A method based on theoretical considerations is presented for estimating stabilized performance of gas wells from data obtained from short-term flow tests. Calculated results are presented for 11 gas wells evaluated with four- to five-hour flow tests. For seven wells on which comparison was available, satisfactory agreement between measured and calculated performance was obtained. Although the final results are favorable, differences in flow capacity k(g)h obtained from flow tests conducted at different rates were observed. This inconsistency did not seriously affect the calculated results in the test wells.
In order to negotiate the sale of gas from newly drilled gas wells and to plan otherwise for the orderly exploitation of gas reserves, it is necessary to predict the long-term withdrawal rates physically possible for each well. Since low permeability gas wells do not stabilize except after a lengthy period of flow, the operator is unable, without wasting large amounts of gas, to measure directly the stabilized performance prior to contracting for sale of the gas and connecting the well to a pipeline. This problem is not new, but is one which has recently received considerable attention because of the rising importance of natural gas in the economy and new emphasis on the development of gas reserves. Thus, a method is needed for estimating stabilized performance from information obtained from short-flow tests in which stabilized flow is not attained. Such short tests would be suitable for wells not connected to a pipeline since the tests require only small amounts of time, labor and production of gas. Other requirements of such a method are short calculation time and use of a model of the gas well and is reservoir drainage area which is as consistent as possible with present theoretical, experimental and field knowledge. The method presented herein was developed in an effort to meet these requirements. It has been named the "two-flow method".In addition to the section on the development and description of the method, a description of field application of the method and a discussion of well test results are included. Finally, there is a section on an additional consideration prompted by some of the test results. The Appendix contains certain details of the derivation of the two-flow method and a sample calculation for Test Well 1.
The method is based upon the following assumptions:
The system can be described as a finite cylindrical gas reservoir produced by a single cylindrical well located at the center of the reservoir. Flow is strictly radial.
Flow obeys Darcy's law except very near the well where a quadratic flow law is assumed.
The system is also characterized by a laminar flow skin factor.
The use of a quadratic or turbulent flow term is more fully justified in Refs. 1 and 2.When such a system is opened at the wellbore to constant rate flow from an initial condition of uniform pressure distribution, the radius or drainage quickly passes through the nondarcy flow and "skin effect" region immediately surrounding the well. After the drainage radius or "transient" has passed beyond this region and before it reaches the external boundary, drawdown can be expressed by
( ) = [ ] + Bq ......................................(1)
After the drainage radius reaches the external boundary, stabilized flow exists and drawdown is given by
( ) = [ ] ...................................(2)
Eqs. 1 and 2 form the basis for the method. The validity of these equations will now be examined in the light of some well test data previously available. It has been found by actual measurements on gas wells that the relationship between drawdown and flow rate, that is the performance curve, is closely approximated by
q = C ( ) ...............................(3)
Eq. 3 is a straight line on log-log graph paper.