Premium
DETERMINATION OF THE NATURE OF RESERVOIRS BY USE OF PATTERN RECOGNITION ALGORITHM WITH PRIOR LEARNING *
Author(s) -
BOIS P.
Publication year - 1981
Publication title -
geophysical prospecting
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.735
H-Index - 79
eISSN - 1365-2478
pISSN - 0016-8025
DOI - 10.1111/j.1365-2478.1981.tb00703.x
Subject(s) - geology , regional geology , process (computing) , telmatology , environmental geology , gemology , economic geology , algorithm , pattern recognition (psychology) , computer science , engineering geology , artificial intelligence , hydrogeology , paleontology , geotechnical engineering , tectonics , volcanism , operating system
A bstract The determination of the nature of reservoirs containing hydrocarbons is approached via a pattern‐recognition method with prior learning. This method consists of comparing the nature of a reservoir that has been drilled into with an unknown reservoir. The known reservoir plays the part of a monitor for the learning process. To this end an algorithm is compiled using the Burg method of the features represented by different portions of traces contained in the reservoirs. An analysis of the dots representing these different features serves as a criterion for deciding whether the two reservoirs have identical or different natures.