University of Houston
Department of Computer Science

In partial fullfillment of the Requirements for the Degree of
Master of Science


Vincent Sookram
will defend his thesis

Classification Of Seismic And Well Log Responses To Geological Formations Using Neural Nets


Abstract

Current methods in petroleum exploration generate a three dimensional view of subsurface formations using surface geology, seismic surveys and well logs. Well log data is used for vertical control in processing and interpreting seismic data. Patterns in this data are also used for characterizing and correlating geological formations Petrophysical properties of subsurface formations and attributes of seismic data are related to physical properties of the rocks and their contained fluids. Acoustic impedance, Poisson’s ratio, bulk, shear and Young’s moduli have been related to sonic and density logs. These logs are used to generate synthetic seismograms which are correlated directly to seismic sections. Using Neural Networks to relate seismic attributes to geophysical well logs shows considerable promise. Additional research needs to be done to explore all the possible relations and uses for Neural Networks in petroleum exploration.


Date: Thursday, April 20,2000
Time: 10:00 AM
Place: 550-PGH


Faculty, students, and the general public are invited
Thesis Advisor: Dr. Olin G. Johnson