Speaker
Description
The simultaneous determination of relative permeability ($k_r$) and capillary pressure ($P_c$) from UnSteady-State (USS) coreflooding data remains a complex estimation problem. Standard interpretation often relies on single-rate experiments, where the cumulative oil production (NP) and differential pressure ($\Delta P$) data may not contain sufficient information to decouple viscous forces from capillary end-effects. This information deficit exacerbates the ill-posed nature of the problem, leading to significant parameter uncertainty, particularly when estimating capillary pressure without independent experimental data. In this work, one investigates a sequential workflow designed to enhance the information content of USS experiments. The proposed methodology treats the standard single-rate coreflood not as a final result, but as a calibration step used to generate prior estimates of rock properties. Using these preliminary estimates, one constructs a Fisher Information Matrix (FIM) based on the sensitivity of the modeled $\Delta P$ and NP responses to the target parameters ($k_r$ and $P_c$ coefficients). This sensitivity analysis is useful to identify specific time windows where the experimental data is potentially uninformative or dominated by parameter correlation. Guided by the FIM, a secondary multi-rate coreflood experiment is designed for the same rock sample, aiming to target flow conditions that maximize parameter distinctness. One present the theoretical framework for this "calibration-then-optimization" approach and discuss its potential to reduce the uncertainty inherent in legacy single-rate datasets. By explicitly incorporating Optimal Experimental Design (OED) principles, this study seeks to provide a more rigorous basis for acquiring physically consistent relative permeability and capillary pressure curves from dynamic displacement data.
| Country | Brazil |
|---|---|
| Student Awards | I would like to submit this presentation into both awards |
| Acceptance of the Terms & Conditions | Click here to agree |
