الفهرس | Only 14 pages are availabe for public view |
Abstract Field having many wells makes it difficult to choose the most efficient Method for increasing productivity of frac, so I try in this thesis using artificial intelligence neutral networks to develop a platform model for selecting the best well candidate for maximum overall productivity of an oil field, study the different affecting parameters on reservoir stimulation and predict the performance and future optimum designs. Artificial intelligence neural network is an information processing system simulating the natural neural system in the human brain. Using it many complex petroleum problems that are difficult for traditional models and computing systems can be solved. It has shown great potential for generating accurate analysis and results from large amount of historical data that otherwise would seem not to be useful in the analysis. It also can make the best selection for any output relevant to several inputs and calculate the optimum value of it for different cases. More than 100 wells with their datasets were used for the study. This data after modeling suggests that on average, over 80% of the stage production performance is controlled by subsurface geological and petrophysical characteristics. Less than 20% of the stage production performance is impacted by completion design parameters. 11% of the stage performance is impacted by total proppant volume. |