When last I wrote in 2021, I noted what I saw as the peak of inflated expectations for machine learning in reservoir simulation, with the topic flooding the list of submitted papers. In 2022, we appear to be rebalancing somewhat, with both a lower proportion of papers on the topic and a distinct increase in papers advocating incorporation of physics or ways to leverage the best of both worlds, classical reservoir engineering fundamentals augmented with these newer approaches.
I find this encouraging.
I also find it encouraging that, anecdotally, I’m seeing more people advocating the need to understand. I’m not talking about understanding how to operate a shrink-wrapped tool rather understanding the physics, the algorithms, and indeed the actual concepts we are trying to simulate. Autopopulating depofacies between well-control points in a model—however clever the algorithm may be—is not a substitute for understanding the concept you are trying to represent, and never will be.
Whether it be Derrick Turk asking us to “resist the temptation to accrue vocabulary rather than understanding” or Mark Bentley telling us that “if you can sketch it you can model it,” there does seem to be a growing pushback against the notions that the modeling/simulation process can be successfully shrink-wrapped and that fundamental understanding is increasingly a legacy requirement.
Sometimes that feeling of imposter syndrome is your subconscious telling you to read a book.
I hope that you enjoy this selection of papers.
This Month’s Technical Papers
Forecasting Technique Bridges Gap Between Material Balance and Reservoir Simulation
Physics-Informed Neural Networks Help Predict Fluid Flow in Porous Media
Study Identifies the Seven Wastes of Reservoir Modeling Projects
Recommended Additional Reading
URTEC 208376 A Model Ranking Approach for Liquid-Loading-Onset Predictions by Hao Jia, China University of Petroleum—Beijing, et al.
SPE 203491 Asphaltene Modeling With Cubic and More-Complex Equations of State by Sukit Leekumjorn, Calsep, et al.
SPE 202303 Openness in Reservoir Simulation: Empowering Flexible Field Management by Wentao Zhou, Schlumberger, et al.
Mark Burgoyne, SPE, is a principal reservoir engineer for Santos. He has more than a quarter-century of industry experience, including technical leadership and subsurface management roles with Santos and hydraulic fracturing, cementing, and coiled tubing roles with Schlumberger. Burgoyne holds a bachelor’s degree in chemical engineering, a master’s degree in petroleum engineering, and an MBA degree.