56 results about "Seismic processing" patented technology

A method for performing a seismic processing task using a machine-learned model developed using an in-domain adversarial attacker. The method includes obtaining a machine-learned model parameterized by a set of weights and generating a synthetic seismic dataset and associated target. The method further includes determining a noise profile for the synthetic seismic dataset in a frequency domain that when added, in a spatial-temporal domain, to the synthetic seismic dataset reduces a performance of the machine-learned model. The method further includes adding the noise profile to the synthetic seismic dataset forming a noisy seismic dataset and updating the set of weights of the machine-learned model based on the noisy seismic dataset and the target. The method further includes receiving a seismic dataset corresponding to a subsurface, processing the seismic dataset with the machine-learned model to form a predicted target, and developing a geological model for the subsurface using the predicted target.

The invention discloses a seismic first arrival wave pickup method and device based on a multi-input attention mechanism network, multi-domain data and multiple network advantages are fused to realize first arrival automatic pickup, and compared with manual pickup, the efficiency is greatly improved, and the result is more objective; compared with a traditional automatic pickup method, the method is higher in precision and stability under different signal-to-noise ratios. Meanwhile, through hyper-parameter optimization, the generalization ability is high, reliable data can be provided for subsequent seismic processing, the advantages are outstanding under complex geological conditions, and the method is of great significance to geological analysis and resource exploration.

A method and system for forming a seismic image of a subterranean region are disclosed. The method includes determining an initial plan for a seismic survey with a value for each member of a set of acquisition parameters and acquiring a first seismic dataset from a first portion of the seismic survey based on the initial plan. The method further includes transmitting the first seismic dataset to a seismic processor, determining a first seismic image from the first seismic dataset by performing expedited seismic processing and determining a first updated plan for the seismic survey based on the first seismic image and acquiring a second seismic dataset from a second portion of the seismic survey based on the first updated plan. The method still further includes transmitting the second seismic dataset to the seismic processor and determining the seismic based on the first seismic dataset and the second seismic dataset.

Systems and methods for noise-robust time-domain multi-scale full waveform inversion using convolved data are disclosed. The methods include obtaining, using a seismic acquisition system, an observed seismic dataset pertaining to a subsurface region of interest; obtaining, using a seismic processor, a seismic velocity model; and iteratively, using the seismic processor, until a stopping criterion is met: selecting a wavelet with a frequency parameter, wherein the frequency parameter increases with each iteration, forming a convolved seismic dataset based on a convolution of the wavelet with the observed seismic dataset, and updating, using a full waveform inversion, the seismic velocity model based, at least in part, on the convolved seismic dataset. The methods further include forming a seismic image of the subsurface region of interest using the updated seismic velocity model.

The invention provides an inverse Q filtering seismic processing system based on artificial intelligence, and the system comprises a historical Q calculation module which calculates a depth domain Q value sequence through a Lee empirical formula, and converts the depth domain Q value sequence into a corresponding frequency band depth domain Q value sequence in combination with a seismic wave dominant frequency; the time-depth conversion module is used for performing convolution on the seismic wavelets and the reflection coefficients to generate synthetic seismic records, and converting a depth domain Q value sequence into a time domain Q value sequence by using a time-depth relationship; the model construction module is used for training a convolutional neural network to generate a Q prediction model by taking the amplitude, the frequency, the phase and the time domain Q value of historical seismic waves as training data; and the inverse Q filtering processing module inputs the multiband parameters of the current seismic wave to the Q prediction model, outputs a time domain Q prediction value, and performs multiband compensation on amplitude attenuation, frequency loss and phase distortion of the seismic wave by adopting an inverse Q filtering formula. According to the method, the defect of insufficient precision of a traditional Q compensation method is avoided, and the calculation efficiency is greatly improved.

