WebWave-equation-based inversion. Thanks to its unmatched ability to resolve CO 2 plumes, active-source time-lapse seismic is arguably the preferred imaging modality when monitoring geological storage (Ringrose 2024).In its simplest form for a single time-lapse vintage, FWI involves minimizing the \(\ell_2\)-norm misfit/loss function between … WebSeismic inversion is generally carried out by iterative data fitting in which the model updates are evaluated by solving the corresponding physics-based forward modeling. Local optimization methods are commonly used for finding an optimal model. Care must be taken to account for the ill posedness of the problem by imposing proper constraints.
Seismic Inversion by Hybrid Machine Learning - NASA/ADS
WebJan 6, 2024 · Deep reflection seismic data are usually accompanied by large-offset data, and the accurate and rapid identification of the first arrivals of seismic records plays an important role in eliminating the effects of topography and other factors that increase with the increasing offsets. In this paper, we propose a method based on convolutional neural … WebJan 7, 2024 · I am a Geophysicist and Data Scientist with 7 years working experience in Mahakam Field. Skilled in seismic interpretation, seismic processing, petroelastic modelling, well correlation, well log interpretation, sedimentology and stratigraphy analysis, velocity modelling, seismic attribute, AVO analysis, Quantitative Interpretation, Rock Physics … files from removable disc
Hybrid and Automated Machine Learning Approaches for Oil …
WebTo mitigate the cycle-skipping problem, Bunks et al. (1995) propose a multiscale inversion approach that initially inverts low-pass-filtered seismic data and then gradually admits … WebSep 29, 2024 · Seismic inversion using a neural network regulariser implemented as an ExternalOperator in Firedrake machine-learning automatic-differentiation autograd partial-differential-equations domain-specific-language seismic-inversion ufl firedrake dolfin-adjoint neural-network-based-regularizer Updated on Feb 3 Python slimgroup / TimeProbeSeismic.jl WebNov 29, 2024 · To resolve those issues, we employ machine-learning techniques to solve the full-waveform inversion. Specifically, we focus on applying convolutional neural network (CNN) to directly derive the inversion operator f-1 so that the velocity structure can be obtained without knowing the forward operator f. files.ggcv.com/download/index.html