Air Separation Unit from COCO Sample Flowsheets

[carolina23]

Hello,

Is it possible to use the ASU model available at to have a nitrogen production rate of 3 kg/s? I have tried to adjust the inlet air mass flow but that does not seem to work properly. Thank you for your help.

Best,
Carolina

[hkooijman]

Yes it is possible but you need to carefully adjust two specifications for the low pressure column that is flow rate specific, namely the side stream rates out of the low pressure column to (1) the waste and (2) the argon side rectification column. I suggest you to first replace them by relative flow specifications (switching "from mole" flow to "mole ratio") of 0.9 and 0.145, respectively. Once you have done that, resolve. Then after that you can adapt the feed inlet in steps of 10 k/s from 100 down to 10 kg/s. I will verify this leads to correct answers and then upload a new version of the ASU flowsheet using relative flow specifications later. Note that the steps need to be small enough in order to keep the low pressure column and the Argon column converging.

Harry

[hkooijman]

I computed the first ratio incorrectly and as a result the product slate changed. I corrected the calculation and saw the flow ratio's needed to be 7.85 and 0.17. I then noticed that the rich liquid feed to the low pressure column was changing its quality (i.e. the ratio of L/V). This is because I had assigned it a fixed duty, whereas that should actually be the duty to run the argon column's condenser. Therefore, I added an energy stream that ensures that the duty of the argon column is used as input for the heat the rich liquid feed experiences. This ensures that the flowsheet converges now over the range of 10 to 100 kg/s (I still advise you to not make the step changes too large). Notice that when you lower the flows the heat loss the columns experience should be adapted correspondingly as well, in order to keep results the same. This matters when you reduce the flow rates, and be aware that each column has a heat loss assigned to it. When you reduce the flows it is better to assign the heat losses first to the lower values expected. The new file for the ASU is now online for download.
H