Thanks for the help dear Trish

I use the advice and define my variable as :

positive variables

usage(n,a,h)

usage.up(n,a,h) = ((E(a)/time(a))+0.2);

loadieq(n,a,h) … (usage(n,a,h)-0.01 )* ((E(a)/time(a))- usage(n,a,h) ) =l= 0 ;

to variable usage be very close to ‘zero’ OR ‘( E(a)/time(a) )’ {are 2 parameter} but the result are wrong and all usage are between 0 and 0.01 !!

i know i can use binary variable but i have to use NLP solver.

any new advice or solution??

tanks you in advance

On Friday, November 14, 2014 11:56:12 PM UTC+3:30, Trish Gillett wrote:

So you mean that X can be either in (-inf, 0.1) or (2, inf)?

For starters, the strict inequalities are a problem. See Arne’s response from this post from a few months ago: https://groups.google.com/d/msg/gamsworld/l3Vkkss6cQQ/ofAZR-vMSQMJ

If you can settle for non-strict inequality, like X >= a and X = 0) and when x >= b (because x-a >= 0 and b-x 2 and X<0.1

How should i write this boundary in GAMS equations?

Thanks for your help in advance!

Azad

–

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.