Full text loading...
-
Application of Probabilistic Neural Network on Complex Carbonate Reservoir Prediction
- Publisher: European Association of Geoscientists & Engineers
- Source: Conference Proceedings, IPTC 2013: International Petroleum Technology Conference, Mar 2013, cp-350-00392
Abstract
Carbonate reservoir in Tarim basin is more complex for its strong heterogeneity and anisotropy. Different from the usual, this paper mainly focused on how to employ PNN as a comprehensive method in detecting the dissolved caves for carbonate reservoir, in which advantages of inversion and attributes are both used. During the entire workflow of PNN, we emphasize the importance of seismic attributes searching and cross validation. And the P-velocity and S-velocity of a test area selected from ZG8 area are predicted by utilizing PNN technique, and the interpretation of Formation Yingshan is conducted on the established indicator of Vp/Vs. Results show that the resolution is greatly boosted and the dissolved caves are distinctively and vividly characterized, which is seldom achieved by pre-stack inversion.