An equation is inconsistent with other equations in the pre-triangular part of the model.

GAMS
1

Hi all,

I use CONOPT to solve a model to initial equilibrium by pinning all variables to their initial equalibrium values. The returned solution is as follows:

S O L V E S U M M A R Y

MODEL FDIHE OBJECTIVE GDP
TYPE NLP DIRECTION MAXIMIZE
SOLVER CONOPT FROM LINE 1751

**** SOLVER STATUS 1 Normal Completion
**** MODEL STATUS 4 Infeasible
**** OBJECTIVE VALUE 0.0000

RESOURCE USAGE, LIMIT 0.031 1000.000
ITERATION COUNT, LIMIT 0 2000000000
EVALUATION ERRORS 0 0
CONOPT 3 24.2.1 r43572 Released Dec 9, 2013 LEG x86_64/Linux


C O N O P T 3 version 3.15M
Copyright (C) ARKI Consulting and Development A/S
Bagsvaerdvej 246 A
DK-2880 Bagsvaerd, Denmark


** Warning ** The variance of the derivatives in the initial
point is large (= 8.1 ). A better initial
point, a better scaling, or better bounds on the
variables will probably help the optimization.


** An equation is inconsistent with other equations in the
pre-triangular part of the model.

Residual= 8.24578221E-02
Tolerance (Rtnwtr)= 2.00000000E-08

The solution order of the critical pretriangular equations is:

Equation INVSAVEQ(ROW) solved with respect to the slack
After the last operation the residual is still -0.0824578221327101

CONOPT time Total 0.034 seconds
of which: Function evaluations 0.009 = 26.5%
1st Derivative evaluations 0.010 = 29.4%


**** ERRORS/WARNINGS IN EQUATION INVSAVEQ(ROW)
1 error(s): Inconsistency in pre-triangular part of model.

I have checked the correctness of equation INVSAVEQ(ROW). The right hand of the equation=2650393.843 and the left hand of the equation=2650393.926. The difference is -0.0824578221327101, as the residual returned by the solution. Is this difference causing the inconsistency in the system? Is this related to the Tolerance (Rtnwtr)= 2.00000000E-08? (I tried to increase the tolerance, but failed.)

Any suggestions are appreciated.

Best,
Qiaomin








\

To unsubscribe from this group and stop receiving emails from it, send an email to gamsworld+unsubscribe@googlegroups.com.
To post to this group, send email to gamsworld@googlegroups.com.
Visit this group at http://groups.google.com/group/gamsworld.
For more options, visit https://groups.google.com/d/optout.

2

Qiaomin:



The feasibility test is scaled with a number that represent the size of the equation. In this case it will be the absolute value of the right hand size or 2.65e6. The scaled infeasibility is therefore 0.08/2.65e6 = 3.e-8 which is larger than the tolerance Rtnwtr = 2.e-8. If you want to change it, do it in an options file and remember to tell CONOPT to use the option file. If you have not used solver options before I recommend that you read the introductory section in the Solver Manual.



Regards



Arne

Arne Stolbjerg Drud

ARKI Consulting & Development A/S

Bagsvaerdvej 246A, DK-2880 Bagsvaerd, Denmark

Phone: (+45) 44 49 03 23, Fax: (+45) 44 49 03 33, email: adrud@arki.dk



From: gamsworld@googlegroups.com [mailto:gamsworld@googlegroups.com] On Behalf Of Qiaomin Li
Sent: Monday, March 17, 2014 4:35 AM
To: gamsworld@googlegroups.com
Subject: An equation is inconsistent with other equations in the pre-triangular part of the model.



