64 results about "Earthquake map" patented technology

The invention provides a three-dimensional seismic fault identification method based on double-attention multi-scale fusion U-Net. The method comprises the following specific steps: constructing a U-shaped network architecture comprising an encoder, a decoder and jump connection; a constructed multi-scale feature fusion module is embedded in the first level of the encoder, and the extraction capability of fault features of different scales is enhanced through a multi-branch structure; introducing the constructed hole fusion modules into the second and third levels of the encoder, designing and expanding a receptive field by using multiple expansion rates, and capturing fault structures of different scales; a double-attention parallel mechanism is integrated in jump connection, and the sensitivity of channel attention and space attention to fault features is improved; constructing a combined loss function; and finally, performing three-dimensional seismic data training and reasoning based on the optimized model to realize high-precision fault identification. The method has high generalization and accuracy, and especially has good performance in the aspect of seismic image fault identification containing a large fault scale span.

A 3D inversion method of DC method using seismic data constraints is disclosed, comprising processing of seismic data and inversion processing of resistivity data; by utilizing seismic wave interval velocity information to constrain the high-density electrical inversion, which can play a complementary role to a certain extent, it is beneficial to supplement the resolution of resistivity inversion with 3D seismic image information, reflect a boundary of subsurface electrical anomalous body, and further characterizes the boundary characteristics of anomalous body through comprehensive geological-geophysical analysis. Seismic data constraints are applied to 3D resistivity inversion, performing numerical simulation and inversion of electrical data under undulating terrain using the finite volume method and Gauss-Newton inversion method. Furthermore, the seismic anisotropic velocity tensor constraint is introduced for inversion to mitigate the non-uniqueness inherent in single-method inversion, yielding a more accurate 3D subsurface electrical structure and improving the resolution of longitudinal boundaries.

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.

The invention belongs to the technical field of seismic exploration and artificial intelligence, and relates to a seismic data high-resolution reconstruction and random noise suppression method, which comprises the following steps of: performing feature extraction and reconstruction on low-resolution noisy seismic data through a classic generator to generate a high-resolution denoised seismic image; and discriminant learning is carried out on real seismic data and a generation result through a quantum enhanced double-path discriminator, and the reconstruction capability of the generator is continuously optimized. According to the quantum enhanced double-path structure designed by the invention, the modeling capability of the model on complex seismic signals can be improved, the reconstructed image is finer, the structure is clearer, the robustness of the network on random noise can be enhanced, and more effective noise suppression is realized. According to the invention, the residual learning mechanism enables the feature interaction between the layers to be more sufficient, and facilitates the transmission and fusion of the seismic features at different layers, thereby improving the overall reconstruction quality. According to the method, the suppression effect on random noise can be remarkably improved while high-resolution reconstruction of seismic data is realized.

Methods and systems are disclosed for automatically integrating subsurface structural maps with strike and dip information measured in subsurface wells. The method includes obtaining a seismic image volume for a subsurface region of interest and a well log for each of a plurality of wellbores penetrating the subsurface region of interest. Further, the method includes determining a seismic map of a geological surface from the seismic image volume, wherein the seismic map comprises an estimated depth and an estimated vector normal to the seismic map at a plurality of horizontal locations and determining an intersection point for each of the plurality of wellbores with the geological surface. Additionally, the method includes forming a cost function based, at least in part, on the seismic map and the intersection points of the plurality of wellbores and constructing a subsurface map by solving a constrained optimization problem based on the cost function.

The invention belongs to the field of seismic data processing and geological modeling, and particularly discloses a seismic image super-resolution reconstruction method based on a multi-scale double-discriminator GAN, and the method comprises the following steps: constructing a synthetic seismic data set, and carrying out the preprocessing; a multi-scale double-discriminator generative adversarial network framework is constructed, the framework comprises a generator and a double-discriminator system, the generator adopts a network structure based on multi-scale residual groups, and each multi-scale residual group comprises parallel multi-branch expansion convolution, an enhanced residual dense block and an improved convolution attention mechanism; the double-discriminator system is composed of a structure discriminator and a frequency domain discriminator, and the authenticity of a spatial geologic structure and the physical rationality of a frequency domain signal are restrained respectively. The generator generates a high-resolution seismic image by minimizing a joint loss function of adversarial loss, reconstruction consistency loss and perception loss. According to the method, the full-scale geologic features can be accurately captured while noise interference is suppressed, and the reconstruction precision of complex structures such as thin layers and faults is improved.

