Analysis of well test data from gas condensate reservoirs using single-phase dry gas methods: guidelines and examples

Abstract

The "problem" with modelling the performance of a well in a gas condensate reservoir system is that for a truly rigorous solution we require a compositional mass balance on the entire system. In particular, the pseudofunction transformations (pseudopressure and pseudotime) are not only functions of pressure, but are also functions of saturation, where saturation is uniquely tied to the compositional material balance of the system--and saturation (even near the well) can not be obtained at a sufficient accuracy to solve this problem in full detail. Our goal is to establish that the single-phase "dry gas" analog can generally be used to estimate formation properties from well test in gas condensate reservoirs. This is not a trivial issue--using the single-phase solution is much easier, we need not account for saturation with this approach. This thesis will provide example analyses and establish practical guidelines for the analysis and interpretation of well test data in gas condensate reservoirs, using the single-phase "dry gas" approach. We use the typical semilog (radial flow) and log-log (type curve) analysis techniques. The rate history is properly incorporated into the time (or pseudotime) functions using the Agarwall "effective" time function or Homer 2 "effective" time function. In addition to traditional well test analysis (various Cartesian, semilog, and log-log plots), we also simulated the entire test sequence (i.e., all data) using the van Everdingen-Meyer approach.(3,4)