Download the latest version of VEDA-FE (45824) and VEDA-BE (492012)

Veda Application Installation guide


Run terminated by solver
#1
Hi,

I'm having trouble with a run not solving. The error message is 'Terminated by solver/No solution returned/NA'.
The run works under a carbon tax scenario but not when I have a cumulative constraint on CO2 emissions.

Below is some selected text from the .LST file.

Any help would be greatly appreciated.

Tamaryn

****                                    $170
**** LINE  13917 BATINCLUDE  C:\VEDA\Veda_FE\GAMS_WRKCLIMATEWORKS\iis-newtechs.dd
**** LINE     64 INPUT       C:\VEDA\Veda_FE\GAMS_WRKCLIMATEWORKS\s09_CW_cumCO2_advIND_500.RUN
482784  CAN.2035.EUSTIISGASCS101.MEUR2005 59.10708804
****                                    $170
**** LINE  13918 BATINCLUDE  C:\VEDA\Veda_FE\GAMS_WRKCLIMATEWORKS\iis-newtechs.dd
**** LINE     64 INPUT       C:\VEDA\Veda_FE\GAMS_WRKCLIMATEWORKS\s09_CW_cumCO2_advIND_500.RUN
482785  WEU.2035.EUSTIISGASCS101.MEUR2005 59.10708804
****                                    $170
**** LINE  13919 BATINCLUDE  C:\VEDA\Veda_FE\GAMS_WRKCLIMATEWORKS\iis-newtechs.dd
**** LINE     64 INPUT       C:\VEDA\Veda_FE\GAMS_WRKCLIMATEWORKS\s09_CW_cumCO2_advIND_500.RUN
482786  EEU.2035.EUSTIISGASCS101.MEUR2005 59.10708804
****                                    $170
**** LINE  13920 BATINCLUDE  C:\VEDA\Veda_FE\GAMS_WRKCLIMATEWORKS\iis-newtechs.dd
**** LINE     64 INPUT       C:\VEDA\Veda_FE\GAMS_WRKCLIMATEWORKS\s09_CW_cumCO2_advIND_500.RUN
GAMS 24.3.1  r47218 Released Jul 31, 2014 WEX-WEI x86 64bit/MS Windows                                                                                                                                                                 07/21/17 11:18:59 Page 9
TIMES -- VERSION 3.6.1 -- Restart (v3.6)
Error Messages

170  Domain violation for element
Reply
#2
The cause of termination would appear lower down in the LST file...

But the cumulative constraint may be too tight. Are the units right? Are dummy variables active for constraints? if not, you can activate in Tools - user options - import settings and reimport this scenario.

and are you OK with the fact that EUSTIISGASCS101 is not defined?
Reply
#3
Hi Amit
Thanks for getting back to me so quickly.
- So where exactly would this be? I don't seem to be able to add the lst file as an attachment to this post so I have pasted more of the file below. Do you have any tips on how to read the LST file?
- Yes, the cumulative constraint is tight but there is a backstop technology for CO2 which should still allow the model to solve.
- Dummy variables for the constraints are also activated.
- How can you tell that EUSTIISGASCS101 is not defined? I don't see this from the text in the LST file.

Thanks,
Tamaryn

               S O L V E      S U M M A R Y
     MODEL   TIMES               OBJECTIVE  OBJZ
     TYPE    LP                  DIRECTION  MINIMIZE
     SOLVER  CPLEX               FROM LINE  779776
**** SOLVER STATUS     4 Terminated By Solver     
**** MODEL STATUS      14 No Solution Returned    
**** OBJECTIVE VALUE               NA
 RESOURCE USAGE, LIMIT         NA         50000.000
 ITERATION COUNT, LIMIT        NA        999999
IBM ILOG CPLEX   24.3.1 r47218 Released Jul 31, 2014 WEI x86 64bit/MS Windows
--- GAMS/Cplex licensed for continuous and discrete problems.
Cplex 12.6.0.1
Reading parameter(s) from "C:\VEDA\Veda_FE\GAMS_WRKCLIMATEWORKS\cplex.opt"
>>  scaind 0
>>  rerun yes
>>  iis yes
>>  lpmethod 4
>>  baralg 1
>>  barcrossalg 1
>>  barorder 2
>>  THREADS=-1
Finished reading from "C:\VEDA\Veda_FE\GAMS_WRKCLIMATEWORKS\cplex.opt"
Resetting option THREADS of value -1 to 7 (8 cores detected).
Space for names approximately 71.28 Mb
Use option 'names no' to turn use of names off
LP status(3): infeasible
Cplex Time: 0.66sec (det. 193.01 ticks)
Presolve found the problem infeasible or unbounded.
Rerunning with presolve turned off.
LP status(3): infeasible
Cplex Time: 0.84sec (det. 218.80 ticks)
Unable to retrieve Cplex solution 1217.=2
No solution returned
---- 779776 Solve Read TIMES         0.031     0.047 SECS  1,069 MB
GAMS 24.3.1  r47218 Released Jul 31, 2014 WEX-WEI x86 64bit/MS Windows                                                                                                                                                                07/21/17 11:18:59 Page 15
TIMES -- VERSION 3.6.1 -- Restart (v3.6)
E x e c u t i o n