The application discloses a kind of multi-information fusion's speed modeling method and device, the method includes, after seismic migration, the imaging data A of acquisition, data phase control type mask B, seismic structure interpretation horizon and fault data C, well logging velocity curve data D and seismic velocity field data E;Edge feature of imaging data A is extracted by edge detection technique, and the structural skeleton data F of seismic image is obtained;Phase control type mask B is fused with structural skeleton data F, and phase control structure guide data G is obtained;Seismic structure interpretation data C is added in phase control structure guide data G, and space interpolation guide data H is formed;Through two-step interpolation extrapolation, the space interpolation distance weight field Q corresponding to space interpolation guide data H and well logging velocity interpolation filling model P are obtained;Classification mask M is established for well logging velocity interpolation filling model P;Two-step filling is carried out to class mask M, and multi-information fusion model R is obtained.

A system and methods are disclosed. The method includes obtaining a seismic dataset including a plurality of recorded multiple events, generating a predicted multiple model using a multiple prediction method and the seismic dataset, and estimating a set of initial matching filters using a matching method, to match the plurality of estimated and recorded multiple events. The method further includes generating a tensor field based on the predicted multiple model, determining a set of structure-oriented matching filters based on the set of initial matching filters and the tensor field, generating a filtered multiple model based on the predicted multiple model and the set of structure-oriented matching filters, and generating a multiple-attenuated seismic dataset based on the filtered multiple model and the seismic dataset, forming a seismic image based, at least in part, on the multiple-attenuated seismic dataset, and determining a location of a hydrocarbon reservoir based on the seismic image.

The invention relates to a seismic data reconstruction method based on dual-domain feature collaborative enhancement, and the method comprises the following steps: obtaining an original seismic data set, carrying out the size normalization and down-sampling processing, and obtaining a lacked-channel seismic data set; the missing channel seismic data set comprises a complete seismic image and missing channel seismic data corresponding to the complete seismic image and missing distance, the missing channel seismic image serves as input of the dual-domain feature-based collaborative enhancement neural network, and the corresponding complete seismic image serves as a label; the method comprises the following steps: constructing a double-domain feature collaborative enhancement neural network, taking a UNet class model as a trunk architecture, performing multiple times of down-sampling on lacked-channel seismic data in the Unet class model through nested connection, and embedding a double-domain feature collaborative enhancement module in each layer of jump connection part; and training the network by using the missing-channel seismic data set, and performing missing-channel seismic data reconstruction by using the trained double-domain feature collaborative enhancement neural network. And high-efficiency reconstruction of a large-area missing region in complex seismic data is realized.

A method may include obtaining seismic data regarding a geological region of interest. The seismic data may correspond to a seismic survey that is divided into various bins in a predetermined bin grid. The method may further include determining a first multiblock bin within the seismic survey. The first multiblock bin may correspond to a source bin and a receiver bin among the bins. The method may further include determining traveltime table data using the seismic data and various multiblock bins that include the first multiblock bin. The method further includes determining migrated data using the seismic data, the traveltime table data, a velocity model, a migration function, and various parallel processors. The method further includes generating a seismic image of the geological region of interest using the migrated data.

The invention belongs to the technical field of geophysics and surveying and mapping science, and discloses an earthquake and level settlement association method and system based on spatial overlay analysis, and the method comprises the following steps: S1, obtaining level measurement data and earthquake directory data; s2, constructing a level line buffer area according to the leveling data; s3, processing the leveling data to obtain level settlement data, and generating a level settlement change point layer according to the level settlement data; s4, performing spatial interpolation according to the level settlement change point diagram layer and the level line buffer area to obtain a settlement change thermodynamic diagram; s5, generating a hierarchical symbolization seismic map layer according to the seismic directory data; s6, the settlement change thermodynamic diagram and the hierarchical symbolization seismic map layer are displayed in an overlapped mode, and an earthquake and level settlement correlation analysis result is obtained. The method disclosed by the invention intuitively reveals and evaluates the spatial correlation between the vertical deformation field and the seismic activity, and provides a brand new technical tool for the conjoint analysis of the earthquake and the level settlement.

