The Effect of Producing Time on Type Curve Analysis

Publisher Website
Google Scholar
Abstract
This paper explores the consequences of the assumption that the producingtime is long compared to the longest shut-in time used in type curveanalysis. The fractured-well type curves of Gringarten et al. and thewellbore-storage type curves of Agarwal et al. are examined in detail.New type curves are presented. Applications of the new type curves tofield data are discussed. Introduction In 1970, Agarwal et al. 1 introduced the idea oflog-log type curve matching to analyze the wellborestorage effect in a well with an annular skin regionadjacent to the sandface. In the same year, Rameydemonstrated that log-log graphs can be used toidentify wellbore storage effects, acidized wells, andfractured wells, since each of these conditionsproduces a specific trace at early or short times. produces a specific trace at early or short times. Ramey defined short-time data as that obtainedbefore the conventional semilog straight line.Since 1970, other type curves have been presented.Of importance to early-time analysis are the wellborestorage type curves of Agarwal et al., McKinley, and Earlougher and Kersch and the fractured-welltype curves of Gringarten et al., Cinco et al., andArgarwal et al. Applications of the solutions toshort-time data are documented in Refs. 2, 5, 6, and10 through 14.Except for McKinley, all have examined onlydrawdown solutions. McKinley examined onlybuildup solutions. However, since he assumed thatthe producing time was much larger than the longestbuildup time, and since the application of drawdowntype curves to pressure buildup or falloff data isbased on the assumption that the longest builduptime is small compared with the producing timebefore shut-in, his results are essentially the same asthose in Refs. 1 and 6.*The assumption that the longest buildup time isshort compared with the producing time beforeshut-in implies that if a well is not produced for asufficient period then the drawdown solutions cannot beused. Unfortunately, many are not aware that thisrequirement exists. Moreover, very little quantitativedocumentation is available to indicate theconsequences of this assumption.The objectives of this paper are (1) to examinecritically the effect of producing time on the log-logtype curve procedure, (2) to establish criteria for useof the drawdown type curves, (3) to discuss errorsthat will result if proper care is not taken in analyzingdata when producing times are small, (4) todemonstrate the use of the findings of this paper bypresenting field case illustrations, and (5) to delineate presenting field case illustrations, and (5) to delineate procedures to be followed to account for producing procedures to be followed to account for producing time. For brevity, only the fractured-well type curvesof Gringarten el al. and the wellbore-storage typecurves of Agarwal et al. will be considered. JPT P. 1053

This publication has 0 references indexed in Scilit: