Disclosed herein is a method of stripping a strong reflection layer based on deep learning. The method establishes a direct mapping relationship between a strong reflection signal and seismic data of a target work area through a nonlinear mapping function of the deep neural network, and strips a strong reflection layer after the strong layer is accurately predicted. A mapping relationship between the seismic data containing the strong reflection layer and an event of the strong reflection layer is directedly found through training parameters. In addition, this method does not require an empirical parameter adjustment, and only needs to prepare a training sample that meets the actual conditions of the target work area according to the described rules.

Computing systems, computer-readable media, and methods for providing multiple computer-generated seismic data interpretation options, of which the method includes receiving a training input, sorting the training input into a first group and a second group, subgrouping the second group into a plurality of subgroups, generating a plurality of trained models based on the plurality of subgroups and the first group, receiving a prediction input having a set of data to be interpreted, generating a plurality of interpretation options for the prediction input by applying the plurality of training models to the prediction input, and outputting the plurality of interpretation options.

A performance-level seismic motion hazard analysis method includes: (1) extracting seismic motion data and denoising the data; (2) extracting feature parameters from the data, and carrying out initialization; (3) generating a training set, an interval set and a test set; (4) training a multi-layer neural network based on the training set; (5) training output values of the neural network based on the interval set, and calculating a mean and a standard deviation of relative errors of the output values; (6) training the neural network based on the test set to determine output values, and calculating a magnitude interval based on an interval confidence; (7) carrying out probabilistic seismic hazard analysis to determine an annual exceeding probability and a return period of a performance-level seismic motion; and (8) determining a magnitude and an epicentral distance that reach the performance-level seismic motion based on the performance-level seismic motion and consistent probability.

A method, computing system, and non-transitory computer-readable medium, of which the method includes receiving offset well data collected while drilling one or more offset wells, generating a machine learning model configured to predict drilling risks from drilling measurements or inferences, based on the offset well data, receiving drilling parameters for a new well, determining that the drilling parameters are within an engineering design window, generating a drilling risk profile for the new well using the machine learning model, and adjusting one or more of the drilling parameters for the new well, after determining the drilling parameters are within the engineering design window, and after determining the drilling risk profile, based on the drilling risk profile.

Systems and methods of placing wells in a hydrocarbon field based on seismic attributes and quality indicators associated with a subterranean formation of the hydrocarbon field can include receiving seismic attributes representing the subterranean formation and seismic data quality indicators. A cutoff is generated for each seismic attribute and seismic data quality indicator. A weight is assigned to each seismic attribute and seismic data quality indicator. The weighted seismic attributes and data quality indicators are aggregated for each location in the hydrocarbon field. A risk ranking is assigned based on the weighted seismic attributes and data quality indicators associated with each location in the hydrocarbon field based on the cutoffs. A map is generated with each location on the surface of the subterranean formation color-coded based on its assigned risk ranking.

A system and method that includes querying a database to obtain offset well data collected while drilling previously drilled wells. The system and method also include determining if at least one risk is identified with respect to a planned well based on the offset well data. The system and method additionally include generating a machine learning model based on the at least one risk that is identified based on the offset well data. The system and method further include predicting at least one drilling risk based on the machine learning model, wherein a drill plan that includes drilling parameters is adjusted based on the at least one predicted drilling risk.

A method for training a backpropagation-enabled segmentation process is used for identifying an occurrence of a sub-surface feature. A multi-dimensional seismic data set with an input dimension of at least two is inputted into a backpropagation-enabled process. A prediction of the occurrence of the subsurface feature has a prediction dimension of at least 1 and is at least 1 dimension less than the input dimension.

In a method to transform logs, an acoustic logging tool inserted into a borehole includes a source and an array of receiver stations. Each station includes a receiver spaced along the tool from the source by successively greater distances. In the method, the source emits energy (I) to cause the propagation towards the stations of plural signals exhibiting paths characteristic of first and second respective modes and (II) to stimulate a receiver of each station to generate an output signal per station that indicates the signal packets and represents the modes in combination with one another. In the method, the output signals are transformed into transformed signals containing phase/amplitude information of each mode. The phase/amplitude are linked by an operator to the slowness and attenuation of the mode and the transmitter-receiver distance of the station. The phase/amplitude are used to extract slowness and attenuation information for each mode.

Methods and devices for seismic exploration of an underground formation including an orthorhombic anisotropic medium or a tilted transverse isotropic medium are provided. Isotropic-type processing techniques use effective elastic parameter values calculated based on elastic parameter values, anisotropy parameter values and azimuth angle values for the orthorhombic anisotropic medium. For the tilted transverse isotropic medium, the effective elastic parameter values depend also on the tilt angle thereof.

A seismic observation device includes: a waveform acquisition unit that acquires waveform data for a predetermined period including an observation start time of a P wave; a delay time specifying unit that inputs the waveform data to a trained model and acquires, from the trained model, a delay time from the observation start time of the P wave to an observation start time of an S wave; and an observation time estimation unit that estimates the observation start time of the S wave based on the observation start time of the P wave and the delay time.