Newsgroups: comp.ai.neural-nets
Path: cantaloupe.srv.cs.cmu.edu!rochester!udel!news.mathworks.com!hookup!news.moneng.mei.com!uwm.edu!cs.utexas.edu!howland.reston.ans.net!ix.netcom.com!netcom.com!park
From: park@netcom.com (Bill Park)
Subject: Re: ANN for wireline log data interpretation
Message-ID: <parkD6B00G.CsF@netcom.com>
Followup-To: comp.ai.neural-nets
Keywords: neural network log analysis tight sand petrochemical
Cc: tfungcc@cc.curtin.edu.au
Organization: Netcom Online Communications Services (408-241-9760 login: guest)
References: <tfungcc.1.000E9E54@cc.curtin.edu.au>
Date: Fri, 31 Mar 1995 12:00:16 GMT
Lines: 55
Sender: park@netcom18.netcom.com

In article <tfungcc.1.000E9E54@cc.curtin.edu.au>,
    tfungcc@cc.curtin.edu.au (Lance Fung) writes:
>
> We are investigating into the application of ANN in wireline log data 
> interpretation.
> 
> We'd appreciate some references in this area and the types of net being 
> used and their performance.
> 
> Lance Fung
> Electronics and Instrumentation Research Group

Mirna Urquidi-Macdonald, Harold S. Javitz, William Park,
Janet D. Lee, and Aviv Bergman, "Feasibility of Calculating
Petrophysical Properties in Tight Sand Reservoirs Using
Neural Networks, Final Report (October 1989-June 1991),"
Information Management and Technology Center, SRI
International, 333 Ravenswood Avenue, Menlo Park, CA 94025,
for Gas Research Institute, Contract No. 5089-260-1845, GRI
Project Manager: Anthony W. Gorody, Ph. D., Physical Science
Department (July 1991).

ABSTRACT

The objective of this research was to determine the
feasibility of using neural networks to estimate
petrophysical properties in tight sand reservoirs.  A second
objective was to gain some experience concerning how to
approach the development of a future prototype, including
what should be done and what should be avoided.

Gas Research Institute (GRI) focused the project on tight
sands because they contain enormous gas reserves and their
complicated lithology represents a challenge to log
analysts.  The data were supplied by GRI from two of its
geographically proximate experimental wells in tight sand
formations.  The nets were tested in sections of those wells
that were not used for training, and in two other wells, one
in a geographically close but geologically unrelated
formation and one in Wyoming.

The feasibility testing demonstrated that the relatively
simple neural networks developed have comparable accuracy
with standard logging analysis estimates in wells that
contributed data to the training set.  Transportability of
the network was tested by using core measurements in two
wells in which the nets were not trained, with inconclusive
results.  Recommendations were made that would increase the
accuracy of the neural networks.

====================

William T. Park, Ph. D.
William T. Park and Associates
Neural Networks for Industrial Applications