---- 779795 Assignment Z             0.000     0.047 SECS  1,069 MB      1
---- 779795 IF-ELSE                  0.000     0.047 SECS  1,069 MB
---- 779796 Put                      0.000     0.047 SECS  1,069 MB
---- 779797 Loop                     0.000     0.047 SECS  1,069 MB
---- 779798 Put                      0.000     0.047 SECS  1,069 MB
---- 779799 Put                      0.000     0.047 SECS  1,069 MB
---- 779800 Loop                     0.000     0.047 SECS  1,069 MB
---- 779801 Put                      0.000     0.047 SECS  1,069 MB
---- 779805 *** ERRORS IN OPTIMIZATION ***
**** Exec Error at line 779805: Execution halted: abort$1 '*** ERRORS IN OPTIMIZATION ***'
**** REPORT FILE SUMMARY
SCREEN C:\VEDA\Veda_FE\GAMS_WRKCLIMATEWORKS\CON
END_GAMS C:\VEDA\Veda_FE\GAMS_WRKCLIMATEWORKS\END_GAMS
QLOG C:\VEDA\Veda_FE\GAMS_WRKCLIMATEWORKS\QA_CHECK.LOG

EXECUTION TIME       =        0.047 SECONDS  1,069 MB  24.3.1 r47218 WEX-WEI

USER: Centre for Energy Policy and Technology        W140317:0740AO-WIN
      Imperial College London                                   DC10905
      License for teaching and research at degree granting institutions

**** FILE SUMMARY
Restart    C:\VEDA\Veda_FE\GAMS_SRCTIMESV361\_times.g00
Input      C:\VEDA\Veda_FE\GAMS_WRKCLIMATEWORKS\s09_CW_cumCO2_advIND_500.RUN
Output     C:\VEDA\Veda_FE\GAMS_WRKCLIMATEWORKS\s09_CW_cumCO2_advIND_500.lst
**** USER ERROR(S) ENCOUNTERED
---- 779805 GAMS Fini                3.453     3.453 SECS  1,069 MB
 
---- Profile Summary (2170 records processed)
    10.391   1.587GB    763082 Equation   EQE_COMBAL (18560)
     9.281   1.545GB    762445 Equation   EQG_COMBAL (45840)
     3.453   1.069GB    779805 GAMS Fini  
     2.875   1.248GB    762072 Equation   EQL_CAPACT (206957)
     2.671   1.646GB    770743 Equation   EQE_UCRTS (10120)
     2.187   1.646GB    766622 Equation   EQ_PEAK (1488)
     1.969   0.729GB    759329 Assignment COEF_PTRAN (904828)
     1.937   1.646GB    779776 GAMS Fini  
     1.547   1.125GB    779049 Load      
     1.500   1.646GB    779776 Solve Fini TIMES (4454982)
Reply
#4
the LST file says presolve found the problem infeasible. clearly, the backstop is not working the way you expect.

$170 below an element indicates that it has not been defined.
Reply
#5
Ok, thanks Amit. That's helpful. I'll investigate the backstop and the undefined elements!

Tamaryn
Reply
#6
Looking at it with a fixed-width font, it looks like it may be the currency, MEUR2005, which is not defined, but would not be causing any infeasibility, just the costs to be ignored. If using multiple currencies, you might also need to define currency conversions.
Note also that TIMES -- VERSION 3.6.1 is from May 2014; the current version is TIMES v4.1.3.
Reply
#7
Thanks, Antti. The currency was indeed not defined. I have added the currency conversions now from the ETSAP version of TIAM and the error is no longer there for the EUSTIISGASCS101 process but is now there for a different technology in the IIS sheet. I think this might have something to do with the user defined commodity group 'IIS_Gases'. I can't see where this is defined anywhere and it doesn't appear in the commodity group master but it does appear in veda FE when looking at the relevant processes. Any idea what the issue might be?

