Dynamic Modeling Technology
One of critical issue in the most of the subsurface studies is how to define the top of the transition zones in the dynamic models, which are direct linked to the shape of lab experimental PC's curves that might not reflect the actual field production tests data. And since the lab experimental PC's curves are always representing thick transition zone definition as initialized in the dynamic models, reservoir simulation engineers are forced to apply nonphysical adjustment to slow-down water movement by introduce unphysical Swcr’s which is pure simulation keyword, or/and using very low Krw’s values, and/or applying unsupported permeability multiplier.
Nevertheless, in order to approve new concept that can be applied to redesign PC’s curves shape while honoring original model fluids distribution and history match profiles, a mechanistic model i.e. (extracted from full field model) has been utilized with only saturation tables allow to be changed between the two cases i.e. (between the original model vs the modify model).
Basically, the results are impressing and showed great match in most of the profiles although the original model is reflecting very thick transition zone, and the modify model is reflecting very thin transition zone.
The main question is: why do we need to interduce thin transition zone definition in the dynamic model if both the original and the modify models showed acceptable history match profiles? The answer is by having thick transition zone define in the dynamic model, that mean all the future infill wells will be located almost at the top of the structure far away from the transition zone in order to prolong their production plateau before water reach them, while having thin transition zone define in the dynamic model, this shall interduce thick oil interval i.e. (thick oil column) that will give more flexibility to deeping the future infill in the reservoir to maximize recovery.
Principal Reservoir Engineer (Simulation)
6 年In this work, we didn’t change anything’s from the original model, we toke the original model as its and we change only the saturation tables, so the modify model it is same as the original model with all include input data. Beside, we got excellent history match after few iterations, in order redesign and implement new Pc’s and Kr’s curves is toke 1 hr timeframe.