English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

The simulation of geological facies in an unobservable volume is essential invarious geoscience applications. Given the complexity of the problem, deepgenerative learning is a promising approach to overcome the limitations oftraditional geostatistical simulation models, in particular their lack ofphysical realism. This research aims to investigate the application ofgenerative adversarial networks and deep variational inference forconditionally simulating meandering channels in underground volumes. In thispaper, we review the generative deep learning approaches, in particular theadversarial ones and the stabilization techniques that aim to facilitate theirtraining. The proposed approach is tested on 2D and 3D simulations generated bythe stochastic process-based model Flumy. Morphological metrics are utilized tocompare our proposed method with earlier iterations of generative adversarialnetworks. The results indicate that by utilizing recent stabilizationtechniques, generative adversarial networks can efficiently sample from targetdata distributions. Moreover, we demonstrate the ability to simulateconditioned simulations through the latent variable model property of theproposed approach.

A stable deep adversarial learning approach for geological facies generation