"Identifying reservoir parameters through well test analysis requires the use of numerous heuristic rules, such as establishing the initial data (time prior to test, flowrates, etc.) and selecting the type of analysis (multiphase, multirate, fracture, etc.), to evaluating the correct data for analysis or matching the best type curve.
This paper describes an application of artificial intelligence through a compilation of facts, rules-of-thumb, and procedures in the form of an interactive computer program, an expert system. Differing from previous work, this program solves well tests in a traditional manner, proceeding forward through the problem from fluid and rock properties, analyzing the pressure and flow rate data, and then conducting the appropriate analyses while employing expertise at every step. Previous works have solved the problem backwards, guessing reservoir parameters until a reasonable match of the pressure test data is found. Programs of the forward type have reflected both extremes: the simpler offers interactive manipulation of the data, while the more complex are "black box" programs. This expert system is between these two extremes.
The system follows the semilog and type curves analyses systematically for any combination of oil, gas (gas pseudo pressure), and/or water flow, multirate, or fracture flow (vertical and dual porosity). Unlike previous works, which offer the options to conduct these procedures separately, the expert system decides which methods should be used and in what order.
The results from this expert system compared well with published or accepted values for several buildup well tests"--Abstract, page iii.
Barr, David J.
Harvey, A. Herbert
Dekock, Arlan R.
Numbere, Daopu Thompson, 1951-
Geosciences and Geological and Petroleum Engineering
Ph. D. in Petroleum Engineering
University of Missouri--Rolla
viii, 103 pages
© 1989 Scott Michael Frailey, All rights reserved.
Dissertation - Restricted Access
Print OCLC #
Electronic OCLC #
Link to Catalog RecordElectronic access to the full-text of this document is restricted to Missouri S&T users. Otherwise, request this publication directly from Missouri S&T Library or contact your local library. http://laurel.lso.missouri.edu/record=b2262648~S5
Frailey, Scott Michael, "A forward thinking expert system in well test analysis" (1989). Doctoral Dissertations. 741.