Fast Real-Time Production Optimization for Integrated Asset Modelling Using Mixed-Integer Non-Linear Programming in Julia Language
Gespeichert in:
| Titel: | Fast Real-Time Production Optimization for Integrated Asset Modelling Using Mixed-Integer Non-Linear Programming in Julia Language |
|---|---|
| Autoren: | Ramiro Canchucaja |
| Quelle: | SPE Latin American and Caribbean Petroleum Engineering Conference. |
| Verlagsinformationen: | SPE, 2023. |
| Publikationsjahr: | 2023 |
| Beschreibung: | This study proposes a method for integrated asset modelling by using machine learning along with operations research algorithms to perform real-time constrained production optimization and maximize operational profit in a real-time basis. The methodology, which is mainly about the transformation of field and well performance to equations, inequalities, and matrixes, was tested successfully in the operation of a gas condensate field where operational profit increased, in abnormal conditions when personnel normally act mostly based on experience, pre-conceived notion, or solutions to previously solved problems. The study provides a solution with full data-driven objectivity for decision-making using the results of a mixed integer non-linear programming problem. |
| Publikationsart: | Article |
| DOI: | 10.2118/213138-ms |
| Dokumentencode: | edsair.doi...........5ceb2e61924c68a0b5859dacdb967327 |
| Datenbank: | OpenAIRE |
Nájsť tento článok vo Web of Science | Abstract: | This study proposes a method for integrated asset modelling by using machine learning along with operations research algorithms to perform real-time constrained production optimization and maximize operational profit in a real-time basis. The methodology, which is mainly about the transformation of field and well performance to equations, inequalities, and matrixes, was tested successfully in the operation of a gas condensate field where operational profit increased, in abnormal conditions when personnel normally act mostly based on experience, pre-conceived notion, or solutions to previously solved problems. The study provides a solution with full data-driven objectivity for decision-making using the results of a mixed integer non-linear programming problem. |
|---|---|
| DOI: | 10.2118/213138-ms |