Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

An investigation of the applicability of expert systems to diagnose formation damage problems

Book ·
OSTI ID:7026023
This dissertation presents the results of a study undertaken to investigate the applicability of expert systems technology to the diagnosis of formation damage problems. Three major contributions were derived from this investigation. First, a list of common types of formation damage, and a complete description of the nature and causes of each, was generated. Second, a methodology to automate the diagnosis task and to store the required knowledge was developed. Third, a computerized knowledge base system implementing this methodology was developed. To automate formation damage diagnosis the required knowledge base was divided into nine parts, which when organized in a hierarchic manner represented the logical flow of information used to diagnose problems in this domain. Each part of the knowledge base is stored in a separate microcomputer file and the pertinent knowledge translated to computer form using the so-called rule-based approach. Using this approach, the diagnostic process was translated to a computerized form by a creative design of the inference tree. Both uncertainty in the reasoning and in the data are used to emulate the inherent judgmental characteristics of the formation damage diagnosis task. Uncertainty is expressed by a numerical certainty factor associated with almost every piece of knowledge encoded in the knowledge base. This knowledge-based system was implemented using the framework offered by the expert systems development tool named PCPLUS. Tests with the system show that formation damage diagnosis can be automated using the approach explained in this dissertation.
Research Organization:
University of Southern California, Los Angeles, CA (USA)
OSTI ID:
7026023
Country of Publication:
United States
Language:
English