9 replies to this topic

[sheiko]

Hello,

Could anybody tell me how to interpret the following message in my convergence report? Does this mean that my Material and Heat balance is incorrect?

Thanks in advance



SIMULATION SCIENCES INC. R PAGE H-2
PROJECT XXX PRO/II VERSION 8.0.1 ELEC V6.6
PROBLEM SWS CALCULATION xxx
HISTORY 12/11/08
==============================================================================

INNER 1 : E(ENTH+SPEC) = 4.366E-01 ALPHA = 0.0001
INNER 2 : E(ENTH+SPEC) = 4.362E-01 ALPHA = 0.0000
INNER 3 : E(ENTH+SPEC) = 4.362E-01 ALPHA = 0.0001
INNER 4 : E(ENTH+SPEC) = 4.360E-01 ALPHA = -0.0008
*** ERROR *** COLUMN calculations TERMINATED. Alpha ( the damping factor )
was BELOW .01 for 4 consecutive iterations after 2 new matrix
calculations.
*** A FATAL ERROR OCCURRED ON ITERATION 3
*** ERROR HISTORY IS NOT AVAILABLE FOR THIS ITERATION
UNIT 1 NOT SOLVED - 'E-2501 '
MAX ERRORS: X=-20.5 (EAU1) T=117. (NH3+H2S)
P=0.
Loop 1 NOT SOLVED after 1 trials
UNIT 2 SOLVED - 'C-2501 '
UNIT 6 SOLVED - 'FV-2502 '
INNER 0 : E(ENTH+SPEC) = 3.559E-01
CALCULATING NEW MATRIX
INNER 1 : E(ENTH+SPEC) = 3.189E-01 ALPHA = 0.9880
INNER 2 : E(ENTH+SPEC) = 2.716E-01 ALPHA = 1.0000
INNER 3 : E(ENTH+SPEC) = 2.505E-01 ALPHA = 1.0000
INNER 4 : E(ENTH+SPEC) = 1.761E-01 ALPHA = 1.0000
INNER 5 : E(ENTH+SPEC) = 1.233E-01 ALPHA = 1.0000
INNER 6 : E(ENTH+SPEC) = 9.974E-02 ALPHA = 0.2500
INNER 7 : E(ENTH+SPEC) = 7.755E-02 ALPHA = 0.2500
INNER 8 : E(ENTH+SPEC) = 6.008E-02 ALPHA = 1.0000
INNER 9 : E(ENTH+SPEC) = 4.419E-02 ALPHA = 0.2500
INNER 10 : E(ENTH+SPEC) = 3.491E-02 ALPHA = 1.0000
ITER 1 E(K) = 3.920E-01 E(ENTH+SPEC) = 3.491E-02 E(SUM) = 2.861E-01
DAMP = 5.000E-01
INNER 0 : E(ENTH+SPEC) = 6.944E-02
INNER 1 : E(ENTH+SPEC) = 4.106E-02 ALPHA = 0.2500
INNER 2 : E(ENTH+SPEC) = 2.153E-02 ALPHA = 1.0000
INNER 3 : E(ENTH+SPEC) = 8.128E-03 ALPHA = 1.0000
INNER 4 : E(ENTH+SPEC) = 7.773E-03 ALPHA = -0.0625
INNER 5 : E(ENTH+SPEC) = 6.976E-03 ALPHA = 0.0625
INNER 6 : E(ENTH+SPEC) = 5.554E-03 ALPHA = 0.2500
INNER 7 : E(ENTH+SPEC) = 2.242E-03 ALPHA = 1.0000
INNER 8 : E(ENTH+SPEC) = 9.140E-04 ALPHA = 1.0000
INNER 9 : E(ENTH+SPEC) = 9.069E-04 ALPHA = 1.0000
INNER 10 : E(ENTH+SPEC) = 5.432E-04 ALPHA = 1.0000
ITER 2 E(K) = 1.949E+00 E(ENTH+SPEC) = 5.432E-04 E(SUM) = 4.872E-01
DAMP = 5.000E-01
INNER 0 : E(ENTH+SPEC) = 5.853E-01
INNER 1 : E(ENTH+SPEC) = 1.925E-01 ALPHA = 0.2457
INNER 2 : E(ENTH+SPEC) = 1.697E-01 ALPHA = 1.0000
INNER 3 : E(ENTH+SPEC) = 1.388E-01 ALPHA = 1.0000
INNER 4 : E(ENTH+SPEC) = 1.081E-01 ALPHA = 0.1100
INNER 5 : E(ENTH+SPEC) = 9.872E-02 ALPHA = 1.0000
INNER 6 : E(ENTH+SPEC) = 9.711E-02 ALPHA = 0.0250
INNER 7 : E(ENTH+SPEC) = 8.187E-02 ALPHA = 1.0000
INNER 8 : E(ENTH+SPEC) = 2.981E-02 ALPHA = 1.0000
INNER 9 : E(ENTH+SPEC) = 1.864E-02 ALPHA = 1.0000
INNER 10 : E(ENTH+SPEC) = 1.767E-02 ALPHA = 1.0000
ITER 3 E(K) = 3.071E-01 E(ENTH+SPEC) = 1.767E-02 E(SUM) = 1.929E-01

Edited by sheiko, 19 July 2011 - 07:44 PM.

