About This Item

Share This Item

The AAPG/Datapages Combined Publications Database

CSPG Bulletin

Abstract


Bulletin of Canadian Petroleum Geology
Vol. 67 (2019), No. 4. (December), Pages 217-230

Utilizing sedimentary process-based models as training images for multipoint facies simulations

Ingrid Aarnes, Helena van der Vegt, Ragnar Hauge, Bjørn Fjellvoll, Kjetil Nordahl

Abstract

Geostatistical facies modeling algorithms are used in reservoir modeling workflows to create geological models which improve the predictive power of the flow simulation models. In heterogeneous reservoirs, it is of key importance to not only apply statistical techniques, but also incorporate prior geological knowledge. Fluvial dominated deltaic deposits can show a high degree of heterogeneity arising from the interaction of stacking of lobate deposits and the continuous erosion and deposition of the distributary channels while building the delta. To simulate these depositional structures, honouring the physical laws of nature, process-based models can be used to generate synthetic deposits. However, such results are driven by physics and therefore cannot be steered to honour exact well data. We address this challenge by integrating physics-driven process-based models with statistical techniques from MPS. Combining these two different methods is MPS relies on discreet geometric patterns. This is addressed by classifying the process-based model results into discreet facies. A major advantage of this integrated technique is the potential to generate multiple MPS training images through simulation of additional process-based model realizations and we also analyze the effect of using one versus multiple process-based models as input. In this work, we show how the best aspects of both process-based models and MPS modeling can be combined to create improved geological models.


Pay-Per-View Purchase Options

The article is available through a document delivery service. Explain these Purchase Options.

Watermarked PDF Document: $14
Open PDF Document: $24