Data Conditioning for Subsurface Models with Single-Image Generative Adversarial Network (SinGAN)

초록

The characterization of subsurface models relies on the accuracy of the models, necessitating the integration of a large amount of information across different sources through model conditioning, such as data conditioning and geological concept conditioning. Conventional geostatistical models demand a trade-off between honoring geological conditioning (i.e., qualitative geological concepts) and data conditioning (i.e., quantitative static data and dynamic data). To resolve this issue, generative artificial intelligence (AI) methods, such as generative adversarial networks (GAN), have been widely applied for subsurface modeling due to their ability to reproduce complex geological patterns. However, the current practices of data conditioning in GANs conduct quality assessment through visual inspection to check model plausibility or some preliminary quantitative analysis of the distribution of the properties of interest. We propose the generative AI realization minimum acceptance criteria for data conditioning, demonstrated with a single-image GAN. Our conditioning checks include static small-scale local and large-scale exhaustive data conditioning checks, local uncertainty, and spatial non-stationarity reproduction checks. We also check conditioning to geological concepts through multi-scale spatial distribution, the number of connected geobodies, the spatial continuity check, and the model facies proportion reproduction check. Our proposed workflow provides guidance on the conditioning of deep learning methods for subsurface modeling and enhanced model conditioning checking essential for applying these models to support uncertainty characterization and decision-making.

키워드