Prediction of immiscible gas flooding performance: a modified capacitance–resistance model and sensitivity analysis
Abstract Reservoir performance prediction is one of the main steps during a field development plan. Due to the complexity and time-consuming aspects of numerical simulators, it is helpful to develop analytical tools for a rapid primary analysis. The capacitance–resistance model (CRM) is a simple tec...
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | article |
| Lenguaje: | EN |
| Publicado: |
KeAi Communications Co., Ltd.
2019
|
| Materias: | |
| Acceso en línea: | https://doaj.org/article/a6b14cedbd334b979a9d9ac081769298 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| Sumario: | Abstract Reservoir performance prediction is one of the main steps during a field development plan. Due to the complexity and time-consuming aspects of numerical simulators, it is helpful to develop analytical tools for a rapid primary analysis. The capacitance–resistance model (CRM) is a simple technique for reservoir management and optimization. This method is an advanced time-dependent material balance equation which is combined with a productivity equation. CRM uses production/injection data and bottom-hole pressure as inputs to build a reliable model, which is then combined with the oil-cut model and converted to a predictive tool. CRM has been studied thoroughly for water flooding projects. In this study, a modified model for gas flooding systems based on gas density and average reservoir pressure is developed. A detailed procedure is described in a synthetic reservoir model using a genetic algorithm. Then, a streamline simulation is implemented for validation of the results. The results show that the proposed model is able to calculate interwell connectivity parameters and oil production rates. Moreover, a sensitivity analysis is carried out to investigate effects of drawdown pressure and gas PVT properties on the new model. Finally, acceptable ranges of input data and limitations of the model are comprehensively discussed. |
|---|