Well Test Analysis In Tight Gas Reservoirs

Abstract

Abstract The subject paper has two main purposes. The first is to describe several interesting field case histories of well tests in tight gas reservoirs. These are given in sufficient detail to act as benchmark examples for comparison of similar analyses in the future by other investigators. The second purpose is to demonstrate the effective application of a simulation model to complement a conventional well test analysis. The use of a single well gas simulation model to characterize the properties of gas wells in tight reservoirs is described. In one case, a long term isochronal test was carried out where none of the flow rates could be stabilized. Conventional analysis is quite arbitrary in this situation. One of the best alternatives was to use a simulation model. In the second case, the simulation model was used to check the results of a suspect conventional analysis. The integration of the two approaches led to a much better understanding of the problem. The simulation model is briefly described. The field examples are given in detail. Some of the pitfalls of using simulation models in this type of pitfalls of using simulation models in this type of analysis are pointed out as well as some of the advantages. Introduction In recent years, with the cost of natural gas going up and the supply going down, much more attention has been focused on producing gas from marginal "tight" gas reservoirs which were formerly passed over as noncommercial deposits. A "tight" gas reservoir is simply a reservoir in a low permeability rock, probably in the range of 10 millidarcies or less. There is no real distinction between a tight gas reservoir and any other kind. The gas is the same, the rocks look similar and the gas flow follows the same physical principles. The real difference is in the mind of the beholder. The same well tests that work very nicely in a 1000 millidarcy permeability 1000 ft thick sandstone may be a little more difficult to perform in a 1 millidarcy 20 ft thick carbonate reservoir. The same test will most likely work, but it will take much more engineering, analysis and patience to get a good test. The flow may be dominated by wellbore effects, turbulence, natural or induced fracture effects, or any number of things that do not cause any real problems in the analysis of high capacity wells. This study concerns itself with describing the application of a single well gas simulation model as the primary tool of well test analysis, as an aid to conventional analyses or as a check on conventional analyses in tight gas reservoirs. Two field studies are described in detail and all the necessary data are given so that the studies can act as a "benchmark" for use by other investigators. STATEMENT OF THE PROBLEM Gas well tests of various types are performed to characterize the formation around the wellbore, to determine the wellbore damage, to define drainage limits, to determine well potential deliverability, and for numerous other reasons. Much has been written about the analyses of these tests using various analytical techniques. For the examples presented in this paper, the conventional well test analysis techniques proved to be inadequate or could be used with only limited confidence. Conventional pressure buildup analysis was attempted for both examples.