A method for predicting a property of a subsurface formation, includes: drilling one or more wellbores into the subsurface formation at a well location; conducting a seismic survey at the well location to construct a primary seismic image volume, the primary seismic image volume dataset comprises at least a pressure wave velocity vp, a shear velocity vs, and a density rho, in both time and frequency domains; conducting well logging at the one or more wellbores to generate logging information at the well location; generating an auxiliary seismic image volume for seismic-guided property prediction using Kriging at the well location; generating a prediction operator based on the primary seismic image volume, the auxiliary seismic image volume and the well logging information; feeding a seismic image volume and the prediction operator into a property prediction engine; and generating the predicted property of the subsurface formation.

This invention provides a method and system for automatic vertex picking of DAS-based microseismic events, relating to the field of microseismic location technology. The method includes acquiring multi-channel data corresponding to microseismic events and performing demodulation analysis to form multi-channel waveform data; performing correlation analysis on the waveform data from different channels to establish an automatic picking correlation model; performing image transformation based on the automatic picking correlation model to form corresponding grayscale images; and performing location discrimination analysis based on the grayscale images to determine the event vertex channel. This method improves the interpretation accuracy and reliability of geological analysis of seismic images by removing noise while preserving geological structural features such as faults and strata.

System methods for generating a seismic image of a subterranean feature is provided. The method includes receiving an input gather representing the subterranean feature. The method further includes applying a pseudo offset header to the input gather to generate a pseudo-offset gather. Finally, the seismic image of the subterranean feature is generated by performing domain processing of the pseudo-offset gather. The method results in an improvement of structural imaging of obscured areas.

Aspects of the technology described herein identify geologic features within seismic data using modern computer analysis. An initial step is the development of training data for the machine classifier. The training data comprises an image of seismic data paired with a label identifying points of interest that the classifier should identify within raw data. Once the training data is generated, a classifier can be trained to identify areas of interest in unlabeled seismic images. The classifier can take the form of a deep neural network, such as a U-net. Aspects of the technology described herein utilize a deep neural network architecture that is optimized to detect broad and flat features in seismic images that may go undetected by typical neural networks in use. The architecture can include a group of layers that perform aspect ratio compression and simultaneous comparison of images across multiple aspect ratio scales.

The application discloses a salt-underground seismic image quality evaluation method and device, equipment and a storage medium. The salt-underground seismic image quality evaluation method comprises the following steps: acquiring a salt-underground seismic image to be evaluated, and uniformly setting at least two quality detection points in the salt-underground seismic image; determining a linear quality score relationship between an expert score result of a quality detection point and an optimal image attribute based on the expert score result of the quality detection point and an attribute value of the optimal image attribute corresponding to the quality detection point; and determining a predicted expert score result of the quality detection point based on the linear quality score relationship, and performing quality grade evaluation on the salt-underground seismic image to be evaluated. Through the technical solution, objective, rapid and accurate evaluation of a large amount of seismic images is realized.

A tangible, non-transitory, machine-readable media, includes instructions configured to cause a processor to determine a residual associated with input seismic data received from a seismic source. The residual is indicative of a difference between expected input seismic data and the input seismic data, and wherein the input seismic data is configured to be combed with an expanded window such that the expanded window comprises data generated by an earlier seismic source excitation and received before a time of a seismic source excitation that generated an input seismic trace corresponding to the input seismic data. The instructions are also configured to cause the processor to determine a deblended output based at least in part on the residual. In addition, the instructions are configured to cause a processor to update the deblended output based at least in part on a result from performing one or more recovery operations configured to recover coherent signals from non-coherent signals of the deblended output. The coherent signals comprise a matching parameter. Further, the instructions are configured to cause a processor to filter the deblended output to remove a portion of the deblended output that is before the time of the seismic source excitation or before a predicted earliest arrival time of a seismic wave travelling from the seismic source to a receiver, to generate an improved deblended output comprising less noise than the deblended output. Still further, the instructions configured to cause a processor to transmit the filtered deblended output for use in generating a seismic image. The seismic image represents hydrocarbons in a subsurface region of Earth or subsurface drilling hazards.

The present invention discloses a method and system for automatically generating sea breeze geological modeling horizons based on image recognition. The method first analyzes seismic images obtained from engineering geophysical surveys to obtain seismic horizon data. Then, based on the seismic horizon data, the coring depth of the station is determined, and coring is performed to obtain geotechnical data. The seismic horizon data is updated based on the geotechnical data, and then coring is guided by the updated seismic horizon data. This process is repeated until the comprehensive stratigraphic similarity between the geotechnical data and the seismic horizon data exceeds a set threshold. Through the interaction between coring and seismic horizon data, iterative updates of the seismic horizon data are achieved, ultimately obtaining accurate stratigraphic horizon data.

