7 results about "Seismic processing" patented technology

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.

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 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 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.

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.