Hi all,



I use CONOPT to solve a model to initial equilibrium by pinning all variables to their initial equalibrium values. The returned solution is as follows:



S O L V E S U M M A R Y

MODEL FDIHE OBJECTIVE GDP
TYPE NLP DIRECTION MAXIMIZE
SOLVER CONOPT FROM LINE 1751

**** SOLVER STATUS 1 Normal Completion
**** MODEL STATUS 4 Infeasible
**** OBJECTIVE VALUE 0.0000

RESOURCE USAGE, LIMIT 0.031 1000.000
ITERATION COUNT, LIMIT 0 2000000000
EVALUATION ERRORS 0 0
CONOPT 3 24.2.1 r43572 Released Dec 9, 2013 LEG x86_64/Linux


C O N O P T 3 version 3.15M
Copyright (C) ARKI Consulting and Development A/S
Bagsvaerdvej 246 A
DK-2880 Bagsvaerd, Denmark


** Warning ** The variance of the derivatives in the initial
point is large (= 8.1 ). A better initial
point, a better scaling, or better bounds on the
variables will probably help the optimization.


** An equation is inconsistent with other equations in the
pre-triangular part of the model.

Residual= 8.24578221E-02
Tolerance (Rtnwtr)= 2.00000000E-08

The solution order of the critical pretriangular equations is:

Equation INVSAVEQ(ROW) solved with respect to the slack
After the last operation the residual is still -0.0824578221327101

CONOPT time Total 0.034 seconds
of which: Function evaluations 0.009 = 26.5%
1st Derivative evaluations 0.010 = 29.4%


**** ERRORS/WARNINGS IN EQUATION INVSAVEQ(ROW)
1 error(s): Inconsistency in pre-triangular part of model.



I have checked the correctness of equation INVSAVEQ(ROW). The right hand of the equation=2650393.843 and the left hand of the equation=2650393.926. The difference is -0.0824578221327101, as the residual returned by the solution. Is this difference causing the inconsistency in the system? Is this related to the Tolerance (Rtnwtr)= 2.00000000E-08? (I tried to increase the tolerance, but failed.)



Any suggestions are appreciated.



Best,

Qiaomin
















\

To unsubscribe from this group and stop receiving emails from it, send an email to gamsworld+unsubscribe@googlegroups.com.
To post to this group, send email to gamsworld@googlegroups.com.
Visit this group at http://groups.google.com/group/gamsworld.
For more options, visit https://groups.google.com/d/optout.

–
To unsubscribe from this group and stop receiving emails from it, send an email to gamsworld+unsubscribe@googlegroups.com.
To post to this group, send email to gamsworld@googlegroups.com.
Visit this group at http://groups.google.com/group/gamsworld.
For more options, visit https://groups.google.com/d/optout.

3

Reply-to: gamsworld@googlegroups.com

Thanks Arne,

Following your advice, I tried to change the tolerance through an option file. The following is the code I learned from CONOPT Sovler manual:

MODEL FDIHE /ALL/;

FDIHE.SCALEOPT=1;
FDIHE.OPTFILE=1;
FILE OPT /CONOPT.OPT/;
PUT OPT;
PUT ‘RTNWTR=6.E-7’/;
PUTCLOSE OPT;
SOLVE FDIHE MAXIMIZING GDP USING NLP;


Because I scaled the model, there is a scale option. I want to increase Rtnwtr to 6.e-7. But the returned solution is as follows:
** An equation is inconsistent with other equations in the
pre-triangular part of the model.

Residual= 1.00000021E-02
Tolerance (Rtnwtr)= 2.00000000E-08

Does that mean the change of tolarance is unsuccessful? If yes, what is wrong with my coding?

Many thanks and best regards,
Qiaomin



On Tuesday, March 18, 2014 5:34:20 AM UTC+13, Arne Stolbjerg Drud wrote:

Qiaomin:



The feasibility test is scaled with a number that represent the size of the equation. In this case it will be the absolute value of the right hand size or 2.65e6. The scaled infeasibility is therefore 0.08/2.65e6 = 3.e-8 which is larger than the tolerance Rtnwtr = 2.e-8. If you want to change it, do it in an options file and remember to tell CONOPT to use the option file. If you have not used solver options before I recommend that you read the introductory section in the Solver Manual.



Regards



Arne

Arne Stolbjerg Drud

ARKI Consulting & Development A/S

Bagsvaerdvej 246A, DK-2880 Bagsvaerd, Denmark

Phone: (+45) 44 49 03 23, Fax: (+45) 44 49 03 33, email: ad…@arki.dk



From: gams...@googlegroups.com [mailto:gams...@googlegroups.com] On Behalf Of Qiaomin Li
Sent: Monday, March 17, 2014 4:35 AM
To: gams...@googlegroups.com
Subject: An equation is inconsistent with other equations in the pre-triangular part of the model.



