I'm not sure I understand your formulation. It wasn't clear to me what you are adding to the standard formulation. You said that your new variables were equal to Va. If that is true they are not necessary since can already put costs directly on Va.
As long as you are only using the mechanisms described in Section 5.3 of the manual, you do not need to modify the Hessian, MATPOWER will take care of forming the appropriate Hessian for you. On Feb 16, 2012, at 4:05 PM, Santiago Chamba wrote: > Yes Prof. I am using Matpower 4.0. In this moment download the lasted version. > > I have thought that in my problem is not necessary modified the balance > restriction. I believe only a fictitious generator and fixed the angle. > > Other question for modified the AC OPF is necessary modified hessian or this > matrix is internal for the optimization tools. > > Thanks > > Santiago > > De: [email protected] > [mailto:[email protected]] En nombre de Ray Zimmerman > Enviado el: jueves, 16 de febrero de 2012 17:32 > Para: MATPOWER discussion forum > Asunto: Re: Modified toogle- reserves > > Aha. Yes, that is (or was) a bug and you found the solution. Sorry, I should > have remembered this issue and guessed the cause immediately. > > I assume you are using MATPOWER 4.0. This bug was fixed in 4.1. Just download > the latest version. > > -- > Ray Zimmerman > Senior Research Associate > 419A Warren Hall, Cornell University, Ithaca, NY 14853 > phone: (607) 255-9645 > > > > > > On Feb 16, 2012, at 12:21 PM, Santiago Chamba wrote: > > > Prof. Zimmerman > > Whit respect H, I think that the error is in get_cost_params.m, specifically > in line 61: > > cp.H = cp.H(idx,:); > > > I think that the cp.H = cp.H(idx,idx), what do you think about this?? This > change generates some errors??? > > Cordially, > > Santiago > De: [email protected] > [mailto:[email protected]] En nombre de Santiago > Chamba > Enviado el: jueves, 16 de febrero de 2012 14:02 > Para: 'MATPOWER discussion forum' > Asunto: RE: Modified toogle- reserves > > I want to coordinate a decentralized DC-OPF with reserves for interconnected > power system. The problem: > > <image001.jpg> > > I add the new variables Tet, what is equal to Va, then add cost alfa for Tet. > > The error is respect to the dimmension of H. For example, for a 4 buses > system and 3 generators, the new reserves variables are 3 and the new > variables for penalizing the angles are 4. In the first iteration for > calculate de cost, H is 7x3, because added the 4 new variables for penalizing > angles. > > I am going to search the error inside, cp = get_cost_params(om, name); > because cp.H=7x3. > > Best Wishes > > Santiago Chamba > > De: [email protected] > [mailto:[email protected]] En nombre de Ray Zimmerman > Enviado el: jueves, 16 de febrero de 2012 13:02 > Para: MATPOWER discussion forum > Asunto: Re: Modified toogle- reserves > > Type ... > > dbstop if error > > ... then re-run your code. When you get the error, check the dimensions of H, > w and Cw to find out where the mismatch is. There is no way for me to suggest > the cause of the error without more details. > > What changes did you make to toggle_reserves? What inputs are you using? ... > etc. > > -- > Ray Zimmerman > Senior Research Associate > 419A Warren Hall, Cornell University, Ithaca, NY 14853 > phone: (607) 255-9645 > > > > > > On Feb 16, 2012, at 8:45 AM, Santiago Chamba wrote: > > > Dear Professor Zimmerman, > > I want to add new variable and linear constraints to the OPF with reserves. > For this cause, I modified toogle_ reserves , but I have problems when the > program calculated to the new cost (compute_cost.m): > > Inner matrix dimensions must agree. > > Error in ==> opf_model.compute_cost at 105 > f = full((w' * H * w) / 2 + Cw' * w); > > The matrix H is define with others dimensions and this case no add the new > cost. > > My new variables are the nodals angles, that are penalizing when not meet > inside specific range > > Please, you explain me the error. > > Best Wishes > > Santiago Chamba > > >
