INTEGRATED CHARACTERIZATION OF HEAVY OIL RESERVOIR USING VP/VS RATIO AND NEURAL NETWORK ANALYSIS

The focus of this study is the southern portion of the Long Lake lease located approximately 40 km southeast of Fort McMurray, Alberta, Canada The lease area is roughly 25,000 hectares and contains over 8 billion barrels of bitumen in place. For heavy oil projects, the V-P/V-S ratio is a good lithology discriminator, and the objective of this paper is to predict a V-P/V-S ratio volume based on neural network analysis Neural network estimation of reservoir properties has proven effective in significantly improving accuracy and vertical resolution in the interpretation of the reservoir The strength of a neural network analysis is the ability to determine nonlinear relationships between logs and several seismic attributes The result is a new lithology calibrated attribute that, when co-rendered with edge detector attributes, can predict the presence of muddy intervals responsible for impacting the propagation of steam through the reservoir, thereby allowing us to more effectively describe enhanced oil recovery in the reservoir