Full text loading...
-
Hydrocarbon Probability Index Based on Ann and Prestack Attributes
- Publisher: European Association of Geoscientists & Engineers
- Source: Conference Proceedings, 67th EAGE Conference & Exhibition, Jun 2005, cp-1-00290
Abstract
F001 HYDROCARBON PROBABILITY INDEX BASED ON ANN AND PRESTACK ATTRIBUTES Summary 1 Risk assessment for hydrocarbon-saturated reservoirs can be improved using neural network classification methods when combined with interpreter's knowledge and multitude of prestack attributes. Training data selected over background events and known hydrocarbon deposits permits calibration of untested reservoirs which in turn improves the pre-drill prediction process as well as the range of possible outcomes thus providing a measure of the uncertainty. The Gulf of Mexico examples presented here demonstrate the potential for improved reservoir assessment and exploration risk reduction with the aforementioned technique. Introduction With the proliferation of