Methods and systems for training a machine learning (ML) network to predict a likelihood of a presence of a geological fracture from an observed drill-bit seismic dataset are disclosed. The method may include obtaining, using a seismic processing system, a plurality of geophysical models, where each geophysical model includes a location of a drill bit. The method may further include simulating, for each geophysical model a corresponding simulated drill-bit seismic dataset for seismic waves emanating from the drill bit and recorded by at least one seismic receiver and forming a training dataset including a plurality of training pairs, with each training pair including a geophysical model from the plurality of geophysical models and the corresponding simulated drill-bit seismic dataset. The method may still further include training, using the training dataset, the ML network to predict the likelihood of the presence of the geological fracture from the observed drill-bit seismic dataset.

Examples of methods and systems are disclosed. The methods may include receiving seismic data regarding a subsurface region of interest, wherein the seismic data comprises a plurality of time-space waveforms, and generating, using statistical sampling, a plurality of pilot waveforms based on the plurality of time-space waveforms. The methods may also include forming a training seismic dataset comprising an input training dataset and an output training dataset, wherein the input training dataset is based on the plurality of time-space waveforms and the output training dataset is based on the plurality of pilot waveforms. The methods may further include training, using the seismic processing system and the training seismic dataset, a machine-learning (ML) model to predict the output training dataset, at least in part, from the input training dataset.

The invention relates to a seismic processing method and device based on seismic wave field numerical simulation, and belongs to the technical field of seismic exploration, and the method comprises the steps: carrying out the polynomial approximate expansion of an anisotropic pure-mode P wave and SV wave frequency dispersion relational expression, and obtaining an expanded polynomial; solving a target function corresponding to the expanded polynomial according to a preset optimization algorithm, and obtaining a polynomial optimization coefficient of the expanded polynomial; converting the optimized frequency dispersion relational expression of the P wave and the SV wave into a time-space domain, and obtaining an anisotropic pure mode wave equation; and solving the anisotropic pure-mode wave equation to realize numerical simulation of a seismic wave field, and performing prediction processing on a real-time earthquake based on a numerical simulation result. By adopting the seismic wave field numerical simulation method, high-precision numerical simulation of the seismic wave field can be realized, an effective wave field extrapolation tool is provided for subsequent anisotropic imaging and inversion, and earthquake prediction processing is realized.

The invention provides a seismic data output method of a seismic processing interpretation system, and belongs to the field of seismic exploration. The method comprises the following implementation steps: step 1, establishing a processing flow; 2, acquiring parameters input by a user; step 3, filtering data of the elastic distributed data set RDD according to parameters input by a user; step 4, outputting the channel header temporary file and the channel data temporary file in parallel; 5, extracting file header information of the header file from the elastic distributed data set RDD, and writing the file header information into an independent info file; step 6, merging the trace header temporary files to obtain an hdr file as an output file; merging the trace data temporary files to obtain a data file as an output file; and outputting the data file, the hdr file and the info file. According to the invention, the output efficiency of the seismic data is effectively improved.

This invention relates to a method for identifying the boundary of anticlines, belonging to the field of oil and gas field exploration and development technology. First, seismic data is acquired, and spectral analysis is performed on the seismic data. The phase of the seismic data is rotated, and the rotated data is interpreted using relative isochronous stratigraphic layers. The interpreted seismic top layer is flattened, and the flattened seismic data is used to determine stratigraphic slices. Based on the amplitude attributes and lateral differences of the stratigraphic slices, the planar morphology of the anticline is identified. This method saves the cost of high-resolution seismic processing and solves the problem of increased uncertainty in imaging results caused by high resolution.

Systems and methods for time preserving full waveform inversion are disclosed. The methods may include acquiring, using a seismic acquisition system, a seismic dataset and, using a seismic processor, obtaining, from the seismic dataset, an event in two-way traveltime (TWT) and a velocity model in TWT, and converting the velocity model in TWT into a velocity model in depth. The methods further include producing a final velocity model in depth, where the final velocity model is produced using a full waveform inversion (FWI), the velocity model in depth, and a TWT-preserving method preserves the event in TWT at each iteration of the FWI; and forming a seismic image in depth using the final velocity model in depth and the seismic dataset. The methods further include identifying, using a seismic interpretation workstation and based on the seismic image in depth, a drilling target; and planning a borehole path to the drilling target.

