CGG SERVICES SAS

Methods and devices according to various embodiments perform reliable automated core-to-log depth matching. A core-to-log depth matching method includes: (i) pairing a core segment of core property values obtained by analyzing rock samples extracted at core depths from a well and a log segment of log property values obtained from measurements performed with a logging tool at log depths in the well, (ii) for each pair, determining a block-shifting of the core segment relative to the log segment so as to maximize a similarity metric between the core property values and the log property values, and (iii) performing an affine warping based on a tuned set of hyperparameters to adjust length of at least one of core segments.

A method for inverting seismic data in elastic full waveform inversion (FWI) includes receiving recorded seismic data; calculating a wavefield with an attenuated interface wave u s based on an attenuation term that describes a wave attenuation along a travelling direction, wherein the attenuated interface wave u s corresponds to a wave traveling along a low-velocity guide or along an interface between two different media and the travelling direction is aligned with the interface; calculating an updated velocity model using (1) the wavefield corresponding to the attenuated interface wave u s and (2) a body wave wavefield u b ; and generating an image of a subsurface based on the updated velocity model of the subsurface.

A device and method for determining surface displacement in a structure based on interferometric synthetic aperture radar (InSAR) includes receiving a stack of wrapped interferograms, performing multilooking on the stack of wrapped interferograms to generate multilooked interferograms, calculating phase closure residual maps based on the multilooked interferograms, calculating average phase closure residual maps based on the phase closure residuals, masking pixels in the multilooked interferograms, based on the average phase closure residual maps, to obtain masked interferograms, unwrapping the masked interferograms to obtain unwrapped, masked interferograms, and generating an InSAR time series product based on the unwrapped, masked interferograms.

A method for extracting and displaying desired information from a set of tables includes storing a set of tables including information associated with a subsurface of the Earth; receiving a user query; selecting a table from the set of tables based on an embedding search performed for the user query, on a vector database of table-column questions of the set of tables; selecting one or more columns from the table based on a likelihood estimation performed in an embedding space, between (1) the user query and (2) a table summary and descriptions of columns for the selected table; determining one or more rows associated with the one or more columns; and displaying one or more answers in response to the user query.

A correlated sparse nodes and mini-streamers system for collecting seismic data includes plural nodes distributed on the ocean bottom, and a mini-streamer spread that includes plural mini-streamers. The plural nodes and the mini-streamer spread are configured to simultaneously collect seismic data from a surveyed subsurface, and wherein a length of the mini-streamers is equal to or less than three times an inline distance between adjacent nodes of the plural nodes.

Methods and seismic data processing apparatuses use a down-going annihilation operator to generate an image from seismic data acquired over a water-covered subsurface formation. The down-going annihilation operator is derived using a down-going wavefield and an estimated water-wave extracted from the seismic data. The down-going annihilation operator may be derived in plane-wave domain.

A quality control method and apparatus generate maps for cycle skipping analysis of velocity models that may be the input or the output of a full waveform inversion. A comparison of observed data and corresponding synthetic data generated using a candidate velocity model yields feature images of shots. The feature images are then grouped in clusters using unsupervised machine learning. Deviation of the feature images from an ideal non-cycle skipped image allows to measure the quality of the candidate velocity model.

A support for directing a fluid flow of a dielectric oil through a server includes an upper deck, a central channel, a cooling channel, a lip, and a channel orifice. The upper deck extends in a longitudinal plane and provides an abutment surface for a connection face of the server. The central channel is formed as a first opening of the upper deck that contains the server. The cooling channel is formed as a second opening of the upper deck, and houses a conduit and a radiator of a heat exchanger. The lip protrudes from the upper deck, and directs the fluid flow of the dielectric oil from the central channel into the cooling channel. Further, the channel orifice fluidly connects the cooling channel to the central channel, and the dielectric oil is transferred from the cooling channel into the central channel through the channel orifice.

A method for raw seismic data inversion includes receiving raw seismic data acquired over an underground formation, receiving an initial velocity model of the underground formation, performing a dynamic resolution full waveform inversion, DR-FWI, on the raw seismic data so that a tomography component of a velocity gradient G is compensated to generate a compensated tomographic component while a diving wave component and a migration component are preserved, outputting an updated velocity model based on an illumination compensated velocity gradient G′, which is calculated based on the compensated tomographic component), and locating natural resources in the underground formation using the updated velocity model updated by the DR-FWI.

Methods of seismic data processing employ neural networks and use a reflectivity image based on the acquired seismic data to generate output training datasets. The neural networks thus trained are used for generating production datasets, without ghosts, source effects, multiples and/or populating a predetermined set of bins in inline-crossline plane for a set of offset classes.