PROGRAM TOPIC: CARBON MANAGEMENT (CO_2 SEQUESTRATION) Modeling analysis of CO_2 Sequestration in Saline Formation Using Artificial Intelligence Technology

摘要

One of the most domineering environmental issues is the increase in atmospheric carbon dioxide (CO_2) concentration ensuing from anthropogenic sources. Sequestration in geological formations is one of the proposed solutions for removing greenhouse emissions from the atmosphere. Since aquifers are considered to be most widely available, there is high potential to find a suitable aquifer with large capacity or close to CO_2 source. The structure and the interconnection of the pores provide flow of gases or fluids through the bed and all these factors make aquifers the second largest, naturally occurring potential store for CO_2. Numerical reservoir simulators are conventionally used to build models of the CO_2 Sequestration process. The sequestration project deals with a wide range of uncertainties. Any comprehensive study or uncertainty analysis of the representative numerical reservoir models would be tedious and time consuming requiring high computational costs. Therefore, comprehensive analysis of such models is quite impractical. This work presents a new artificial intelligence base technique known as Surrogate Reservoir Model (SRM) that can mimic the behavior of the commercial reservoir model with high accuracy in fractions of a second. Application of SRM to Mattoon field, located in the eastern three quarters of section 8 of Mattoon Township, Coles County, IL, is presented in this article. Upon validation of SRM Key Performance Indicators (KPIs) of the simulation model are identified to help reservoir engineers concentrate on the most influential parameters on the model's output when studying the reservoir and performing uncertainty analysis. These indicators can be used so as to build a spatiotemporal model which can deliver dynamic properties such as pressure, water saturation and CO_2 mole fraction at each particular location of the reservoir in a specific time.