The invention relates to the technical field of petroleum seismic exploration software engineering, and particularly discloses an operation state management method and device for a seismic processing system, and the method comprises the steps: submitting an operation to a Yarn platform, and enabling the operation to flow according to the operation state of the Yarn platform; the method comprises the following steps: a Pi platform obtains an operation state of circulation of a Yarn platform, and converts the operation state of circulation of the Yarn platform into an operation state of circulation of the Pi platform; and carrying out business processing on the operation state transferred by the Pi platform through the operation management service of the Pi platform. According to the method, after a job is submitted to Yarn, the job state is transferred according to the state of Yarn, pi obtains the job state of the corresponding job, and the job state is converted; different business processing flows are executed according to different operation states and operation management services, so that correct operation in an operation life cycle is ensured, actual production requirements are met, and the stability of cluster operation and the smoothness of business processing are ensured.

A computer-implemented method for seismic processing includes receiving a seismic training input image, generating, using a first portion of a machine learning model, a first output based at least in part on the seismic training input image, generating, using a second portion of the machine learning model, a second output based at least in part on the seismic training input image, generating a loss function based at least in part on comparing at least two of the first output, a deterministic first label synthetically generated and representing a deterministic ground truth for the first output, the second output, and a non-deterministic second label representing a non-deterministic ground truth for the second output, and refining the first portion, the second portion, or both of the machine learning model based at least in part on the loss function.

The invention relates to the field of seismic processing system workflow data circulation, and particularly discloses a seismic processing workflow channel header word data circulation method and device, and the method comprises the steps: creating a global channel header word memory management object; performing dynamic memory management on each module object, a relationship between modules and a relationship between a corresponding module and a corresponding header word in the workflow based on the global header word memory management object; dynamically expanding and updating the header word; and dynamically obtaining the latest header word information of the corresponding module, and carrying out related display or application calculation. According to the seismic processing work flow channel header word data circulation method provided by the invention, a global header word memory management data structure is designed, so that dynamic management recording of a module relationship is realized, circulation of latest header word information of a module in a work flow is realized, and the flow efficiency is improved. And the dynamic change circulation requirement of the processing business process on the header word is effectively met.

The present invention relates to the field of oil and gas field development and applied geophysics, and specifically to a method for dynamic monitoring of oil and gas reservoirs by combining active and passive seismic data. The method comprises the following steps: designing active and passive seismic combined oil and gas reservoir monitoring acquisition parameters according to the target area range of oil and gas reservoir monitoring; continuous acquisition of passive seismic data and phased acquisition of active seismic data; processing each phase of active seismic data; using each phase of active seismic processing results as constraints to process passive seismic data; performing dynamic monitoring and interpretation of oil and gas reservoirs; and adjusting development plans according to the interpretation results. The method of the present invention combines active seismic and passive seismic data to achieve higher-precision monitoring of oil and gas reservoir dynamics, reduce monitoring costs, and make up for the deficiencies of the prior art.

This invention provides a high-resolution intelligent seismic processing method and system incorporating physical information. The processing method includes: acquiring stacked one-dimensional seismic data; preprocessing the one-dimensional seismic data to obtain a training set and estimating the initial non-attenuated wavelet; constructing a deep network model based on the absorption and attenuation characteristics of the one-dimensional seismic data; learning an attenuation compensation operator; and applying the attenuation compensation operator to non-stationary seismic records to obtain stationary seismic records. This method does not require known Q-values ​​and fully utilizes existing well logging data and non-attenuated wavelets, enabling the acquisition of more stable compensation results.

