Αναζήτηση

Πλοήγηση

• Κοινότητες
• Συγγραφείς
• Έτος
• Τίτλοι
• Θέματα

Ο Χώρος μου

• Είσοδος

An accurate method to generate composite PVT data for black oil simulation

Gaganis Vasileios, Kourlianski Ilias, Varotsis Nikolaos

Πλήρης Εγγραφή

---

Περίληψη

Lookup Black Oil Tables (BOT) are widely utilized by professional software packages for providing fluid properties during reservoir, wellbore and pipeline flow simulation instead of having to solve repeatedly the highly non-linear phase behavior problem at a very high CPU cost. These Tables, are either generated using an Equation of State (EoS) model already tuned to the available lab data or from the data itself or by applying available correlations of general use. Given the considerable effort and expertise required for achieving a comprehensive EoS tuning, the methods for generating the BOT utilizing the PVT lab data are still very popular. Such methods currently available are based on simplifying assumptions and often fail to match accurately the required composite vaporization Bo and Rs curves along the entire operating pressure range. In addition, these methods do not allow the recalculation of the BOTs when the selected separation train conditions are altered unless such a PVT lab separation test data is already available. In this work, an accurate method is presented to derive BOTs by combining PVT lab depletion data with properties values which can be generated by running simplified surface flash calculations. Separator gas and tank oil volumes are obtained by flashing the reservoir oil at the selected pressures using k-values generator methods and by taking advantage of the fact that, at surface train conditions, equilibrium coefficients are practically independent of composition. The results obtained can be also matched against available PVT lab report flash data or field measurements. The method was applied to several reservoir oils and the accuracy of these test cases proved to be of the order of 1% and 2% for the prediction of Bo and Rs respectively all along the operating pressure range. Finally, an approach is presented by which the generated BOTs can be accurately adjusted according to any variations of the separator train conditions without the need of additional lab data.

Υπηρεσίες