Hi all,



I use CONOPT to solve a model to initial equilibrium by pinning all variables to their initial equalibrium values. The returned solution is as follows:



S O L V E S U M M A R Y

MODEL FDIHE OBJECTIVE GDP
TYPE NLP DIRECTION MAXIMIZE
SOLVER CONOPT FROM LINE 1751

**** SOLVER STATUS 1 Normal Completion
**** MODEL STATUS 4 Infeasible
**** OBJECTIVE VALUE 0.0000

RESOURCE USAGE, LIMIT 0.031 1000.000
ITERATION COUNT, LIMIT 0 2000000000
EVALUATION ERRORS 0 0
CONOPT 3 24.2.1 r43572 Released Dec 9, 2013 LEG x86_64/Linux


C O N O P T 3 version 3.15M
Copyright (C) ARKI Consulting and Development A/S
Bagsvaerdvej 246 A
DK-2880 Bagsvaerd, Denmark


** Warning ** The variance of the derivatives in the initial
point is large (= 8.1 ). A better initial
point, a better scaling, or better bounds on the
variables will probably help the optimization.


** An equation is inconsistent with other equations in the
pre-triangular part of the model.

Residual= 8.24578221E-02
Tolerance (Rtnwtr)= 2.00000000E-08

The solution order of the critical pretriangular equations is:

Equation INVSAVEQ(ROW) solved with respect to the slack
After the last operation the residual is still -0.0824578221327101

CONOPT time Total 0.034 seconds
of which: Function evaluations 0.009 = 26.5%
1st Derivative evaluations 0.010 = 29.4%


**** ERRORS/WARNINGS IN EQUATION INVSAVEQ(ROW)
1 error(s): Inconsistency in pre-triangular part of model.



I have checked the correctness of equation INVSAVEQ(ROW). The right hand of the equation=2650393.843 and the left hand of the equation=2650393.926. The difference is -0.0824578221327101, as the residual returned by the solution. Is this difference causing the inconsistency in the system? Is this related to the Tolerance (Rtnwtr)= 2.00000000E-08? (I tried to increase the tolerance, but failed.)



Any suggestions are appreciated.



Best,

Qiaomin
















\

To unsubscribe from this group and stop receiving emails from it, send an email to gamsworld+...@googlegroups.com.
To post to this group, send email to gams...@googlegroups.com.
Visit this group at http://groups.google.com/group/gamsworld.
For more options, visit https://groups.google.com/d/optout.

–
To unsubscribe from this group and stop receiving emails from it, send an email to gamsworld+unsubscribe@googlegroups.com.
To post to this group, send email to gamsworld@googlegroups.com.
Visit this group at http://groups.google.com/group/gamsworld.
For more options, visit https://groups.google.com/d/optout.

4

Reply-to: gamsworld@googlegroups.com



On Tuesday, March 18, 2014 at 10:11:40 AM UTC+4:30, Qiaomin Li wrote:

Thanks Arne,

Following your advice, I tried to change the tolerance through an option file. The following is the code I learned from CONOPT Sovler manual:

MODEL FDIHE /ALL/;

FDIHE.SCALEOPT=1;
FDIHE.OPTFILE=1;
FILE OPT /CONOPT.OPT/;
PUT OPT;
PUT ‘RTNWTR=6.E-7’/;
PUTCLOSE OPT;
SOLVE FDIHE MAXIMIZING GDP USING NLP;


Because I scaled the model, there is a scale option. I want to increase Rtnwtr to 6.e-7. But the returned solution is as follows:
** An equation is inconsistent with other equations in the
pre-triangular part of the model.

Residual= 1.00000021E-02
Tolerance (Rtnwtr)= 2.00000000E-08

Does that mean the change of tolarance is unsuccessful? If yes, what is wrong with my coding?

