Virtual intelligence technology offers a different paradigm for computing, as biologically-inspired computing (soft computing) is radically different from conventional computing (hard computing). In discussing virtual intelligence, it is common to present it using a biological metaphor. We have some understanding of how humans process information and learn. Virtual intelligence techniques such as artificial neural networks (ANNs), genetic algorithms, and fuzzy logic are used as rough models of mental processes. As these techniques operate in a massively parallel fashion, they can process information and carry out solutions almost simultaneously in an intuitive manner. This is why some interactive non-linear processes that cannot be tracked through the diligent use of analytical formulations can be practically solved using soft computing techniques. In this presentation, four ANN applications to four different reservoir engineering problems are discussed. The first application demonstrates the utilization of an artificial neural network for predicting the relative permeability characteristics. The second application is designed to overcome some of the significant challenges that are faced in positioning a new well during a field development study. The third application shows the use of soft computing methodology in the analysis of pressure transient data. Finally, the fourth application involves the use of artificial neural networks as a screening tool for CO2 sequestration process in coal seams. In all of these applications the question that we are trying to answer is whether the virtual intelligence is capable to minimize a cost function, energy function, time function, or a complex combination of these functions in finding the optimum solution to reservoir engineering problems which fall under one of these groups.
Register @ www.spegcs.org
or by fax: (713) 779-4216
(Please print / bring credit card receipt if paid on-line.)