[joerd]

The error means that your column did not converge the first time around. This doesn't have to be a problem since sometimes the simulator recovers after a couple of recycle iterations.
You should verify the bottom part of the simulation trace report and see if there are any errors near the end of the iterations. The simulator should report something like "Simulation Converged" (I forgot what the exact ProII lingo is).

[sheiko]

The error means that your column did not converge the first time around. This doesn't have to be a problem since sometimes the simulator recovers after a couple of recycle iterations.
You should verify the bottom part of the simulation trace report and see if there are any errors near the end of the iterations. The simulator should report something like "Simulation Converged" (I forgot what the exact ProII lingo is).

At the end i have the following, so i think this is fine. Thanks

*** PROBLEM SOLUTION REACHED

*** THIS RUN USED 125.02 PRO/II SIMULATION UNITS

SIMULATION SCIENCES INC. R PAGE H-28
PROJECT INEOS PRO/II VERSION 8.0.1 ELEC V6.6
PROBLEM SWS CALCULATION LSH
HISTORY 12/12/08
==============================================================================

*** RUN STATISTICS
STARTED 16:05:36 12/12/08 3 ERRORS
FINISHED 16:05:40 12/12/08 NO WARNINGS
RUN TIMES NO MESSAGES
INTERACTIVE 0 MIN, 0.00 SEC
CALCULATIONS 0 MIN, 4.37 SEC
TOTAL 0 MIN, 4.37 SEC

[Andrei]

sheiko,

The best way to get quick and reliable support is to contact Simsci, any legitimate user will get their attention. Their response should be the one than counts, any response from here should be considered only an opinion.
Speaking about your error, it is hard to say for sure without looking to the actual simulation file. Was the simulation solved in the end? Seams to me yes.
Your error was caused by that negative alpha value, a mathematical nonsense in a way, equivalent to divided by zero mathematical operation. Alpha gets values from 0 to 1. 0 means solving procedure divergence, and 1 convergence. There may be several reasons for this error, but most commonly this is caused by the specifications you are using. Either you have to tight specifications or conflicting specifications. The conflicts most commonly are between your input data and the program internal databanks. Since the calculations didn't stop, the error was covered by the calculations error margin and ignored. Again, this is only an opinion, the error may be very well caused by something else.
In any case, it is a good idea to check your results, and even try to run the same file on a different machine to see the differences and if you get the same error. You may want to reinitialize your simulation by exporting to .inp and importing back to PRO/II, and see if you get the same error.

Good luck

[smalawi]

try these in the order listed

update the initial estimates in the column with your solution

re-solve,

rest your solution and resolve

re-solve

export import

re-solve

if you still get this problem then contact help

[sheiko]

try these in the order listed

update the initial estimates in the column with your solution

re-solve,

rest your solution and resolve

re-solve

export import

re-solve

if you still get this problem then contact help

Thanks but how to export/import and generate .inp files?

[smalawi]

try these in the order listed

update the initial estimates in the column with your solution

re-solve,

rest your solution and resolve

re-solve

export import

re-solve

if you still get this problem then contact help

Thanks but how to export/import and generate .inp files?

file ==> export ==> keword data file

if you have cutmoized pro II then ask simsci help, it could be deferent

[sheiko]

Thank you very much

I have sent the convergence report to Simsci and they confirmed that the results are OK

Additionnal questions about Pro/II please:

1/ how to turn-off the energy-balance aspect to perform a mass-balance-only simulation?

2/ How to create a heat or duty stream (to model an interchanger by 2 simple heaters blocks connected by a heat stream for example)?

[smalawi]

Thank you very much

I have sent the convergence report to Simsci and they confirmed that the results are OK

Additionnal questions about Pro/II please:

1/ how to turn-off the energy-balance aspect to perform a mass-balance-only simulation?

2/ How to create a heat or duty stream (to model an interchanger by 2 simple heaters blocks connected by a heat stream for example)?

Hi,

Good news from simsci !

1/ not sure this is possible, there are material only operations I guess but I'm not familiar how to do this step.

2/ Pro II dont have energy stream like hysys, but it has the calculater which can add / subtract duties. you can put a calculater block with optimizer to do a similar thing. For exmaple, duty 1 + duty 2 = duty 3 in calcualtor , optimizer or controller if one element only, duty 3 / duty 4 = 1 spec by changing duty 1 & 2.

hope this helps

cheers,

SM

[Andrei]

Sheiko,

1. Personally I do not know the reasons why someone would like to turn off thermal balance calculations in a flowsheet. I would like to remind that mass balance calcs use properties variable with temperature. I cannot imagine how can you determine temperatures without heat balance.
You may choose to define same temperatures and zero duties along your flowsheet and I guess that is equivalent to performing only mass balance calcs somehow, but the actual calculations are not turned off.
2. There is "Define" statement in the specifications of any heat exchanger. You can define the duty of any exchanger to be equal to any other one from your flowsheet. You can find this in the "Specifications" window of the heat exchanger. You can also perform simple operations via define statement. "Define" statement is present in other unit operations and it doesn' have a graphical representation, you have to remember that one parameter is set by "Define" at a certain level in relation with another one elsewhere on your flowsheet.