Many thanks and best regards,
Qiaomin



On Tuesday, March 18, 2014 5:34:20 AM UTC+13, Arne Stolbjerg Drud wrote:

Qiaomin:



The feasibility test is scaled with a number that represent the size of the equation. In this case it will be the absolute value of the right hand size or 2.65e6. The scaled infeasibility is therefore 0.08/2.65e6 = 3.e-8 which is larger than the tolerance Rtnwtr = 2.e-8. If you want to change it, do it in an options file and remember to tell CONOPT to use the option file. If you have not used solver options before I recommend that you read the introductory section in the Solver Manual.



Regards



Arne

Arne Stolbjerg Drud

ARKI Consulting & Development A/S

Bagsvaerdvej 246A, DK-2880 Bagsvaerd, Denmark

Phone: (+45) 44 49 03 23, Fax: (+45) 44 49 03 33, email: ad…@arki.dk



From: gams...@googlegroups.com [mailto:gams...@googlegroups.com] On Behalf Of Qiaomin Li
Sent: Monday, March 17, 2014 4:35 AM
To: gams...@googlegroups.com
Subject: An equation is inconsistent with other equations in the pre-triangular part of the model.



Hi all,



I use CONOPT to solve a model to initial equilibrium by pinning all variables to their initial equalibrium values. The returned solution is as follows:



S O L V E S U M M A R Y

MODEL FDIHE OBJECTIVE GDP
TYPE NLP DIRECTION MAXIMIZE
SOLVER CONOPT FROM LINE 1751

**** SOLVER STATUS 1 Normal Completion
**** MODEL STATUS 4 Infeasible
**** OBJECTIVE VALUE 0.0000

RESOURCE USAGE, LIMIT 0.031 1000.000
ITERATION COUNT, LIMIT 0 2000000000
EVALUATION ERRORS 0 0
CONOPT 3 24.2.1 r43572 Released Dec 9, 2013 LEG x86_64/Linux


C O N O P T 3 version 3.15M
Copyright (C) ARKI Consulting and Development A/S
Bagsvaerdvej 246 A
DK-2880 Bagsvaerd, Denmark


** Warning ** The variance of the derivatives in the initial
point is large (= 8.1 ). A better initial
point, a better scaling, or better bounds on the
variables will probably help the optimization.


** An equation is inconsistent with other equations in the
pre-triangular part of the model.

Residual= 8.24578221E-02
Tolerance (Rtnwtr)= 2.00000000E-08

The solution order of the critical pretriangular equations is:

Equation INVSAVEQ(ROW) solved with respect to the slack
After the last operation the residual is still -0.0824578221327101

CONOPT time Total 0.034 seconds
of which: Function evaluations 0.009 = 26.5%
1st Derivative evaluations 0.010 = 29.4%


**** ERRORS/WARNINGS IN EQUATION INVSAVEQ(ROW)
1 error(s): Inconsistency in pre-triangular part of model.



I have checked the correctness of equation INVSAVEQ(ROW). The right hand of the equation=2650393.843 and the left hand of the equation=2650393.926. The difference is -0.0824578221327101, as the residual returned by the solution. Is this difference causing the inconsistency in the system? Is this related to the Tolerance (Rtnwtr)= 2.00000000E-08? (I tried to increase the tolerance, but failed.)



Any suggestions are appreciated.



Best,

Qiaomin
















\

To unsubscribe from this group and stop receiving emails from it, send an email to gamsworld+...@googlegroups.com.
To post to this group, send email to gams...@googlegroups.com.
Visit this group at http://groups.google.com/group/gamsworld.
For more options, visit https://groups.google.com/d/optout.






Hi dear Qiaomin
I have a similar problem. Would you please say that how could you change rtnwtr scale? (we should write codes in main page of gams or write in another place?) I don’t know anythings about conopt control program!!
Regards
Hamid

–
To unsubscribe from this group and stop receiving emails from it, send an email to gamsworld+unsubscribe@googlegroups.com.
To post to this group, send email to gamsworld@googlegroups.com.
Visit this group at http://groups.google.com/group/gamsworld.
For more options, visit https://groups.google.com/d/optout.