Take yourself to greater depths right from your desktop with SPE Webinars and Online Education.
Join our industry experts as they explore solutions to real problems and discuss trending topics.

Smart Proxy Modeling for Numerical Reservoir Simulations – Big Data Analytics in E&P

Recorded On: 10/06/2015

Computational science, addressing numerical solution to complex multi-physic, non-linear, partial differential equations, is at the forefront of engineering problem solving and optimization. Numerical Reservoir simulation, the application of computational science in petroleum engineering, is computationally expensive. Proxy models (statistical response surfaces, or reduced physics) attempt to make it practical to use the simulation models for field development planning and uncertainty quantification by addressing their computational footprint (with limited success rate).

Data-Driven Smart proxies take advantage of the “Big Data" solutions (machine learning and pattern recognition) to develop highly accurate replicas of numerical models with very fast response time. The novelty of Smart Proxy Modeling stems from the fact that it is a complete departure from traditional approaches to modeling in the oil and gas industry and constitutes a major advancement in utilization and incorporation of Big Data solution in the E&P industry.

Instead of starting with first principle physics, smart proxies are models that are built based on observation of system behavior, through data, much like how human brain learns. Just imagine that a single run of a one-million grid block reservoir simulation model that includes 100 time-steps will generate 1,000,000 x 100 = 1x108 examples of pressure and saturation changes at the grid block level to learn from. Furthermore, only by making 10 simulation runs, the number of training examples will increase to a billion records. A large amount of information and knowledgeis embedded in this one billion example of how pressure and saturation in a reservoir changes as a function of initial and boundary conditions as well as a function of all other static and dynamic characteristics of the reservoir being modeled. Surrogate Reservoir Model (SRM) is the smart proxy of numerical reservoir simulation.

This web event includes:

Introduction to Big Data Analytics in E&P
Introduction to Numerical Reservoir Simulation
Description of Smart Proxy Model
Surrogate Reservoir Model (Smart Proxy of Reservoir Simulations)
Case Studies (Production optimization in Carbonates, CO2 Storage, History Matching)

SPE Webinars are FREE to members courtesy of the

Shahab Mohaghegh


Shahab D. Mohaghegh, a pioneer in the application of Artificial Intelligence and Data Mining in the Exploration and Production industry, is the president and CEO of Intelligent Solutions, Inc. (ISI) and Professor of Petroleum and Natural Gas Engineering at West Virginia University. He holds B.S., MS, and PhD degrees in petroleum and natural gas engineering

He has authored more than 150 technical papers and carried out more than 50 projects for NOCs and IOCs. He is a SPE Distinguished Lecturer and has been featured, four times, in the Distinguished Author Series of SPE's Journal of Petroleum Technology (JPT). He is the founder of SPE's Petroleum Data-Driven Analytics Technical Section that focuses on the application of AI and data mining in the upstream. He has been honored by the U.S. Secretary of Energy for his technical contribution in the aftermath of the Deepwater Horizon (Macondo) incident in the Gulf of Mexico and was a member of U.S. Secretary of Energy's Technical Advisory Committee on Unconventional Resources (2008-2014). He represents the United States in the International Standard Organization (ISO) on Carbon Capture and Storage.

SPE Webinars are FREE to members courtesy of the


Web Event
10/06/2015 at 9:00 AM (EDT)   |  90 minutes
10/06/2015 at 9:00 AM (EDT)   |  90 minutes Scheduled for 90 minutes.
CEU Credit
0.15 CEU credits  |  Certificate available
0.15 CEU credits  |  Certificate available CEU Credit