Abstract: Efficient Modeling of the Performance of Non-Conventional Wells in Heterogeneous Reservoirs

DURLOFSKY, LOUIS, Stanford University, Stanford, CA

The use of non-conventional wells (e.g., horizontal, deviated or multilateral wells) is now quite common throughout the industry. However, performance predictions for these wells are often inaccurate. This may be due in some cases to an inadequate representation of reservoir heterogeneity in the simplified models used to estimate well performance. In recent work (Wolfsteiner et al., SPE 56754 and SPE 59399), we developed a semi-analytical approach for modeling non-conventional well performance. The method accounts approximately for the effects of reservoir heterogeneity and wellbore hydraulics and is very efficient compared to performing full finite difference simulations of detailed heterogeneous permeability fields. We refer to our semi-analytical approach as the s-k* method because we represent heterogeneity in terms of a near-well skin (s) and a constant background permeability (k*).

In this paper we apply the method to two distinct problems involving primary production. The first is a detailed assessment of the variation in well performance observed for different geostatistical realizations of the permeability field. We quantify the variation in productivity for horizontal and multilateral wells in permeability fields of prescribed statistics and demonstrate how this variation is affected by the relevant reservoir and wellbore system parameters. Next, in an attempt to capture the effects of near-well heterogeneity, we introduce the skin computed in our semi-analytical method into coarse, homogeneous (of permeability k*) finite difference simulation models of two-phase flow. We show that this approach is highly efficient and reasonably accurate relative to the reference fine grid solution for some flow quantities.

AAPG Search and Discovery Article #90911©2000 AAPG Pacific Section and Western Region Society of Petroleum Engineers, Long Beach, California