons and fault data in a time domain to establish a geological structure model; by means of a stratigraphically controlled and fault-block-division-based velocity assignment method, converting the geological structure model into an interval velocity model; establishing virtual well control points in the study area, and extracting from the interval velocity model an interval velocity curve at each virtual well control point (S400); converting the interval velocity curve at each virtual well control point into an average velocity curve (S500); using a stratigraphically controlled and fault-block-division-based virtual well control point spatial interpolation method to construct a three-dimensional average velocity volume; and performing time-depth conversion on target horizons. Further provided is an apparatus for the time-depth conversion of a concealed structure in a foreland thrust belt with a low exploration degree. By means of an obtained depth-domain structural map, the identification of a concealed structure in a foreland thrust belt with a low exploration degree can be realized, thereby providing reliable depth data for the detailed description of traps.

The invention relates to the frontier field of the geological exploration technology, and discloses a three-dimensional seismic fine interpretation and multi-attribute combined complex fracture system analysis method, which comprises the steps of S1, data acquisition and data preprocessing; s2, constructing a sequence stratigraphic framework; s3, three-dimensional seismic interpretation; s4, performing multi-attribute analysis; s5, carrying out geometric and kinematic analysis on the fracture system; s6, constructing a dynamic evolution mode; and S7, analyzing the influence of oil and gas accumulation. According to the method disclosed by the invention, the three-dimensional seismic data, the logging data and the sequence stratigraphic framework are comprehensively utilized, and a comprehensive fracture system analysis means is provided. Through multi-attribute analysis, the accuracy of fracture system identification and the meticulous degree of analysis are improved. A multi-stage dynamic evolution mode is constructed, the long-term evolution law of the fracture system and the deep influence of the fracture system on oil and gas accumulation are disclosed, the blank in the aspect of multi-stage dynamic evolution analysis of the stretching strike-slip fracture system in the prior art is filled, and scientific basis and technical support are provided for oil and gas exploration.

The invention relates to a pre-drilling stratum depth prediction method, and belongs to the technical field of oil and gas field exploration and development. According to the seismic interpretation result and the geological data, determining well position coordinates of a design well or an intention well in the target area, converting the well position coordinates of the design well or the intention well into geographic latitude and longitude coordinates, and obtaining well point elevation data values of the design well or the intention well according to the current latitude and longitude coordinates and the topographic data of the target area; performing well-seismic joint time-depth conversion work, determining sand body top and bottom positions and a horizontal section track of a well section of the design well according to the horizon after depth conversion and a transverse change trend of seismic inversion, and predicting a horizontal section track depth value of the design well or the intention well according to a seismic datum plane and an elevation data value of the design well or the intention well; according to the method, well point elevation data can be rapidly and accurately obtained, and the well point depth value can be rapidly and accurately determined.

A method includes receiving seismic data of a subsurface. The method also includes color processing the seismic data to produce color-processed seismic data. The method also includes performing a seismic interpretation on the color-processed seismic data to identify regional geologic features. The method also includes performing first mapping along the regional geologic features to identify local geologic features. The method also includes extracting reservoir bodies from the local geologic features based upon the first mapping. The method also includes performing second mapping along the local geologic features to embed the local geological features into the regional geologic features. The method also includes delineating boundaries of reservoirs or seals in the subsurface based upon the local geologic features that are embedded into the regional geologic features. The method also includes de-risking a fluid prospect in a fluid reservoir based upon the boundaries of the reservoirs or the seals.

The invention discloses a novel method for establishing a high-precision three-dimensional velocity field in a horizontal well deployment area, and the method comprises the steps: firstly determining two or more oil layer group-level seismic standard layers near a target layer, then building a vertical reasonable constraint frame model, calibrating a velocity spectrum through a fitting relation, and then building an initial three-dimensional velocity model. And performing geological mode constrained filtering processing on the velocity curve of each well to obtain a smooth velocity curve capable of reflecting reservoir change, training the model by adopting a multi-layer feedforward neural network intelligent algorithm, and applying a model training result to the initial three-dimensional velocity model to obtain a high-precision three-dimensional velocity field. The time-depth relationship, the three-dimensional velocity spectrum, the seismic interpretation result and other information of the well are fully applied, and through pseudo well velocity analysis, velocity curve filtering, artificial intelligence algorithm fusion and the like, the error between a depth domain result and the well is reduced, so that the accuracy of horizontal well region velocity establishment is improved.

