Proxy Models, an alternative to reduce the computational time during numerical reservoir simulation
No Thumbnail Available
Date
Advisors
Evaluators
Journal Title
Journal ISSN
Volume Title
Publisher
Universidad Industrial de Santander
Abstract
Description
La simulación numérica de yacimientos es una herramienta que ha venido evolucionando de la mano de la industria petrolera, su objetivo fundamental es predecir el comportamiento de un yacimiento dado y posteriormente encontrar el escenario óptimo para aumentar la recuperación de hidrocarburos. Sin embargo, los estudios de simulación numérica pueden llegar a ser muy costosos por factores como: tamaño de la grilla y la robustez del estudio a realizar (asociado a mayores tiempos de cómputo y de análisis de resultados); además del tiempo que implicaría llegar a evaluar los múltiples escenarios posibles para definir la mejor estrategia de explotación. Es por esta razón, que los ingenieros de yacimientos se apoyan en herramientas estadísticas para aplicar los estudios de simulación numérica orientados al ajuste histórico, análisis de sensibilidad, análisis de incertidumbre, predicción y estrategias de optimización de producción. Dentro de estas herramientas estadísticas se encuentran los Modelos Proxy, los cuales surgen como alternativa para reducir los tiempos de cómputo en estudios de imulación numérica que no requieran un alto grado de exactitud en sus resultados. Este artículo busca contextualizar al lector sobre el significado de los modelos proxy y su utilidad durante los estudios de simulación numérica de yacimientos, además de presentar un ejemplo de aplicación.
The reservoir simulation is a tool which had been evolving at the same time than the petroleum industry; its main purpose is to predict the behavior of a reservoir and afterward finding the optimum scenario in order to increase the oil recovery. However, numerical simulation studies could be very expensive because factors such as grid size and detail of the study to develop (It is associated to long computational and analysis times). Furthermore the time it would take to evaluate multiple possible scenarios to define the best exploration strategy. For this reason, field engineers use statistical tools during numerical simulation studies oriented to history match, sensitivity analysis, uncertainty analysis, optimization and forecasting. One of these statistical tools is the Proxy model, it appears as an alternative to reduce computational times which does not need a high accuracy grade in the results. This article looks for a reader contextualization about the meaning of Proxy model and its value during reservoir simulation studies, and also to present an application study.
The reservoir simulation is a tool which had been evolving at the same time than the petroleum industry; its main purpose is to predict the behavior of a reservoir and afterward finding the optimum scenario in order to increase the oil recovery. However, numerical simulation studies could be very expensive because factors such as grid size and detail of the study to develop (It is associated to long computational and analysis times). Furthermore the time it would take to evaluate multiple possible scenarios to define the best exploration strategy. For this reason, field engineers use statistical tools during numerical simulation studies oriented to history match, sensitivity analysis, uncertainty analysis, optimization and forecasting. One of these statistical tools is the Proxy model, it appears as an alternative to reduce computational times which does not need a high accuracy grade in the results. This article looks for a reader contextualization about the meaning of Proxy model and its value during reservoir simulation studies, and also to present an application study.
Keywords
Numerical Reservoir Simulation, Proxy Model, CO2 injection, Simulación Numérica de Yacimientos, Modelos Proxy, Inyección de CO2