When test in g a production well to determine the characteristics of a fluid filled reservoir, one usually waits until wellbore storage is over, and then one determines both the slope of the downhole pressure versus log (time) plot ( to calculate kh/u), and the intercept of the line ( to obtain @chre2). However, in a geothermal field it may not always be possible to run a test for a sufficiently long time to insure an accurate measurement of these parameters. The testing of a geothermal field requires instrumentation that can withstand high temperatures and high salinities, and, at present, available instrumentation is limited. Another problem is that non-isothermal effects in the bore increase the time of wellbore storage. The slow heating of the fluid in the well results in a slight change in slope of the p versus log t plot, and the duration of this heating effect can be much longer than wellbore storage due to pressure changes alone. in addition, the slope of the p versus log t graph can be very flat because of the large values of kh/v in geothermal fields. With a positive skin effect wellbore storage lasts longer, so the slope will be even flatter in this pseudo-steady region.

Very small changes in pressure must be measured over long times requiring accurate instrumentation. It is desirable to be able to use the pressure transient data taken while wellbore storage is important. The transient test can be relatively short (say 20 minutes) and the changes in pressure are still large enough ( say on the order of a psi/minute) so that the error in the measurements because of the accuracy and resolution of the pressure gauge is small. The pressure data taken at early times can be used i f the response of the wellbore is eliminated from the well test data, and a variable flowrate pressure transient analysis is performed. By modeling the transient flow in the well, it is also possible to explain differences between the pressure transient data from a geothermal field and that of an oil field.

The response of the flow in the wellbore is eliminated by calculating the actual conditions at the sandface (well/reservoir boundary). Given the sandface flow and enthalpy and knowing the downhole pressure, one can use a variable well test analysis method to determine kh/v and @chre*. reservoir, a variable well test method that uses a minimization technique is available (Benson and McEdwards, 1980). For a two-phase reservoir, a method of analyzing transient flow data is more difficult than from a single phase reservoir for even a constant mass flowrate. However, one could use a numerical simulator and try to match the pressure data by varying the reservoir parameters until a "best fit" is obtained.

A numerical model of transient , one-dimensional two-phase flow in a well has been developed ( Miller, 1979). This model is used to simulate the wellbore flow in the calculations below. Numerous steady state wellbore flow models have been reported (Gould, 1974; Nathenson, 1974; Sugiura and Farouq, 1979). However, in such models, one must naturally have the flow into the bore equal to the flow out of the bore. Therefore, these models cannot be used to obtain the sandface flowrate when wellbore storage is important.

The main purpose of this work is to show that it is possible to calculate the sandface flow rate given wellhead conditions and the downhole pressure transients. It is not necessary to know anything about the reservoir itself. first, it is of interest to look at the nonuniform pressure changes in the well , and to illustrate nonisothermal effects on pressure transient data.

Press the Back button in your browser, or search again.

Copyright 1980, Stanford Geothermal Program: Readers who download papers from this site should honor the copyright of the original authors and may not copy or distribute the work further without the permission of the original publisher.