A method includes receiving field seismic data that represents a subsurface, identifying features in the field seismic data using a machine learning model that was trained using at least one first synthetic seismic data set that includes one or more features and one or more labels of the features, and at least one second synthetic seismic data set, the first and second synthetic seismic data sets both generated based on a geological model. Noise is injected into the second synthetic seismic data based on the geological model. The method also includes generating a model of the subsurface based at least in part on the features that were identified in the field seismic data using the machine learning model.

Examples of methods and systems are disclosed. The methods may include obtaining, by a geological modeling system, geological data and seismic data regarding a subsurface region of interest, generating a paleogeographic surface using the geological data and the seismic data, and selecting an adjustable geological attribute from the paleogeographic surface. The methods may also include generating, using a seismic interpretation system coupled to the geological modeling system, a surface of a seismic attribute using the seismic data, where the seismic attribute is related to the adjustable geological attribute. The methods may further include generating, by the geological modeling system, an updated paleogeographic surface, iteratively or recursively, until a stopping condition is reached, based on a difference between the paleogeographic surface and the surface of the seismic attribute, and determining a validated geological model of the subsurface region of interest based, at least in part, on the updated paleogeographic surface.

The invention provides a time-depth conversion method and device for a hidden structure of a low-exploration-degree foreland thrust zone, and belongs to the technical field of geological exploration. The method comprises the steps of obtaining speed change characteristics of a research area; establishing a geologic structure model by using the seismic interpretation horizon and fault data of the time domain; converting the geologic structure model into an interval velocity model through an interval control breaking block velocity assignment method; establishing virtual well control points in the research area, and extracting an interval velocity curve at each virtual well control point from the interval velocity model; converting the interval velocity curve at the virtual well control point into an average velocity curve; constructing a three-dimensional average velocity body by using a layer control breaking block virtual well control point spatial interpolation method; and performing time-depth conversion on the target layer. The variable-speed mapping is performed based on the time-depth conversion method realized by the invention, and through the obtained depth domain structure map, the identification of the hidden structure of the low-exploration-degree foreland thrust zone can be realized, and reliable depth data is provided for fine description of entrapment.

A method is described for generating multispectral variance volumes from seismic data. The method may include receiving, at a computer processor, a seismic image; analyzing the seismic image to determine frequency content and vertical seismic resolution; performing structural oriented filtering to generate a filtered seismic volume; performing spectral enhancement of the filtered seismic volume using the determined frequency content and vertical seismic resolution to generate a plurality of frequency-dependent seismic volumes; calculating covariance of the frequency-dependent seismic volumes to generate a plurality of covariance matrices; combining the covariance matrices to create a multispectral variance volume. The method is executed by a computer system.

A method for estimating higher-order dependencies from input data includes receiving input data. The method also includes generating multi-tapered spectral estimates based upon input data. The method also includes determining dependencies based at least partially upon the multi-tapered spectral estimates. The method also includes building or updating a deep learning foundation model to employ the dependencies.

The invention relates to a method for generating a karst cave 3D printing digital model, a storage medium and equipment, and the generation method comprises the following steps: analyzing a seismic data file in an SEGY format; arranging and storing X and Y coordinates, trace gathers (CDP) and seismic attribute data of a specified main measuring line seismic interpretation profile (Inline) and a tie line seismic interpretation profile (Xline) corresponding to each seismic data point in an index mode; generating a corresponding orthogonal grid model on the basis of the seismic data points; searching and extracting a karst cave in the orthogonal grid model according to the seismic data information of the orthogonal grid model and each grid unit, then identifying all boundary surfaces participating in forming a karst cave boundary, and processing the boundary surfaces to obtain a karst cave contour represented by a triangular grid; and outputting and storing the three-dimensional contour of the karst cave oil reservoir model obtained in the step 3.4 according to an STL file format. The STL model is constructed by identifying and analyzing the seismic data in the SEGY format, and a new tool is provided for smooth development of a 3D printing physical model.

The invention provides a well control constrained seismic data frequency expanding method based on broadband wavelets, and the method comprises the steps: S1, determining the feasibility of frequency expanding through the analysis of an amplitude spectrum and a signal-to-noise ratio spectrum of original seismic data in combination with the drilling condition of an actual work area; s2, performing high-frequency seismic forward modeling on the geologic model of the multi-well target sand body to determine a wavelet dominant frequency capable of identifying the target sand body; and obtaining three-dimensional matching factor data according to the wavelet dominant frequency and completing data frequency expansion processing with the constructed broadband wavelet. The spatial matching factor library can be obtained by fully utilizing broadband information of well data in an actual work area and existing mature geological knowledge, wavelet side lobes are further suppressed through amplitude envelope, broadband wavelets are constructed, dominant frequency improvement and frequency band expansion of seismic data are achieved, and the requirements of seismic fine interpretation are met.