The application discloses a method and equipment for picking up a seismic first arrival wave based on a multi-input attention mechanism network. The application realizes automatic picking up of the first arrival by fusing multi-domain data and the advantages of various networks. Compared with manual picking up, the efficiency is greatly improved and the result is more objective. Compared with a traditional automatic picking up method, the accuracy and stability are higher under different signal-to-noise ratios. Meanwhile, after hyperparameter optimization, the generalization ability is strong, reliable data can be provided for subsequent seismic processing, the advantage is prominent under complex geological conditions, and the method has great significance for geological analysis and resource exploration.

The invention belongs to the technical field of exploration-seismic processing interpretation, and particularly relates to a non-stretching distortion change correction method for split moving. According to the method, pre-stack seismic data to be dynamically corrected and time-speed pairs are input, the wavelength corresponding to each time-speed pair is calculated, the event center track and the dynamic correction value corresponding to each time-speed pair are calculated, dynamic correction value assignment is carried out in the half-wavelength range of the two sides of the event center, and the dynamic correction value is calculated. The method comprises the seven steps of obtaining the dynamic correction values of sampling points in other areas, splitting moving dynamic correction and outputting a non-stretching distortion change correction result, so that sampling points of an event in each seismic channel have the same dynamic correction value, and in the moving process, the sampling points before dynamic correction are split into data after dynamic correction. It is guaranteed that the amplitude of one sampling point is moved only once, and therefore stretching distortion is effectively avoided. The method makes full use of the number of effective channels for superposing the shallow and middle layers, and improves the imaging quality of the superposing shallow and middle layers.

The present disclosure relates to the technical field of seismic processing, and particularly relates to an interpolation method and device based on prior data, equipment, a storage medium and a computer program. The method comprises the following steps: acquiring seismic data, performing τ-P positive conversion on the seismic data in a preset data processing domain to obtain converted data; selecting unaliasing and preset numerical aliasing data in the converted data, and collecting the selected unaliasing and preset numerical aliasing data into to-be-processed data; performing five-dimensional interpolation on the to-be-processed data to obtain an interpolation result; and performing five-dimensional regularization processing on the seismic data according to the interpolation result to obtain a processing result. The present application can adapt to interpolation processing based on prior data.

This invention relates to a pre-stack noise suppression method for seismic data, belonging to the field of seismic exploration technology. The method includes the following steps: 1) After decompiling and defining the observation system of the acquired seismic data, static correction, spherical diffusion compensation, and surface wave suppression are performed sequentially; 2) The first arrival time of the surface wave suppressed seismic data is corrected to a relatively uniform time, and then random noise and residual noise from the surface wave suppression process are suppressed; 3) The first arrival time of the noise-suppressed seismic data is reverse-corrected to restore the first arrival time to its state before correction. This method can effectively suppress shallow noise interference in near-offset data while better preserving the detailed structure of the seismic data, improving the accuracy of seismic processing; and avoids other noise interference introduced by general processing methods that apply strong noise suppression to achieve the purpose of suppressing shallow noise in near-offset data, thus improving the amplitude fidelity of the processing.

The invention discloses an oil-gas-water interface prediction method and system based on a time-lapse earthquake, and belongs to the technical field of development-earthquake processing interpretation. The prediction method comprises the following steps: constructing a three-dimensional porosity model; establishing an original water saturation model, and determining an original oil-gas-water interface height; determining rock mechanical parameters; based on the three-dimensional porosity model, the original water saturation model and the rock mechanical parameters in combination with a time-lapse seismic amplitude difference formula, calculating a residual value of a model time-lapse seismic amplitude difference and an observation time-lapse seismic amplitude difference, and searching for an optimal oil-gas-water interface height change and a porosity value which enable the residual value to be minimum; according to the optimal oil-gas-water interface height change, the porosity value and the original oil-gas-water interface height, current water saturation change is predicted, and remaining oil-gas distribution is obtained. By inverting the change of the height of the oil-gas-water interface before and after the time-lapse earthquake, the water saturation change condition of the target horizon can be obtained at the same time, and the problems that the nonlinear inversion data calculation amount is large and the result is not convergent are effectively solved.

