SYMPOSIA PAPER Published: 01 January 1991
STP17626S

Expert Systems for Classification and Identification of Waterborne Petroleum Oils

Source

Expert systems are among the most fascinating application of computers which attempt to emulate human cognition and perceptual abilities. In the environmental field, expert systems have been used for spectral pattern recognition and have proven useful for classification and identification of waterborne petroleum oils.

This paper, after reviewing expert systems in general, describes a few successful attempts by the authors for classification of synchronous fluorescence spectra of petroleum oils. A typical scheme for developing an expert system for the classification and identification of petroleum oils is also presented. The scheme suggested uses pattern recognition and artificial intelligence (AI) for modeling expert knowledge and spectral processing. Using this model, an expert system called Spectral Expert system (SES) was designed. SES comprised of five primary components, fact gathering, knowledge/rule base, knowledge organization/learning, inference engine and expert/user interface. All these components were developed and integrated into a working system. The system is successfully used to classify several oil samples.

Author Information

Siddiqui, KJ
Creighton University, Omaha, NE
Lidberg, RL
Lockheed Engineering and Sciences Company, Las Vegas, NV
Eastwood, D
Lockheed Engineering and Sciences Company, Las Vegas, NV
Gibson, G
Lockheed Engineering and Sciences Company, Las Vegas, NV
Price: $25.00
Contact Sales
Related
Reprints and Permissions
Reprints and copyright permissions can be requested through the
Copyright Clearance Center
Details
Developed by Committee: D19
Pages: 598–612
DOI: 10.1520/STP17626S
ISBN-EB: 978-0-8031-5163-5
ISBN-13: 978-0-8031-1407-4