ABSTRACT
Experimental and numerical studies investigating the validity of the prediction of reservoir matrix blocks performance from water-oil imbibition tests on small samples are reported.
The reliability of the numerical predictive method is confirmed. It was checked against a complete set of laboratory imbibitions tests on homogeneous synthetic samples and heterogeneous outcrop cores for water-wet conditions. The ability to simulate a wide range of blocks heights and several different experimental boundary conditions allowing counterflow is satisfactorily verified.
Experimental results highlight the effects of heterogeneity on oil recovery rate. Oil production cannot be matched when the simulation does not take the heterogeneity into account. Any attempt to use a modified capillary pressure curve in fitting experimental data can lead to large errors in recovery rate prediction. When spatial distributions of petrophysical parameters are introduced in the simulation, good agreement is obtained with the laboratory results. Finally, guidelines are proposed for scaling-up the imbibition process from laboratory to the field.
