Capillary pressure correction of cuttings

Abstract

The accurate characterization of capillary pressure is essential in determining multiphase flow behavior in subsurface conditions. It is also essential in quantifying reservoir rock quality, reservoir fluid saturations, and the thickness of the transition zone. Mercury injection has become a routine measurement for capillary pressure characterization, but the existing technology is primitive. For samples with irregular shapes, such as cuttings, unconfined pieces are placed in an empty cell before injection. The raw capillary pressure measurements show unrealistic entry pressure corresponding to filling the empty cell and closing microcracks. This study proposes a simple relation for determining accurate entry pressure. The proposed relation is applied to the actual measurements of seven shale samples, and its performance is improved using k-nearest neighbors (KNN), locally selective combination in parallel outlier ensembles (LSCP), and Savitzky–Golay (SG) filters. The optimal solution is obtained by combining the simple relation with unsupervised machine learning and noise filtering techniques in series. The proposed relation, which is corroborated by high-resolution images, provides a new approach to determining true entry pressure and has applications in characterizing multiphase flow in unconventional formations.