... Corrections for source and receiver positioning and clock-drift errors are early steps in the processing of ocean bottom seismic (OBS) data. We have implemented an optimization process to correct jointly for these errors, while a water velocity model gets updated from an initial uniform estimate...

... We developed a method to predict ocean-water temperature and salinity from water-column seismic reflections. In our approach, we first use a prestack waveform inversion with genetic algorithm (GA) optimization and estimate the oceanwater sound-speed from seismic data. Then, from the UNESCO...

... accelerated convex optimization algorithm, which applies an accelerated augmented Lagrangian to perform a convex optimization of the objective function. The joint TRSVD and truncated nuclear norm technique reduce the computational effort of FWI while maintaining adequate information. By applying truncated...

... states government artificial intelligence exploration geophysicist optimization geologist deep learning machine learning maintenance shoreline iot operator automated asset performance optimization climate change american association international methane emission monitoring cloud energy...

... an optimization-based workflow implemented as neural network training was developed to construct an RGT cube that can satisfy the trace-to-trace correlations which is captured by flow field and seismic amplitude. This workflow does not rely on interpreter label input. A public dataset has been tested...

... is developed for shale gas production performance analysis. First, a machine learning model is built as the surrogate of shale gas production. Then, Shapley values based on game theory principles are employed on top of the surrogate model for key factor analysis. Finally, bi-objective optimization that jointly...

... artificial intelligence algorithm deep learning machine learning configuration boundary reflection international upstream oil & gas optimization workflow optimization algorithm minima optimization workflow applied geoscience american association wavefield snapshot exploration...

... Microseismic data has become a routine tool to help understand hydraulic completions. This data is often used to optimize production and better understand the natural and induced fractures in and around the reservoir. Additionally, well-constrained moment tensors provide information about how...

... image improvement can be achieved. rtm optimization frequency attenuation reservoir characterization hz rtm migration expanded abstract upstream oil & gas international enhancement american association subsalt imaging applied geoscience exploration geophysicist gulf basement...

... achieving human level-performance in both the adaptive subtraction and quality control (QC). We studied the noise in labels for supervised training and provided a series of strategies to overcome the challenges. We selected Bayesian optimization as an engine to search the highdimensional hyper-parameter...

... automatically in isolating diffracted multiples in Radon domain. This method first computes an apex-shifted Radon transform using a L0 optimization that gives the sparsest Radon coefficients. Those sparse coefficients provide a guide for the time-space localization of multiples. Using the focused energy from...

.... The amount of uncertainty is also a larger function of assumptions made to convert the underdetermined problems to an overdetermined one. In this work, we present some practical advantage of using multimineral analysis using non-linear optimization algorithms, which are based on broad classes of stochastic...

... at every location. To address the challenge, we proposed a semi-supervised learning approach with a hybrid optimizer, i.e., Genetic-evolutionary adaptive moment estimation (G-ADAM) for acoustic impedance estimation from seismic data. In this approach, reflectivity and seismic loss functions are used...

... is described. We take advantage of the introduce physics constraints in composite loss functions to domains of local optimization and of the machine learning act as regularization mechanisms toward nonphysical (ML) or deep learning (DL) technique to generate efficient solutions for the derivation...

...) or the use of speci cally matitsch and Tromp, 1999), which handles surface topography amplitude-sensitive mis t functions (Pan and Wang, 2020). extremely well. For our optimization, we use the parameters , Cycle-skipping, non-linearity, and the non-convexity of the VS, VP, Q 1 and Q µ 1. The variation...

... Accurate rainfall estimation is critical to modeling and predicting landslides. Often sparse rain gauge networks need to be expanded for credible precipitation inputs. One of the chief optimization problems is to minimize the estimation errors from spatial interpolation of rain gauge values...

... skylake node reid atcheson exploration geophysicist 10 numerical algorithm group upstream oil & gas performance comparison optimization seg international exposition kernel reservoir characterization seismic application nec corporation summary application problem size nec...

...A set of seismic technology for optimizing horizontal well fracturing in tight sandstone gas reservoir Wang Yijun* ,DuYubin ,Guo Yabin ,Zhang Linke ,Gao Yang ,Chen Dahong ,Wang peng BGP,CNPC Changqing Oilfield, CNPC Summary sandstone gas reservoir determine that the inevitable choice...

... reservoir characterization is also based on 3D sedimentary characteristics of uranium reservoir by the seismic, drilling, logging and other geological data, and extraction and optimization of seismic attribute. Then, the involves steps including well seismic calibration, structural lithological...

... nonlinear optimization problem where a genetic algorithm (a type of algorithm in the artificial intelligence spectrum) replaces the time-consuming, manual trial-and-error process of adjusting properties and fitting the input logs in conventional multimineral analysis. The method requires interpretative...