Data-driven based investigation of pressure dynamics in underground hydrocarbon reservoirs

The process of storing natural gas in geological formations involves applying pressure to force the into and out porous permeable reservoir. In response extraction/withdrawal storage/injection, reservoir compresses expands as a major consequences fluid pore variations. challenge associated with this type energy systems is learning variations within grid being injected and/or withdrawn. As such, it essential identify realistic model that accounts for at any point time. paper, we present data-driven technique called Dynamic Mode Decomposition (DMD) investigate dynamics an underground hydrocarbon relation injection/storage. For demonstration purpose, first implement using benchmark data Society Petroleum Engineers (SPE1) Comparative Solution Project. Applying DMD field simulated model, show capable approximating average change over time up 99% accuracy. Given depleted reservoirs are already developed reservoirs, conclude could serve reliable tool fast evaluation storage given its low complexity insignificant loss prediction