This paper documents the development of an expert system for screening well
s that could be drilled underbalanced, and for aiding in the preliminary se
lection of appropriate underbalanced drilling fluids for a given range of w
ellbore and reservoir conditions. This approach combines a qualitative rule
-based analysis for assessing formation damage and lost circulation potenti
als with quantitative analysis for assessing wellbore stability using geome
chanical and petrophysical data. To make the analysis complete, a variety o
f other factors such as pipe sticking potential, wellbore geometry, type of
fluid influx anticipated, pore pressure value, and cost benefit are also i
ncluded in the expert-system decision trees. The main advantage of the expe
rt system, developed in this study, is the use of fuzzy logic for handling
cases that lend themselves to partial truth. This feature makes the system
a powerful tool for analyzing ambiguous drilling scenarios.
Results of compared field cases are encouraging and conformal to field prac
tices. Allowing for human input, when essential data are lacking, makes the
system a useful tool that can help less experienced individuals function n
ear the level of proficient drilling engineers. (C) 2001 Elsevier Science B
.V. All rights reserved.