Loading...

Automatic Well Planning to Optimize ROP Using Previous Wells Data

Heidari, Amir Hossein | 2022

134 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 55236 (06)
  4. University: Sharif University of Technology
  5. Department: Chemical and Petroleum Engineering
  6. Advisor(s): Jamshidi, Saeed
  7. Abstract:
  8. Today, with the reduction of world oil prices and the development of competing energy sources, fossil fuels such as nuclear energy, electricity, wind, etc., in order to increase the economic efficiency of oil sales, we must Reduce production costs to a minimum. If the components of oil extraction costs are examined, the cost of drilling production wells is one of the main economic components that affect the cost of producing oil. Therefore, according to the above explanations, the cost of drilling wells should be reduced.To reduce the cost of drilling a well to be drilled, it is better to predict the drilling parameters (for example ROP) before starting the drilling process. For this prediction, the well plan must be designed before drilling. To design this program, wells similar to the target well must be identified and the data of similar wells must be used to design the target well program. In the old methods for determining wells similar to the target well, only geological layers were studied based on UGC maps and well distance between two wells, but in this study, the similarity of the two wells is defined based on the overall similarity index. The closer this index is to one, the more similar the two compared wells are, and the closer it is to zero, the more different the two compared wells are. For this overall similarity index, the partial indices should be calculated based on the distance of the wells, the trajectory of the wells, the casing design of the walls, and the formation prognosis of the wells. Afterward, the arithmetic average of these partial similarity indices should be calculated. At this stage of the research, the two wells are compared in terms mentioned aspects. In the next step of this research, similar wells must be grouped together in a cluster or group. For this classification, artificial intelligence methods are used. The input of these artificial intelligence methods is the partial similarity indicators calculated in the previous step. By identifying well clusters, well data that is in the same cluster as the target well is used to plan the target well
  9. Keywords:
  10. Drilling ; Artificial Intelligence ; Wells Clustering ; Production Costs ; Oil Production ; Similarity Index

 Digital Object List

 Bookmark

No TOC