The invention discloses an ancient landform restoration method based on a geological process and related equipment. The method is specifically implemented by the following key steps. Comprising the steps of recognizing a current stratum interface through seismic interpretation, analyzing and correcting the stratum denudation amount through seismic structure analysis, calculating the original deposition thickness through well-seismic combination, distinguishing rock correction difference compaction, removing the influence of a newly-added containable space through containable space analysis and correcting the surface influence through a sedimentary analogy terrain trend, and finally obtaining the comparable deposition thickness. According to the method, the original landform before sediment deposition is obtained through mirroring, the original landform morphology is obtained through recovery of all geological elements in the geological formation process on the basis of the geological evolution process, and all the elements influencing the landform are stripped and recovered from the final sediment to be corrected step by step. According to the method, all factors influencing the ancient landform restoration are comprehensively considered, and the accuracy of the ancient landform restoration result is improved.

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 data bidirectional frequency band expansion method based on carrier modulation, which relates to the field of seismic exploration, and comprises the following steps: processing original seismic data through time window segmentation, calculating an autocorrelation function and carrying out time domain continuation; solving a prediction coefficient based on entropy maximization constraint to generate an amplitude spectrum; calculating a boundary attenuation rate through boundary frequency point positioning; constructing an attenuation compensation function to generate a frequency domain carrier function; applying sparse constraint to solve a weighting factor and iteratively calculating a modulation amplitude spectrum; and finally, carrying out combined inverse transformation with an original phase spectrum to obtain expanded data. According to the method, the resolution and the signal-to-noise ratio of the seismic data are effectively improved, and the seismic interpretation precision is improved.

The invention belongs to the technical field of geologic structure interpretation and modeling, and particularly relates to a multi-data fusion seismic interpretation and geologic modeling method for a complex structural zone. The invention discloses a multi-data fusion seismic interpretation and geologic modeling method for a complex tectonic zone. The method comprises the following steps: observing and judging formation and fault development characteristics in a work area through an earth surface outcrop fault; interpreting a stratigraphic dip angle by adopting imaging logging to identify a fault; using 3D-VSP data to enhance seismic imaging and identify faults; a low-angle fault is recognized through conventional logging interpretation; identifying a velocity abrupt change interface by adopting a seismic velocity spectrum to determine a fault; through fault comprehensive judgment identified by different data and methods, the fault position is determined, and a geological structure mode of a research work area is constructed. According to the method, through fusion interpretation, fault position identification, seismic interpretation and geologic structure modeling of various data such as outcrop, logging data, seismic data and velocity field, the accuracy of geologic structure modeling can be improved, the success rate of oil-gas exploration is improved, and meanwhile, the cost is saved.

The invention provides a tight gas reservoir fractured horizontal well fracture parameter optimization method, system and equipment. The method comprises the steps that a river channel sand body space distribution model is determined according to geological parameters; determining a channel sand body heterogeneous attribute model according to the horizontal shaft parameters and the channel sand body spatial distribution model; determining an actual gas reservoir numerical simulation model of the heterogeneous river channel sand body according to the simulation production dynamics determined by the horizontal shaft parameters, the working system parameters and the heterogeneous attribute model of the river channel sand body and the actual production dynamics corresponding to the parameters; determining a heterogeneous river channel sand body crack parameter optimization mathematical model; and establishing a gas reservoir numerical simulation model considering the spatial distribution characteristics and the heterogeneous characteristics of the river channel sand body through the micro-seismic interpretation picture and the logging interpretation data of the river channel sand body of the target gas reservoir. Through fracturing fracture parameter optimization, the cumulative gas production rate is maximized, optimal matching between fracture parameters and sand body distribution is achieved, and effective guidance is provided for efficient development of tight gas reservoirs.