The invention discloses an OVT domain self-supervision seismic data reconstruction method and device, electronic equipment and a medium. The method comprises the following steps: taking seismic data with data missing as an original data set, and establishing a training data set, a verification data set and a test data set; constructing a convolutional neural network, training the convolutional neural network through the training data set, and performing verification and testing through the verification data set and the test data set to obtain a final convolutional neural network; and inputting seismic data to be reconstructed into the final convolutional neural network to obtain reconstructed OVT domain data. According to the method, the seismic data are converted into the OVT domain to serve as network input, the seismic image features of the OVT domain are extracted, the ability of the network in learning different information of the seismic data is enhanced, local detail features can be well reconstructed while main information is reserved, the reconstruction quality of the seismic data is improved, and the method is suitable for being used for the field of seismic data reconstruction. And the method is of great significance to high-precision seismic data reconstruction processing.

A method for producing a high-fidelity, high-resolution seismic image of a subsurface of a wellsite. The method includes receiving seismic data from wellsite equipment that is disposed at a wellsite. An initial estimation of an earth model is recovered from the received data and a depth migration workflow is performed that is based on the initial estimation of the earth model. The depth migration workflow may be a least-squares FWI-derived reflectively (LS-FDR) workflow. The method also includes revising the initial estimation of the earth model based on the results of the depth migration workflow in order to produce an optimal estimation of the earth model. A seismic image may then be generated from the optimal estimation of the earth model and displayed on a screen for a user. The user may then perform a wellsite action that is based on the generated seismic image.

Systems and methods are disclosed. The method may include receiving observed seismic data, and a first and a second seismic velocity model, each pertaining to a subterranean region of interest and, iteratively, determining synthetic reflection data based on the first and second seismic velocity model, determining traveltime shift data between the synthetic observed seismic data, and determining warped observed seismic data by applying the traveltime shift data to the observed seismic data. The method further includes determining conditioned traveltime shift data using local similarity based on shaping regularization from the synthetic reflection data, the warped observed seismic data, and the traveltime shift data, and determining a seismic velocity model based on the first seismic velocity model and the conditioned traveltime shift data. The method also includes determining a seismic image from observed seismic data and the seismic velocity model, and a location of a hydrocarbon reservoir using the seismic image.

A computer-implemented method for processing a seismic image obtained from seismic measurements performed on a geological formation includes determining a relative geological time (RGT) image based on the seismic image; obtaining setpoints for the RGT image; retrieving an RGT surface of the RGT image for each of the setpoints, wherein each point of an RGT surface is defined by a vertical position and at least one horizontal position; and determining a reference RGT surface by combining the RGT surfaces determined for the setpoints. The vertical position of a point of the reference RGT surface is obtained by combining the vertical positions of the points of the RGT surfaces having the same at least one horizontal position as the considered point of the reference RGT surface. The combining of the vertical positions uses weighting coefficients.

A computer-implemented method for determining modeled seismic data of a subsurface region includes receiving observed seismic data from the subsurface region captured by one or more seismic receivers as one or more reflected seismic signals; determining, based on a velocity model of the subsurface region and the observed seismic data, a traveltime function associated with the subsurface region; migrating, using the traveltime function and a Kirchhoff migration operator, the observed seismic data to produce a migrated seismic image; and performing reverse-time demigration on the migrated seismic image, using the velocity model and a solved full wave-equation, to produce modeled seismic data.

A computer-implemented method for determining modeled seismic data of a subsurface region includes receiving observed seismic data from the subsurface region captured by one or more seismic receivers as one or more reflected seismic signals; determining, based on a velocity model of the subsurface region and the observed seismic data, a traveltime function associated with the subsurface region; migrating, using the traveltime function and a Kirchhoff migration operator, the observed seismic data to produce a migrated seismic image; and performing reverse-time demigration on the migrated seismic image, using the velocity model and a solved full wave-equation, to produce modeled seismic data.

