Flow Sheet Solver

[nrgeng]

The flow sheet requires much time to solve. At times, it ran for about 8 hours without a solution. Using trial and error, I was able to solve the flow sheet if I disconnected the various process blocks. Please refer to the diagram "Flow Sheet Block Connections."

Simplified Representation of Flow Sheet

Flowsheet Block Connections.JPG (97.39 KiB) Viewed 8662 times

The procedure used was to disconnect the blocks at the four Disconnect Points shown. Then Block 1 was solved and Point 1 was connected between Blocks 1 and 3. When Blocks 1 and 3 solved, Point 2 was connected and then Blocks 1, 3, 5 and 7 were solved. Point 3 was connected and Blocks 1, 3 , 5 and 7 were solved with the one recycle stream to Block 1. Lastly, Point 4 was connected and the four blocks were solved with the second recycle stream to Block 5. As you can surmise, any change requires that the entire tedious procedure be repeated.

This is a very time consuming procedure to use. I hope that you can suggest one or more faster methods for the solution of this flow sheet. Flow Sheet Options settings, special Unit Operations, and/or other methods are solicited.

[jasper]

Without further information it is hard to say anything. Which variable has trouble converging? (you can get this information by looking at the variable with the highest residual, if you set the verbosity level to Debug)

Interestingly you pick streams 2, 3, 4 that break 1 recycle. The streams going into the mixers break multiple recycles (and I imagine they are selected as cut streams). So does stream 1 and the stream that connects blocks 5 and 7.

I reckon not every change requires repetition of the procedure as the procedure itself involves changes. So it sounds like there are problems converging particular equations. Can you send me the fsd or is that not an option? (don't do so please if there is sensitive information in there or in case the flowsheet contains CAPE-OPEN components that require a separate license).

[nrgeng]

The flow sheet contains sensitive information so I will not be sending it for review. Too bad, that would have helped. I will try the debug suggestion. Also, I unfortunately only labeled the Disconnect Points and not the streams, so the Stream Numbers that you cite are not streams. I will correct my error and add Stream Numbers to the Diagram for a better understanding of your explanation.