The invention relates to the technical field of seismic interpretation, and discloses a seismic interpretation method and device based on a parallel processing framework, and electronic equipment. The method comprises the following steps: transmitting configuration parameters of a seismic interpretation module; constructing a data reading buffer, and adopting a full-time window data reading buffer; when the data is read and buffered, the seismic interpretation module is executed in parallel to perform seismic interpretation calculation; and submitting a processing result of the seismic interpretation module. According to the device, an introduction module is used for introducing configuration parameters of a seismic interpretation module; the construction module is used for constructing a data reading buffer; and the calculation module is used for parallel seismic interpretation calculation. According to the technical scheme of the invention, a universal parallel processing framework is realized to bear each seismic interpretation module, the expandability is high, and the software design efficiency is higher; a typical producer-consumer mode calculation framework is realized, and the calculation efficiency is improved. The spelling algorithm thread is realized, and the parallel processing efficiency is improved.

A system and method in accordance with the present disclosure include a workflow applied to identify areas of confidence in seismic interpretations that meets a pre-selected threshold by utilizing machine-learning (ML) techniques. The ML techniques enable one or more filters to be applied, based on a range of confidence values, to areas of seismic interpretation both in the vicinity of the faults but also away from fault locations. The contribution of the various inputs is evaluated and weighted by the ML model, which reduces the time to prepare input seismic interpretation data and to obtain results from executing the model.

The present application relates to the technical field of rock lithology exploration, and more particularly to a dolomite lithology prediction method, system and device; comprising the following steps: S1, obtaining drilled well data and seismic data of a region to be identified; S2, performing standardization processing; S3, performing well control feature frequency extraction; S4, performing fine production of a synthetic record for the region to be identified; S5, determining a seismic interpretation mode scheme for the region to be identified; S6, determining dolomite attributes; S7, obtaining a first content of dolomite in the region to be identified; S8, determining a second content of dolomite in the region to be identified; S9, determining whether the dolomite attributes determined in the step S7 are qualified according to the first content of dolomite and the second content of dolomite; and quantitatively predicting the content of dolomite in a carbonate rock formation.

This invention relates to a method, apparatus, migration method, device, and medium for establishing an anisotropic parameter field. The method includes: obtaining the thickness of the target logging layer at each well point based on well logging layer information; obtaining the thickness of the target seismic interpretation layer corresponding to the well point based on seismic interpretation layer information; calculating the thickness factor of the well point at the target logging layer based on the thickness of the target seismic interpretation layer and the thickness of the target logging layer; and correcting the target logging layer based on the thickness factor of each well point and the thickness of the target seismic interpretation layer to obtain the corrected target logging layer, thereby establishing an anisotropic parameter field. In this embodiment, by constraining the logging layers between well points under the lateral trend of the seismic interpretation layer, the difference between the established anisotropic parameter field and the anisotropic parameter field of the actual subsurface medium is reduced.

The application discloses a multi-azimuth post-stack seismic attribute interpretation method, device, medium and equipment, and the method comprises the following steps: acquiring a target layer seismic interpretation horizon and a plurality of azimuth seismic attribute data bodies; setting a window used for calculating eigenvalues; selecting an eigenvalue; for the selected eigenvalue, in the window, based on the plurality of azimuth seismic attribute data bodies, attribute optimal eigenvalues of each position point in the plurality of azimuth seismic attribute data body spaces are obtained; based on the attribute optimal eigenvalues of each position point, an eigenvalue data body is obtained; and an attribute plan view is extracted in the eigenvalue data body along the target layer seismic interpretation horizon. The application acquires the most representative attribute advantage eigenvalues in different azimuths, and constructs a new data body from the eigenvalues, so that the purpose of highlighting the geological abnormal response characteristics and reducing the dimension interpretation is achieved, and the efficiency and precision of the post-stack multi-azimuth seismic interpretation are effectively improved.

The invention belongs to the technical field of earthquake comprehensive interpretation research, and discloses a lithofacies identification method based on difference analysis, which comprises the following steps: firstly, calculating to obtain difference data of a longitudinal wave velocity logging curve, difference data of a transverse wave velocity logging curve and difference data of a density logging curve; performing clustering analysis on the difference data of the longitudinal wave velocity logging curve, the difference data of the transverse wave velocity logging curve and the difference data of the density logging curve by using a Gaussian Mixture method according to the drilling lithofacies type to obtain logging lithofacies data; analyzing the lithofacies depth trend of the well logging according to the lithofacies data of the well logging, and predicting the lithofacies space distribution of the well logging by taking the lithofacies depth trend as a constraint and utilizing a Bayesian method; the invention further provides a lithofacies recognition device and equipment based on difference analysis and a storage medium. The lithofacies can be accurately identified, the identification working efficiency is improved, and the method has a good application effect and a good popularization prospect.