Traditional reservoir modeling is based on our understanding of physics of fluid flow in the porous media. Complex nature and large variety of hydrocarbon reservoirs continuously challenge our assumptions during the modeling process. Mature assets are usually blessed with massive amounts and large variety of data that have been collected throughout the years. In many cases, they are the manifestation of “Big Data” in the exploration and production industry.
In this webinar a novel approach to reservoir modeling that is based on measured data is presented. This technology that has been named “Top-Down Modeling – TDM” integrates fundamentals of reservoir and production engineering with latest advances in machine learning and predictive analytics. It is a formalized, comprehensive, empirical, and multi-variant, reservoir model, developed solely based on field measurements (logs, cores, well tests, seismic, etc.) and historical production/injection data.
Application of this technology to a prolific mature giant oilfield in the Middle East producing from multiple horizons with production data going back to mid-1970s is presented. Periphery water injection in this filed started in mid-1980s. The field includes more than 400 producers and injectors. The production wells are deviated or horizontal and have been completed in multiple formations.
The constructed reservoir model is conditioned to all available field data (measurements) such as production and injection history, well configurations, well-head pressure, well logs, core analysis, time-lapse saturation logs, well interventions and well tests. The well tests were used to estimates the static pressure of the field as a function of time. Time-lapse saturation logs were available from large number of wells indicating the state of water saturation in multiple locations in the field at different times.
A full field TDM was trained and history matched by machine learning technology using data from 1975 to 2001. The history matched model was deployed in predictive mode to generate (forecast) production from 2002 to 2010 and the results were compared with historical production (Blind History Match). Finally future production from the field was forecasted. The main challenge was to simultaneously history match static reservoir pressure, water saturation and production rates (constraining well-head pressure) for all the wells in the field. The model was used to identify infill locations and water injection schedule in this field.
Shahab D. Mohaghegh
Intelligent Solutions, Inc. & West Virginia University
Shahab D. Mohaghegh is the president and CEO of Intelligent Solutions, Inc. (ISI) and Professor of Petroleum and Natural Gas Engineering at West Virginia University. A pioneer in the application of Artificial Intelligence and Data Mining in the Exploration and Production industry, 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 many of them with major international companies. He is a SPE Distinguished Lecturer and has been featured in the Distinguished Author Series of SPE’s Journal of Petroleum Technology (JPT) four times. He is the program chair of Petroleum Data-Driven Analytics, SPE’s Technical Section dedicated to data mining. 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 is a member of U.S. Secretary of Energy’s Technical Advisory Committee on Unconventional Resources. Shahab is the designated U.S. liaison (WG4) representing the Unites States in the International Organization for Standardization (ISO) for CO2 capture and storage.