Loading...

Characterization and Modeling of Heavy Oil Fields Using Artificial Neural Networks to Optimize Horizontal Wells Selection

Al Ali, Najeh | 2009

956 Viewed
  1. Type of Document: Ph.D. Dissertation
  2. Language: Farsi
  3. Document No: 39455 (06)
  4. University: Sharif University of Technology
  5. Department: Chemical and Petroleum Engineering
  6. Advisor(s): Pishvaie, Mahmoud Reza; Taghi Khani, Vahid; Buzorgmehri, Ramin
  7. Abstract:
  8. Steam-Assisted Gravity Drainage (SAGD) is one of the best and the most effective methods of enhanced oil recovery (EOR) which is used in the production from Heavy Oil Reservoirs which constitute large amount of oil throughout the globe. In a typical process of SAGD, steam is injected into a horizontal well and the oil is produced several meters deeper in the ground. In order to increase the Recovery of this process, it is better to review on the Pre-heating methods around the injecting and producing wells. Pre-heating is important when the viscosity of the oil is large enough to prevent oil from any motion at the beginning of the SADG operation, and makes it necessary to to use a Pre-heating method. An ideal form of this method (Pre-heating) is that it can be performed in a short time. Here, the effect of various methods on production process at the beginning of the production is studied using numerical simulating. Variables such as Recovery Factor (RF), Temperature Distribution (TD) and Production Rate (PR) are reviewed from the results obtained from the Simulation. Using the results obtained, we come to this conclusion that vapor injection cycle, as a pre-heating method, is the best method to be used. In addition, we found that low viscosity, high thickness of the Reservoirs and large amount of soluble gas causes an increase of production. The results of STARS were analyzed in terms of sensitivity, and the effect of parameters that involved in this process on the results of the production in the well was evident. From the results of this sensitivity analysis, the parameters that had the most influence on this process came to be known. The results were applied in the Artificial Neural Network which studied the application of this network as a substitute for common simulation of SAGD to be able to forecast based on this process to reduce cost, time and calculations and better results from Simulation. We could achieve a neural network that has an estimation and prediction of Recovery Factor for a period of 10 years. This research presents a method that can solve the problem of Simulators specially their time consuming and expensive price, and models the steam injection into the horizontal wells and production using Steam-Assisted Gravity Drainage. In this method, the distribution of stability of time distribution in front of vapor interface (Steady State) has been modeled in the form of Unsteady State rather than Steady State. We could calculate Vapor Interface position and Interface velocity in the Reservoirs in any time and in terms of every Recovery Factor
  9. Keywords:
  10. Heavy Oil ; Artificial Neural Network ; Analytical Modeling ; Steam Assisted Gravity Drainage (SAGD) ; Stars Software

 Digital Object List

 Bookmark

No TOC