Loading...

Development of Artificial Intelligence Model to Optimize Dynamic Parameters during Acidizing Operation

Mousavi Badjani, Amir Hossein | 2023

85 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 56018 (06)
  4. University: Sharif University of Technology
  5. Department: Chemical and Petroleum Engineering
  6. Advisor(s): Ayatollahi, Shahaboddin; Aghaei, Hamed
  7. Abstract:
  8. At each stages of oil and gas production, from the time of drilling of the wells to the state of production and development of the reservoirs, formation damage would hinders the oil/gas rate and causes high pressure drop in the drainage area. To eliminate the damage caused by the formation blockage, several remediation techniques are used, which are called well stimulation methods. The most common method for the past tens of years is matrix acidizing which lead to the improvement of the performance of the well. To optimize this operation, optimal acidizing design is needed, otherwise the acidizing process face failure and lead to blocking of the well through acid-induced damage. One of the most important issues during acid treatment is to inject the acid at the optimal flow rate in such a way that with the least amount of acid used, the most effective wormholes are formed. Otherwise, the excessive consumption of acid causes lack of proper dissolution of rock and the damage cannot be removed. To determine the optimal acid injection rate for acidizing operations, acid injection tests are performed on the reservoir core plugs. These tests are very risky due to the sensitivity of acid test. Besides, these tests are costly and critically increase the laboratory charge of this process. It should be mentioned that these tests are time consuming as well. In this project, the optimal acid rate is predicted by applying artificial intelligence (AI) techniques on a collected database related to acid core flooding tests. In the first stage a data base is created using a wide range of information available in the literature, then it was enriched by several tests performed on real core plugs from an Iranian gas field. Performing these acid injection tests on carbonate cores, the optimal flow rate of injection and the volume of acid used for acid breakthrough were determined. The proposed AI algorithms Based on the data of the database and comparing their prediction results with the laboratory results, it was found that the SVM algorithm was able to predict the optimal flow rate and PVBT with high accuracy
  9. Keywords:
  10. Acidizing Treatment ; Artificial Intelligence ; Artificial Neural Network ; Regression Analysis ; Carbonated Rock ; Formation Damage ; Carbonate Resrevoirs ; Optimal Flow Rate

 Digital Object List

 Bookmark

No TOC