This invention discloses an adaptive multi-scale high-resolution seismic processing method that integrates time-frequency features, belonging to the field of seismic exploration data processing technology. This method constructs a Transformer-based encoder-decoder network architecture, combines continuous wavelet transform to achieve time-frequency feature fusion, and employs an adaptive multi-scale partitioning strategy based on K-Means energy clustering and a dual attention mechanism to capture local and global dependencies in seismic data, ultimately improving the resolution of the seismic data. This invention addresses the over-reliance of traditional methods on geological assumptions and professional knowledge, and the shortcomings of existing deep learning methods in utilizing time-frequency features and multi-scale modeling. It possesses stronger generalization and transfer capabilities, making it suitable for processing seismic data with complex geological structures.

This invention discloses a method for elastic modulus inversion based on diffraction wave information. The method includes: acquiring a three-dimensional seismic data volume of diffraction waves; obtaining near-, middle-, and far-field angularly stacked data of diffraction waves based on the three-dimensional seismic data volume; obtaining a P-wave diffraction coefficient formula suitable for pre-stack inversion based on the P-wave reflection coefficient formula; determining the diffraction coefficient based on the P-wave diffraction coefficient formula; obtaining near-, middle-, and far-field angularly stacked data of the whole wavefield based on layer interpretation; obtaining wavelet data corresponding to near-, middle-, and far-field gathers based on the near-, middle-, and far-field angularly stacked data of the whole wavefield; converting the layer velocity in the whole-wavefield seismic processing into layer velocity P-wave and S-wave modulus and density; using the layer velocity P-wave and S-wave modulus and density as background parameters of the inversion objective function; constructing and inverting the diffraction wave elastic modulus inversion objective function based on the wavelet data and diffraction coefficients; obtaining the elastic modulus anomaly in the target layer caused by heterogeneous disturbance by solving the objective function; and using the elastic modulus anomaly to identify and predict heterogeneous bodies in the strata.

The invention provides an application multistage communication method and device based on the same message body and a medium, and belongs to the field of oil-gas exploration and development. The method comprises the following steps: step 1, establishing a message body; step 2, establishing a message receiving and sending service; and step 3, performing multi-stage communication of the message based on the message body. According to the method, the message body is established, the messages can be transmitted between the applications (or processes) based on the message body, each application (or process) can receive the messages of other applications (or processes) and can also send the messages to other applications (or processes), and multi-stage message communication is formed between the applications (or processes) in the processing and interpreting integrated platform. The usability of the earthquake processing platform is greatly improved, the data processing time is shortened, the data processing timeliness is improved, and the service is provided for exploration and development of conventional oil gas and coal bed gas.

A high-resolution seismic processing method. It mainly solves the problem that the existing seismic processing methods have a narrow frequency range, insufficient resolution and cannot identify thin reservoirs. It includes the following steps: obtaining seismic data, using the well logging acoustic wave velocity curve and density curve, and using the well-seismic correlation coefficient constrained inverse distance weighted algorithm to interpolate and calculate the longitudinal wave impedance model; constructing a high-frequency seismic wavelet, and performing convolution operation with the longitudinal wave impedance model to obtain high-frequency synthetic seismic data; performing frequency fusion calculation with the measured seismic data to obtain broadband seismic data. This high-resolution seismic processing method can obtain broadband and high-resolution seismic data, so that the seismic data can improve the resolution of thin reservoirs with a thickness of more than 3m, and improve the exploration and development effect.

