For library components, COSTALD is a very good prediction method. There are more complex equations of state that will do better, but when you stick to cubic equations of state, using COSTALD is probably the best option (see JM Campbell article on liquid density). However, once you start using hypothetical components, the COSTALD method may not be the best method anymore. A common reason for needing the volume translation option is that your PVT package may have provided liquid density information in the form of volume translation constants. In general you should be careful to check if the units of the constants you received match the units used in the simulator. And in particular for data sourced from PVTSim, be aware that the sign of those constants should be changed before entering them in the simulator!
Models can be slow, ugly or even be plain wrong. This is my attempt to help prevent that kind of small or not so small disasters.
Sunday, 15 March 2015
Did you know series: You can use the Peneloux volume translation in HYSYS - Why would you?
For library components, COSTALD is a very good prediction method. There are more complex equations of state that will do better, but when you stick to cubic equations of state, using COSTALD is probably the best option (see JM Campbell article on liquid density). However, once you start using hypothetical components, the COSTALD method may not be the best method anymore. A common reason for needing the volume translation option is that your PVT package may have provided liquid density information in the form of volume translation constants. In general you should be careful to check if the units of the constants you received match the units used in the simulator. And in particular for data sourced from PVTSim, be aware that the sign of those constants should be changed before entering them in the simulator!
No comments:
Post a Comment