... Using a physics-based shear-tensile crack model, we develop a Bayesian approach to simultaneously calculate source mechanisms for a set of microearthquakes and rigorously quantify the uncertainties of model parameters. To that end, we use the normalized displacement amplitudes of direct P-waves...

... Full waveform inversion (FWI) uses all information in seismic data to estimate high-resolution earth model parameters. It is an ill-posed nonlinear inverse problem that is very computationally demanding and is prone to sub-optimal convergence in local minima. In this work, we address...

..., such as those contained in wells. We have developed a method for EFWI that enforces explicit physical relationships between the model parameters. The underlying assumption is that a strong correlation between the desired physical parameters can be observed from existing data. We have examined constraining...

... acquired at the Containment and Monitoring Institute Field Research Station in Newell County, Alberta, Canada. where d contains the dispersion curve data de ned above, m contains The results are consistent with the regional lithology background. the model parameters, namely the shear-wave velocity, density...

... to achieve the spectrum whitening function. Accordingly, the virtual source is computed by summing all of the single-frequency misfits. Based on the adjoint method, the gradients corresponding to model parameters are calculated via frequency-domain gradient formulas, where the phase-sensitive detection (PSD...

... in the multilayered model, which are substituted into the approximation form for the extension. Note that these effective parameters are not the real model parameters but are reminiscent of the acoustic ORT model parameterization that is used only for the modeling computational purpose. These effective parameters...

... tensor gravity gradiometry (FTG) data collected in the Nordkapp Basin in Barents Sea, Norway. stabilizer constable boundary artificial intelligence data collected model parameter upstream oil & gas imaging barent sea reservoir characterization applied geoscience joint inversion...

... to order these patches into a Theory and Method regular smooth sequence, and use a simple smoothing filter to obtain a stable reconstructed result. Eventually, we apply FWI tries to update model parameter through minimizing an the proposed multi-scale full-waveform inversion constrained with patch-ordering...

... of the method. illustrates the effectiveness of this method. Theory Introduction 1. PPDF of model parameter constrained by seismic data Seismic inversion is a significant technology that converts the seismic amplitude to certain elastic properties, such as In Bayesian inversion, the model parameter is assumed...

... that the simultaneous optimizations of multiple Markov chains can achieve the efficient simulation of the posterior probability density distribution of model parameters, which is helpful for the uncertainty analysis of the inversion and sets a theoretical fundament for reservoir characterization. Presentation Date...

... method and reservoir simulator. The prior model parameters are then updated to account for the information available from acoustic impedance changes using an iterative ensemble smoother. The outcome is a new ensemble of changes in pressure-saturation and porosity fields that honors the seismic data...

... any uncertain There has seen a growing interest in enhancing reservoir model parameters can be estimated and various types of characterization through history matching of different measurements can be readily incorporated. Ensemble geophysical measurements (Katterbauer et al., 2016; Liang assimilation...

... parameters using all the information available in seismic data and can provide FWI is a seismic imaging and velocity estimation technique valuable information about subsurface properties. It is a very computer demanding algorithm and subject to local minima that attempts to find a set of model parameters...

... tradeoffs, 1986) extensive computation cost and low convergence rate. xx = ( + 2µ ) vx + vz , (1) An FWI problem is usually solved by adjoint state methods t x z (Plessix, 2006; Liu and Tromp, 2006), in which the gradients to update the model parameters are derived explicitly, and then zz...

... machine learning deep learning model parameter seg international exposition artificial intelligence model misfit iteration reservoir characterization inverse problem magnetotelluric inverse problem inversion algorithm data misfit rnn-based gradient prediction gradient neural network...

... resistivity index model parameter seg international exposition experimental data upstream oil & gas porosity exploration geophysicist 10 reservoir log analysis water saturation formation evaluation determination nuclear magnetic resonance log data well logging geometric mean permeability...

... build a theoretical rock physics model using a conventional RPM workflow. We use Hertz-Mindlin grain contact theory to estimate the moduli of rock frame with optimized model parameters. We then use this rock physics model to create synthetic well logs by perturbing the rock properties, such as lithology...

... inversion covariance matrix exploration geophysicist 10 azimuth upstream oil & gas sayer expanded abstract symmetry hydraulic fracturing vertical fracture phase ii model inversion interface monoclinic avaz inversion reservoir characterization bachrach model parameter seg international...

... Time: 3:05 PM Location: Poster Station 7 Presentation Type: Poster upstream oil & gas model parameter seg international exposition isotropic background medium machine learning coefficient approximation coefficient frazer reservoir characterization approximation geophysics stova...

... model parameter seg international exposition dataset workflow inversion joint inversion daniele colombo Physics-driven deep learning joint inversion Daniele Colombo1, Weichang Li2, Diego Rovetta3, Ernesto Sandoval-Curiel1, Ersan Turkoglu1 1Geophysics Technology, EXPEC Advanced Research Center...