This invention discloses a high-resolution seismic processing method and system, storage medium, and electronic device, applied to the field of seismic data processing. Addressing the problem that existing techniques suffer from poor spectral extension effects and difficulty in identifying thin-layer earthquakes, this invention, based on time-frequency analysis, performs deconvolution, harmonic decomposition, and basis pursuit inversion on the time-frequency spectrum of seismic data. This achieves high-frequency reconstruction of the time-frequency spectrum, improving the resolution of seismic data and forming a novel spectral extension method. Compared to previous spectral recovery methods, this method achieves better spectral extension effects on more complex strata, with higher resolution and signal-to-noise ratio.

The present invention provides a method, device, equipment and storage medium for acquiring surface data, the method comprising: using a controllable vibrator to excite the first end of the source receiving channel to acquire first surface data of the controllable vibrator; using an artificial weight to excite the first end of the source receiving channel to acquire second surface data of the artificial weight; wherein the artificial weight and the controllable vibrator are located at the same first end of the source receiving channel; converting the first surface data into first surface data that conforms to a target format, and converting the second surface data into second surface data that conforms to the target format; based on the first surface data that conforms to the target format and the second surface data in the target format, acquiring first fused surface data corresponding to the first end. The surface data acquired by this method can solve the interference problem of the controllable vibrator on the near-source receiving channel and improve the resolution of seismic processing data.

The invention relates to the technical field of seismic data processing and analysis, in particular to a seismic data storage method and system based on lake-warehouse integration, and the method comprises the steps: collecting seismic data, and constructing a data storage module based on a Hudi frame through a lake-warehouse integrated architecture; designing a random forest-isolated forest hybrid model, inputting the seismic data into the random forest-isolated forest hybrid model, sequentially performing noise suppression, missing value interpolation and abnormal seismic phase detection, and extracting seismic features to obtain seismic processing data; training a random forest-isolated forest hybrid model based on the seismic processing data, and inputting to-be-detected seismic data into the trained random forest-isolated forest hybrid model to obtain a prediction result; storing the seismic processing data and the prediction result to a data storage module; and in combination with a Hudi framework and a Spark Streaming stream processing engine, performing minute-level dynamic rendering on a prediction result, and then performing visual display. And full-link optimization of seismic data from acquisition, cleaning, storage to analysis is realized.

A method of modeling a subsurface area of a seabed using fewer seismic sensors than a seismic sensors that generate sufficient data to permit a desired model of the subsurface area. Seismic datasets are received from the fewer seismic sensors. The fewer seismic sensors are located at points defined in relation to the seabed. Boundary conditions of a seismic wavefield at a fluid-solid interface between the seabed and water above the seabed are identified. An estimation, from the seismic datasets and the boundary conditions, is made of one or more virtual datasets. Each of the virtual datasets includes corresponding estimated data for additional points defined in relation to the seabed. The points and the additional points, combined, are sufficient to permit the desired model of the subsurface area. A subsurface model of the subsurface area is generated by modeling the subsurface area using the seismic datasets and the virtual datasets.

The invention belongs to the technical field of reservoir prediction, and particularly provides a reservoir gas-bearing evaluation method based on a seismic wave elastic parameter difference ratio, and the method comprises the steps: obtaining a seismic processing sub-offset superposition data volume of a target region; well seismic fine calibration and horizon tracking are carried out; performing pre-stack inversion to obtain a transverse wave impedance data body and a Poisson's ratio data body of the target area; calculating a transverse wave impedance difference and a Poisson's ratio difference between the target reservoir and the cover layer by using the transverse wave impedance data body and the Poisson's ratio data body based on horizon calibration; determining an effective closure threshold value of the cover layer; and according to the construction formula, calculating the sealing coefficient of the target reservoir and the cover layer, and according to the effective sealing threshold value of the cover layer and the sealing coefficient of the target reservoir and the cover layer, determining whether the sealing performance is effective or not, and determining the gas-bearing property of the target reservoir. The problem that the gas-bearing property of the unconventional natural gas reservoir cannot be effectively predicted in the prior art is solved, and the effectiveness of the gas-bearing property of the unconventional natural gas reservoir such as limestone and deep coal rock gas can be evaluated.