Using commercial simulators for determining flash distillation curves for petroleum fractions
This work describes a new method for estimating the equilibrium flash vaporisation (EFV) distillation curve for petroleum fractions by using commercial simulators. A commercial simulator was used for implementing a stationary model for flash distillation; this model was adjusted by using a distillat...
Gespeichert in:
Veröffentlicht in: | Ingeniería e investigación 2008-01, Vol.28 (1), p.50-56 |
---|---|
Hauptverfasser: | , , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | This work describes a new method for estimating the equilibrium flash vaporisation (EFV) distillation curve for petroleum fractions by using commercial simulators. A commercial simulator was used for implementing a stationary model for flash distillation; this model was adjusted by using a distillation curve obtained from standard laboratory analytical assays. Such curve can be one of many types (eg ASTM D86, D1160 or D2887) and involves an experimental procedure simpler than that required for obtaining an EFV curve. Any commercial simulator able to model petroleum can be used for the simulation (HYSYS and CHEMCAD simulators were used here). Several types of petroleum and fractions were experimentally analysed for evaluating the proposed method; this data was then put into a process simulator (according to the proposed method) to estimate the corresponding EFV curves. HYSYS- and CHEMCAD estimated curves were compared to those produced by two traditional estimation methods (Edmister’s and Maswell’s methods). Simulation-estimated curves were close to average Edmister and Maxwell curves in all cases. The proposed method has several advantages; it avoids the need for experimentally obtaining an EFV curve, it does not depend on the type of experimental curve used to fit the model and it enables estimating several pressures by using just one experimental curve as data. |
---|---|
ISSN: | 0120-5609 2248-8723 |
DOI: | 10.15446/ing.investig.v28n1.14867 |