Improving Genetic Algorithms for Optimum Well Placement
Abstract/Contents
- Abstract
- Optimum well placement can help reservoir management teams in developing a field development plan that could result in substantial increase in productivity and profitability of any new or existing field. The proposed location and configuration for new producers and injectors is usually nontrivial, due to the complexity of the fluid flow in highly heterogeneous reservoirs. The objective of this work was to understand the steps involved in the optimum well placement by GAs, and to introduce enhancements to the algorithm that could increase the possibility of obtaining promising solutions. Based on the success of binary GAs in optimum well placement problems and motivated by the advantages observed in application of continuous GAs in other fields, here we attempt to use continuous GAs in for field development. To meet our objectives, we have investigated the design of continuous GA that retain the important benefits characteristics of binary GAs, while solving some of the problems associated with binary GAs. The implementation of continuous GA was designed to avoid generating invalid wells during the reproduction process. Continuous GAs have shown considerable potential to achieve higher fitness values. The gradual progress of a continuous GA during the generations, compared to the stepwise evolution observed in a binary GA, has the possibility of achieving more desirable outcomes. However, the design of a powerful optimization tool with continuous GAs is harder because of the higher number of GA parameters involved. The study also implemented dynamic mutation to take advantage of the exploring capacity of mutation in each period of the evolution. Furthermore, it has been shown that the efficiency of the GA search can be increased by imposing a minimum Euclidian distance between the individuals within the population, utilizing our engineering knowledge by requiring a minimum physical distance between all the wells in a reservoir. Finally, a model was introduced to include curved wells during the search. Through this model the possibility of capturing straight wells still exists, while providing the opportunity of exploring more promising configurations. Throughout this work we have demonstrated that various improvements introduced lead to finding higher objective values in shorter time.
Description
| Type of resource | text |
|---|---|
| Date created | June 2008 |
Creators/Contributors
| Author | Moravvej Farshi, Mohammad |
|---|---|
| Primary advisor | Aziz, Khalid |
| Degree granting institution | Stanford University, Department of Energy Resources Engineering |
Subjects
| Subject | School of Earth Energy & Environmental Sciences |
|---|---|
| Genre | Thesis |
Bibliographic information
Access conditions
- Use and reproduction
- User agrees that, where applicable, content will not be used to identify or to otherwise infringe the privacy or confidentiality rights of individuals. Content distributed via the Stanford Digital Repository may be subject to additional license and use restrictions applied by the depositor.
Preferred citation
- Preferred Citation
- Moravvej Farshi, Mohammad. (2008). Improving Genetic Algorithms for Optimum Well Placement. Stanford Digital Repository. Available at: https://purl.stanford.edu/tf860rn9355
Collection
Master's Theses, Doerr School of Sustainability
View other items in this collection in SearchWorksContact information
- Contact
- brannerlibrary@stanford.edu
Also listed in
Loading usage metrics...