The primary objective of this paper is to demonstrate the practical value of type curve matching, as a complimentary technique to standard diagnostic plots, in the interpretation of production data from complex multi-stage horizontal wells.
The approach is based on the usage of Compound Linear Flow Type Curves (Liang et al, 2012). Our methodology involves identifying the transition from early to late (compound) linear flow that would typically be expected in a multi-stage fractured completion. The "shape" of this transition can be a good indicator of the geometry of the system- specifically degree of stimulation efficiency along the lateral, in relation to the effective fracture half-length. This approach has been applied to several producing wells in the Eagle Ford (nanodarcy reservoir) and Bakken (microdarcy reservoir).
As a result of using CLF type curves, we are able to estimate stimulation efficiency along the lateral, with reasonable confidence. The stimulation efficiency is defined as the effective stimulated horizontal well length divided by the actual horizontal well length. In addition, we are able to estimate stimulated reservoir width in a reasonably objective and consistent way. These results can be used to seed analytical models for subsequent history matching and forecasting. Diagnostic methods currently used for interpreting performance of multi-stage horizontals are based on identifying specific flow regimes using specialized plots (eg- square root time and flowing material balance). These methods are limited, in that they don't consider the transition periods, which in some cases can dominate the production history. Type curve matching is a good compliment to the standard diagnostic plots, providing an independent and objective method of characterizing the reservoir.
Historically, type curve matching has been proven as a valid technique for interpreting production data (Fetkovich, Blasingame, Agarwal et al). However, with the advent of complex multi-stage completions in tight fractured reservoirs, the existing conventional models are no longer useful. The recently introduced CLF type curves are more appropriate for unconventionals. This work represents the first detailed case study where type curve matching is applied to an unconventional oil and gas play.