A sonic tool is activated in a well having multiple casings and annuli surrounding the casing. Detected data is preprocessed using slowness time coherence (STC) processing to obtain STC data. The STC data is provided to a machine learning module which has been trained on labeled STC data. The machine learning module provides an answer product regarding the states of the borehole annuli which may be used to make decision regarding remedial action with respect to the borehole casings. The machine learning module may implement a convolutional neural network (CNN), a support vector machine (SVM), or an auto-encoder.

A method for generating an image of a subsurface based on blended seismic data includes receiving the blended seismic data, which is recorded so that plural traces have uncontaminated parts with different uncontaminated recording time lengths, selecting plural subgroups (SG1, SG2) of traces so that each subgroup (SG1) includes only uncontaminated parts that have a same uncontaminated recording time length, processing the traces from each subgroup to generate processed traces, mapping the processed traces to a same sampling, combining the processed traces from the plural subgroups (SG1, SG2) to generate combined processed traces, and generating an image of a structure of the subsurface based on the combined processed traces.

A sonic logging method is provided that transmits acoustic signals using a high order acoustic source and processes waveform data to identify a set of arrival events and time picks by automatic and/or manual methods. Ray tracing inversion is carried out for each arrival event over a number of possible raypath types that include at least one polarized shear raypath type to determine two-dimensional reflector positions and predicted inclination angles of the arrival event for the possible raypath types. One or more three-dimensional slowness-time coherence representations are generated for the arrival event and raypath type(s) and evaluated to determine azimuth, orientation and raypath type of a corresponding reflector. The method outputs a three-dimensional position and orientation for at least one reflector. The information derived from the method can be conveyed in various displays and plots and structured formats for reservoir understanding and also output for use in reservoir analysis and other applications.

Simultaneous sources are separated with a deblending method wherein the fullband data deblended on a source-by-source basis. The fullband data is first decomposed into multiple subbands, and then a non-equispaced subband Radon transform is used to transform the decomposed data into the Radon domain. The deblending process is solved subband by subband by minimising a nonlinear objective function. The solution of the nonlinear objective function is found using a multi-step procedure.

A sonic tool is activated in a well having multiple casings and annuli surrounding the casing. Detected data is preprocessed using slowness time coherence (STC) processing to obtain STC data. The STC data is provided to a machine learning module which has been trained on labeled STC data. The machine learning module provides an answer product regarding the states of the borehole annuli which may be used to make decision regarding remedial action with respect to the borehole casings. The machine learning module may implement a convolutional neural network (CNN), a support vector machine (SVM), or an auto-encoder.

A method of generating target-oriented acquisition-imprint-free prestack angle gathers using common focus point (CFP) operators includes receiving a plurality of seismic traces associated with a target point in a reservoir. A first angle domain common image gather (ADCIG) is generated based on the received plurality of seismic traces. A plurality of synthetic traces associated with the target point is generated. A second ADCIG is generated based on the synthetic traces. An enhanced ADCIG is generated using the first ADCIG and the second ADCIG.

Post-stack seismic data is received. Transformed seismic data is created from the received post-stack seismic data, including performing a super-resolution radon transform on the post-stack seismic data. Signal and noise regions are separated using the transformed seismic data, including using a defined muting function to remove unwanted noise. An inverse radon transform is performed using the separated signal and noise regions, outputting only signals.

A sonic logging method is provided that transmits acoustic signals using a high order acoustic source and processes waveform data to identify a set of arrival events and time picks by automatic and/or manual methods. Ray tracing inversion is carried out for each arrival event over a number of possible raypath types that include at least one polarized shear raypath type to determine two-dimensional reflector positions and predicted inclination angles of the arrival event for the possible raypath types. One or more three-dimensional slowness-time coherence representations are generated for the arrival event and raypath type(s) and evaluated to determine azimuth, orientation and raypath type of a corresponding reflector. The method outputs a three-dimensional position and orientation for at least one reflector. The information derived from the method can be conveyed in various displays and plots and structured formats for reservoir understanding and also output for use in reservoir analysis and other applications.

Computing device, computer instructions and method for up-down separation of seismic data. The method includes receiving the seismic data, which includes hydrophone data and particle motion data; performing a first up-down separation, which is independent of a ghost model, using as input the hydrophone data and the particle motion data, to obtain first up-down separated data; performing a second up-down separation by using as input a combination of (i) the hydrophone data and/or the particle motion data and (ii) the first up-down separated data, wherein an output of the second up-down separation is second up-down separated data; and generating an image of the subsurface based on the second up-down separated data.

A method of generating target-oriented acquisition-imprint-free prestack angle gathers using common focus point (CFP) operators includes receiving a plurality of seismic traces associated with a target point in a reservoir. A first angle domain common image gather (ADCIG) is generated based on the received plurality of seismic traces. A plurality of synthetic traces associated with the target point is generated. A second ADCIG is generated based on the synthetic traces. An enhanced ADCIG is generated using the first ADCIG and the second ADCIG.