ARBOGAST, JEFF S.¹ and STEVEN M. GOOLSBY^2
1Petroleum Software Technologies, Denver, CO
²Goolsby Brothers and Associates, Inc., Denver, CO

Abstract: Resistivity Modeling and Neural Network Synthetics -- Powerful New Exploration and Development Tools

The most fundamental application of wireline log analysis has been the calculation of formation fluid saturation. This process requires accurate resistivity and porosity information.

Accurate resistivity information has been difficult to obtain because of poor vertical resolution of available logs and invasion and borehole effects on those logs. In the last decade, slow and somewhat cumbersome forward modeling and inversion processing methods have been successfully applied to overcome the limitations of resistivity tools used to measure true formation resistivity (Rt). Very recently, the availability of inexpensive, high-speed PC hardware and newer processing software has vastly reduced the processing time requirements and transformed this technique into a practical solution for multi-well applications.

Complementary to this technique, neural networks can provide high-quality synthetic reservoir information using available data from wireline logs, cores, or cuttings in nearby wells if porosity and permeability information is inaccurate, incomplete, or simply not available.

The combination of these two new methods represents a powerful exploration and development tool. Applications include equity determinations, completion strategies, reservoir modeling, and the identification and evaluation of bypassed pay zones. This presentation demonstrates several successful applications of these techniques.

AAPG Search and Discovery Article #90928©1999 AAPG Annual Convention, San Antonio, Texas