**** LINE 11826 BATINCLUDE C:\VEDA\Veda_FE\GAMS_WRKCLIMATEWORKS\iis-newtechs.dd
**** LINE 64 INPUT C:\VEDA\Veda_FE\GAMS_WRKCLIMATEWORKS\s09_CW_cumCO2_advIND_500.RUN
480729 MEX.2006.IISSINTER01.IIS_Gases.ANNUAL 2
**** $170
**** LINE 11827 BATINCLUDE C:\VEDA\Veda_FE\GAMS_WRKCLIMATEWORKS\iis-newtechs.dd
**** LINE 64 INPUT C:\VEDA\Veda_FE\GAMS_WRKCLIMATEWORKS\s09_CW_cumCO2_advIND_500.RUN
480730 AUS.2006.IISSINTER01.IIS_Gases.ANNUAL 2
**** $170
**** LINE 11828 BATINCLUDE C:\VEDA\Veda_FE\GAMS_WRKCLIMATEWORKS\iis-newtechs.dd
**** LINE 64 INPUT C:\VEDA\Veda_FE\GAMS_WRKCLIMATEWORKS\s09_CW_cumCO2_advIND_500.RUN
480731 CSA.2006.IISSINTER01.IIS_Gases.ANNUAL 2
**** $170
**** LINE 11829 BATINCLUDE C:\VEDA\Veda_FE\GAMS_WRKCLIMATEWORKS\iis-newtechs.dd
**** LINE 64 INPUT C:\VEDA\Veda_FE\GAMS_WRKCLIMATEWORKS\s09_CW_cumCO2_advIND_500.RUN
480732 EEU.2006.IISSINTER01.IIS_Gases.ANNUAL 2
**** $170
**** LINE 11830 BATINCLUDE C:\VEDA\Veda_FE\GAMS_WRKCLIMATEWORKS\iis-newtechs.dd
**** LINE 64 INPUT C:\VEDA\Veda_FE\GAMS_WRKCLIMATEWORKS\s09_CW_cumCO2_advIND_500.RUN
GAMS 24.3.1 r47218 Released Jul 31, 2014 WEX-WEI x86 64bit/MS Windows 07/21/17 15:10:10 Page 9
TIMES -- VERSION 3.6.1 -- Restart (v3.6)
Error Messages


170 Domain violation for element
Reply
#8
Looks strange to me, because it appears to be the same DD file as earlier, but this error did not appear earlier?

I can only suggest now to check that you didn't change your Syssettings file with respect to the User-Defined CG sheet when adding the currency conversions.  If IIS_Gases is a user-defined group, it should be defined there in SysSettings, and should appear in the GommGrp Master in VEDA-FE.
Reply
#9
Maybe the error was already there? She had sent only a section of the LST file...
Reply
#10
You are right, Amit, maybe it was. I thought all the GAMS warnings were shown, but I see now it cannot be verified.   Blush
Reply
#11
Dear Antti and Ajay

Thanks for your responses last week. It was indeed only some of the errors that I had copied and I think both were present in the previous version. I have now corrected both of these by adding the currency conversion and by adding the commodity group definition in syssettings for the user-defined commgrp. Thanks for these tips!

Unfortunately, however, the model is still not solving. Strangely, the CO2 backstop technology is being read fine in Veda FE so I am not sure why this would not be working suddenly. Also, even if I loosen the cumulative CO2 constraint the model still doesn't solve although it does solve if there is just a carbon price.

Any suggestions as to how to find what the problem is? Is there a way to see if there are competing constraints or if the model is over constrained? Dummys are turned on but as the model is not solving at all I can't import into back end to see which dummy's are being activated.

Thanks,
Tamaryn
Reply
#12
Post the scenario file where you have defined this cumulative constraint.
Reply
#13
Attached is the constraint file. This is how we have done it for quite a while and it has worked fine before so I doubt that it is the problem. The main difference between this model version and what we have run before is that I have integrated the new industry and IIS files from the ETSAP TIAM version. This includes revised files on CCS etc. so I think it is something in those files that might be causing the problem.

Tamaryn


Attached Files
.xls   Scen_UC_CUMUCO2_2dt.xls (Size: 31.5 KB / Downloads: 10)
Reply
#14
Just in case the problem still persists, small additional comments below:

If your backstop process(es) can consume TOTCO2 from all regions, and they are supposed to be able to reduce emissions in any amounts (i.e. even more than the emissions of TOTCO2 in each period), then I think it is clear that your backstop(s) either do not work, AND/OR you have not enabled negative emissions. Surely, if the backstops can reduce any amounts of TOTCO2, you should be able to reach cumulative emissions that are as low as you like.

Recall also that enabling negative emissions requires defining the associated VAR_COMNET variables with a negative lower bound, e.g. -INF (at least for TOTCO2, probably needed also for GHG). As I remember, in the ETSAP TIAM model there is a specific scenario file for that purpose, which should be included in the run. By default, most variables are non-negative in TIMES (including VAR_COMNET and VAR_COMPRD). A shortcut for defining -INF as the lower bound of VAR_COMNET(r,t,c,s) in each period is to set: COM_BNDNET(r,'0',c,'ANNUAL','N')=-1;
Reply
#15
Good idea Antti. Given all we know at this point, this looks like the most likely culprit.
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)