| Title | INTEGER PROGRAMMING OPTIMIZATION OF PRODUCTION WELL PLACEMENT |
|---|---|
| Authors | R. Adiga, J. OSullivan and A. Philpott |
| Year | 2017 |
| Conference | New Zealand Geothermal Workshop |
| Keywords | Well placement, optimization, resource estimation, resource management, TOUGH2, Python, PyTOUGH, Gurobi |
| Abstract | Geothermal power generation is not keeping pace with other renewable energy technologies. This is due to a number of factors including the industry’s high capital cost, of which wells account for a significant portion. Hence, it is imperative to maximize value from wells drilled by selecting them optimally. An important technology used when making well placement decisions is computer simulation of production. This is usually done manually, with experts creating reservoir models, simulating wells at candidate locations and comparing the predicted production scenarios. Manual selection in this manner is slow and labor intensive, requiring weeks or months for expert modelers to make recommendations. Recently, various heuristics have been investigated to try and automate this process. Examples include Particle Swarm Optimization, Genetic Algorithms, Simulated Annealing, and Gradient descents. However, no strict form optimization that guarantees the best solution has been attempted for the complex problem of selecting multiple production wells to maximize value. This paper uses Integer Programming to address this problem. An economic model was used to calculate Net Present Values (NPVs) for a set of candidate wells and the interactions between them using AUTOUGH2 simulation results of an example geothermal system. Binary decision variables were used in the optimization to select the combination of wells that would maximize total NPV. Unlike with heuristics, the solution is guaranteed to be optimal, at least with respect to the economic model. It was also found to be optimal with respect to the AUTOUGH2 model for the reservoir used. |