... zero-type regularization lawi information depth reservoir characterization inversion time geophysics upstream oil & gas distance gabor deconvolution international robust localized adaptive waveform inversion american association iteration Robust localized adaptive waveform...

... way to make CSEM appealing for 4D programs would be to increase value of information and reduce costs by performing operations where seismic and CSEM are acquired in the course of the same survey and at comparable data density. To that end, we are proposing to develop a next-step Multiphysics...

... We introduce a power-law distribution to describe the fault characteristics of the prior information in the geosteering inverse problem. For two faulted examples in the Marcellus shale, we retrospectively compared our computed geosteering results to the original, manual interpretations...

... Interpretation tasks of subsurface data, for example in the form of seismic, logs, and their derivatives, have been automated to some extent with deep learning solutions such as automated fault interpretation. These solutions extract information (e.g. geologic objects and their attributes) from...

... the complexity of field data inversion. The first inverts for an effective source field that addresses complex near surface wavefield propagation and incomplete source signature information. The second is a special data driven parameterization that prevents cross-talk in the inverted models. production...

... characterization upstream oil & gas neural network deep learning hydrocarbon machine learning exploration geophysicist prospect offshore australia energy 10 american association expression prospect identification information australia applied geoscience society hydrocarbon migration...

... activities become more desirable. Although they appeared random, ambient seismic signals contain rich information about the underground. Cross-correlation based methods rely on the coherent stack of noise recordings from stationary sources, and extract the empirical Green’s function for further analysis...

...-level hazard that can be relatively easily mitigated through appropriate decision making during operational planning and/or design. Key to making these decisions is the availability of information that allows the hazard to be quantified. In this paper, I look at ways in which the hazard can be better...

... SmartTensors (https://github.com/SmartTensors) is a novel framework for unsupervised and physics-informed machine learning for geoscience applications. The methods in SmartTensors AI platform are developed using advanced matrix/tensor factorization constrained by penalties enforcing robustness...

... provided a seamless 3D-high resolution KDM image across the original two seismic surveys with clear definition of major and minor faults and well-preserved amplitude and azimuth information for improved reservoir characterization, leading to more accurate and detailed delineations of Bahrah reserves...

... multi-parameter fwi information inversion exploration geophysicist artificial intelligence evolutionary algorithm international da silva woodside energy ltd energy 10 semi-global multi-parameter fwi waveform inversion american association applied geoscience umpleby implementation cloud...

... source rock. The mineral composition was also investigated using cuttings from the exploration well for information related to potential reservoirs. Additionally, a seismic interpretation is made around the exploration well area to look for potential traps and closures. Some closures were identified...

... but was not presented at IMAGE 2022 in Houston, Texas. boundary applied geoscience fusion characteristic information international structural subdivision boundary information small-scale boundary information reservoir characterization american association energy 10 amplitude upstream oil & gas...

... machine learning information neural network international exploration geophysicist accuracy american association energy 10 expanded abstract applied geoscience society upstream oil & gas artificial intelligence time-velocity sequence picking comparison heap detection summary...

..., we integrate low-frequency eFWI models, facies information from wells, and least-squares reverse time migration (LSRTM) imaged stacks to constrain high-frequency, absolute-band inversions. structural geology reservoir characterization information inversion machine learning exploration...

... CCS storage sites fast. To do this, accessing regional quality broadband seismic information would be a significant move in that direction. An integrated G&G workflow has been developed and implemented over a proof-of-concept (PoC) area initially considering two aspects of the CCS storage...

...; 2) the learned model has good generalization abilities for divergent geological subsurface formations; and 3) the framework can work with nonconforming discretization used by different physical models. In addition, prior information can be easily integrated into the framework during the training...

... diffractions, providing interpreters with detailed structural and stratigraphic information. diffraction gmdf algorithm imaging algorithm diffraction imaging decomposition international geophysical prospecting reservoir characterization geophysics upstream oil & gas exploration...

... sufficient information about 3D geology. This difficulty can be alleviated by using our newly developed ergodic sampling using irregular samples. Ergodic sampling theory uses optimization to form sparse irregular patterns that maximize the information contained in the same number of samples. A specific...

... information to generate synthetic data from the subsurface AI distribution. We also add Bayesian layers to the first stage of the network to evaluate the model errors. The proposed probabilistic approach to deep learning allows one to estimate the uncertainty of the obtained parameters, which enhances...