Setting the RATE_A column to zero will remove the flow limit constraint, for both transmission lines and transformers. Setting it to a large number leaves the constraint in the problem, but the constraint will never bind if the limit is large enough. I recommend the former.
However, if you want to simulate adding a 2nd identical parallel branch you do need to change the impedance between the connected buses. You can do this by either adding a duplicate branch explicitly or by simply dividing the R and X of the existing branch by 2. -- Ray Zimmerman Senior Research Associate 419A Warren Hall, Cornell University, Ithaca, NY 14853 phone: (607) 255-9645 On Dec 6, 2012, at 11:16 AM, Panagis Vovos <[email protected]> wrote: > Hello Vagelis, > > You should make the amendment in the mpc.branch array. I am not sure > about transformers, but thinking about their model probably it will > work there,too. Why don't you just give it a try? Try first with a > very tight limit, so that the OPF will not converge and then run again > with zeros at the appropriate branch columns. > > Best regards, > > Panagis Vovos > > On 6 December 2012 17:52, Evangelos Galinas <[email protected]> wrote: >> Hallo Panagis, >> >> and thanks for your answer. >> >> so just making the mpc.gen(:,6)=0 will disregard the constraint without >> having to change the impedance. This is great. >> >> and if for example I do not want the transformers to be a bottleneck, can I >> just make their 6th column 0 while keeping the same correct value for the >> line branches? >> >> Thanks a lot. >> >> Vagelis >> >>> Date: Thu, 6 Dec 2012 12:12:52 +0200 >>> Subject: Re: Disregard the branch constraints >>> From: [email protected] >>> To: [email protected] >> >>> >>> Dear Vagelis, >>> >>> I think there is no flow constraint if you set the MVA limit to 0 in >>> "branches". I remember such an if/case in the code, but I am not sure >>> if this is about the constraint. >>> >>> Best regards, >>> >>> Panagis Vovos >>> >>> On 5 December 2012 20:52, Evangelos Galinas <[email protected]> >>> wrote: >>>> Dear all, >>>> >>>> I have a small question regarding the max P constraints of the branches. >>>> >>>> With Regard to how matPower is implemented, how would be better to >>>> disregard >>>> the branch constraints? >>>> >>>> I cannot just make the constraints too high (e.g. 9999) because the R >>>> and X >>>> do not change so this will affect the flow. >>>> >>>> So, does multiplying the branch's actual limit by 10 and at the same >>>> time >>>> decreasing its R and X tenfold solve the issue? >>>> >>>> Isn't that the correct way to simulate a second // branch to the initial >>>> one? (i.e. Pmax (x)2 and R (/)2 X (/)2 ) >>>> >>>> Regards, >>>> >>>> Vagelis >>>> >>>> >>> >
