Skip to main content Skip to secondary navigation
Main content start
Postdoctoral Scholar

Suihong Song

Postdoctoral Scholar, Energy Resources Engineering
Suihong Song collaborates with Professor Tapan Mukerji at the Stanford Center for Earth Resources Forecast (SCERF) as a postdoctoral scholar. His research is centered on integrating machine learning with geosciences, specifically focusing on machine learning-based reservoir characterization and geomodelling, Physics-informed Neural Networks (PINNs) and neural operators as well as their applications in porous flow simulations, neural networks-based surrogate and inversion, decision-making under uncertainty, and machine learning-based geological interpretation of well logs and seismic data. These research endeavors have practical applications in managing underground water resources, oil and gas exploration, geological storage of CO2, and the evaluation of hydrothermal and natural hydrogen, among others.Song proposed GANSim, an abbreviation for Generative Adversarial Networks-based reservoir simulation, which presents a reservoir geomodelling workflow. This innovative approach has been successfully implemented in various 3D field reservoirs by international oil companies, including ExxonMobil.

Education

Doctor of Philosophy, China University of Petroleum (2021)
Ph.D, China University of Petroleum-Beijing, Geological Resources and Geological Engineering (2021)
Visiting Ph.D., Stanford University, Energy Sciences & Engineering (2020)
M.Eng., China University of Petroleum-Beijing, Geological Resources and Geological Engineering (2017)
B.Eng., China University of Petroleum-Beijing, Petroleum Geology (2013)