Next-Generation Interpretation Workflows

Abstract

Artificial intelligence (AI) and data-driven innovation impacted multiple industries throughout the previous decade; however, the oil and gas industry has yet to take full advantage of these technologies and increase their operational efficiencies. Applying various AI-based solutions across the exploration and appraisal lifecycle is discussed and how these solutions collectively can revolutionize the current method of building subsurface models. Subsurface models provide important information about quantifying the potential exploration risk, evaluating prospects, and calculating reserve estimates and recoverable reserves; therefore, it has a direct impact on revenues, cash flows, and capital expenditures for any exploration and production (E&P) operator. However, current interpretation and static modeling techniques are often repetitive, time consuming, computationally intense, and subject to interpreter bias. Research across multiple domains (geosciences, petrophysics, and reservoir engineering) has successfully demonstrated that AI combined with domain expertise and powered by cloud computing can help build more accurate and faster structural frameworks, property models, and reservoir models, causing improved and real-time decision making. Well log and seismic data interpretations can be semi-automated, causing at least ten times faster interpretations for the entire seismic volume and well data rather than constraining interpretations to only the zone of interest. Similarly, machine learning-embedded property modeling techniques prove to be approximately 20% more accurate and more than seven times faster than the geostatistical approach. Overall, these solutions overcome some of the limitations of conventional techniques used during the previous few decades without compromising the quality of models.

AAPG Datapages/Search and Discovery Article #90350 © 2019 AAPG Annual Convention and Exhibition, San Antonio, Texas, May 19-22, 2019