This invention relates to a desert earthquake noise suppression method based on a multi-scale attention interaction network, belonging to the fields of machine learning and seismic image processing. Addressing the problem of low-frequency noise suppression in desert earthquake images, this invention proposes a multi-scale attention interaction network. This network first downsamples the input seismic data, then adaptively learns the complex features of the multi-scale seismic data using dual-branch convolutional layers with different kernel sizes. It then employs multiple attention mechanisms, including permutation attention and coordinate attention, to integrate and interact with multi-scale features across channel, spatial, and coordinate dimensions to obtain continuous seismic signals. This method uses a hybrid loss function of mean squared error and mean cosine similarity, combining the multi-scale strategy with the attention mechanism to improve the denoising effect of desert earthquake images. Compared with single-scale denoising convolutional neural networks, this invention significantly suppresses desert noise and outperforms single-scale denoising convolutional neural networks in restoring the continuity of the phase axis.

A method (600) and a system (1000) for generating an adaptive migration taper for a pre-stack seismic dataset are disclosed. The method (600) includes obtaining the pre-stack seismic dataset (602) and a seismic velocity model of a subterranean region (604). The method (600) also includes generating the adaptive migration taper based, at least in part, on the pre-stack seismic dataset (606), and forming a migrated seismic image using a migration function, the seismic velocity model, the pre-stack seismic dataset, and the adaptive migration taper (608). The method (600) further includes determining a location of a hydrocarbon reservoir based, at least in part, on the migrated seismic image (610).

The invention provides a step-by-step frequency expanding method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring a pre-stack CRP gather; performing interpretive CRP gather superposition on the pre-stack CRP gather to obtain superposed seismic data; performing spectrum equalization processing of a scale domain on the superposed seismic data to obtain a reconstructed seismic image; carrying out smooth filtering processing on the reconstructed seismic image by adopting anisotropic diffusion filtering; selecting a scale range with a high signal-to-noise ratio in the reconstructed seismic image after smooth filtering processing to extrapolate other scale segments to obtain seismic data of a plurality of scale segments; and performing signal reconstruction on the seismic data of all the scale segments to obtain broadband seismic data. According to the method, the relative effective frequency band of seismic data can be remarkably widened, stratum details are well depicted, the resolution of the seismic data is effectively improved, and the method has high industrial practical value and application prospects in oil and gas seismic exploration.

The invention belongs to the field of seismic data processing and geological modeling, and particularly discloses a seismic image super-resolution method and system based on a residual feature distillation generative adversarial network, and the method comprises the following steps: carrying out the preprocessing of paired high-resolution seismic image and low-resolution seismic image samples; a generative adversarial network framework is constructed and comprises a generator and a discriminator, the generator adopts an enhanced residual characteristic distillation network, and the enhanced residual characteristic distillation network is formed by cascading a plurality of enhanced residual characteristic distillation blocks. Each enhanced residual feature distillation block comprises a residual feature distillation mechanism and a parallel channel attention and space attention mechanism; a generator and a discriminator are optimized in an alternate training mode, and the generator generates a high-resolution seismic image by minimizing a joint loss function of pixel reconstruction loss, perception loss and adversarial loss. According to the method, the reconstruction precision of the seismic image can be effectively improved while the model parameter quantity and the calculation quantity are reduced.

A computer-implemented method for detecting geological elements or fluid in a subsurface from seismic images is disclosed. Seismic data may be analyzed to identify one or both of fluid or geologic elements in the subsurface. As one example, the analysis may include unsupervised learning, such as variational machine learning, in order to learn relationships between different sets of seismic data. For example, variational machine learning may be used to learn relationships among partially-stack images or among pre-stack images in order to detect hydrocarbon presence. In this way, an unsupervised learning framework may be used for learning a Direct Hydrocarbon Indicator (DHI) from seismic images by learning relationships among partially-stack or pre-stack images.

The application provides a seismic wave impedance inversion method and system based on deep learning, comprising the following steps: S1, constructing and preprocessing a data set; S2, constructing and training a DA-TransUNet model; extracting features of an input seismic image through a convolutional neural network of the DA-TransUNet model, extracting local features, and generating skip connection features; extracting global features through a Transformer network based on the local features; S3, performing upsampling processing through the convolutional neural network of the DA-TransUNet model, fusing the skip connection features with features of a current layer of a decoder, and obtaining a fusion result as an upsampling result; and S4, constructing a final seismic image prediction result. Through the DA-TransUNet network architecture, the application solves the problems of insufficient precision and low efficiency of traditional seismic wave impedance inversion technology in processing complex geological data.