Analysis Procedure for Variable-Rate Pressure Drawdown Data
Academic Article
Overview
Identity
Additional Document Info
Other
View All
Overview
abstract
Use of pressure drawdown tests to characterize gas wells has long been accepted, in principle, to be an excellent formation evaluation technique. In practice, however, the theoretical requirement that flow rates must be maintained strictly constant during the test has led to two major problems:the constant flow rate requirement is frequently ignored, and use of conventional analysis on such tests has led to serious error in many cases, leading to a loss of confidence in drawdown testing; orthe constant flow rate is maintained, but only with a great deal of difficulty.Winestock and Colpitts showed that both problems could be avoided if gas-well drawdown tests were run with a fixed choke - usually resulting in a slowly decreasing flow rate during the test - if a simple modification of conventional analysis theory is employed. In fact, Winestock and Colpitts claimed that the method they proposed would be adequate in general for smoothly varying flow rates, even though the total change in flow rate from beginning to end of a test was large. This claim requires justification because the method proposed by Winestock and Colpitts is an approximation, and it clearly must fail when the rate of change in flow rate is sufficiently large. The purpose of our investigations, therefore, was to find the limits of applicability of the Winestock and Colpitts method.First, we reviewed their proposed method and the reason for the method. A frequently used equation describing a pressure drawdown test in a gas well is pressure drawdown test in a gas well is+F (1)In this equation all terms except flowing bottom-hole pressure, Pwf, and test time, t, are assumed to be constant. Winestock Pwf, and test time, t, are assumed to be constant. Winestock and Colpitts point out that use of the equation in this form can lead to serious errors if it is applied to analyze drawdown tests in which flow rates vary even slightly during the test. They propose, however, that for a test with flow rate varying slightly, the test is modeled adequately by a simple rearrangement of Eq. 1:= 712 (2)They propose that the entire group on the left side of Eq. 2, consisting of two variables, bottom-hole flowing pressure, Pwf, and gas production rate qr, both measured pressure, Pwf, and gas production rate qr, both measured during the test, be plotted against the logarithm of test time, t. The permeability thickness product would then be a function of the slope of the resulting line. The turbulence constant, F B, must be obtained before this graph can be prepared.P. 115
