| Autor: |
Magdi, Dalia Ahmed, Elgendy, M. Y. |
| Předmět: |
|
| Zdroj: |
Egyptian Computer Science Journal; Sep2016, Vol. 40 Issue 3, p48-57, 10p |
| Abstrakt: |
One of the major current and future challenges facing the oil and gas industry is to maximize the oil recovery factor. One optimization technique is Horizontal wall-bore drilling methodology, "Geo-steering". This type of techniques depends on adjusting the borehole position on-the-fly to reach one or more geological targets; these adjustments are based on the petro-physical information gathered during drilling requires real-time drilling directions decisions. A main step needed for this methodology is building a geological model before drilling for the specified regions based on surrounding pre-drilled wells. In this paper, a proposed solution based on "Deep Belief Network" is presented to construct the geological model and propose a drilling path to follow. In the presented work, a deep architecture "Deep Belief Network" (DBN) with a fine tuning training is used to construct the geological model for a new well before drilling starts. The proposed solution has shown superiority on the traditional Neural Network for determining constructing the model. The proposed system generated a proposed solution achieved more than 92% accuracy based on real training drilling measurements from surrounding 5 wells. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
