No. Maximizing social welfare is not the same objective as minimizing losses. Maximizing social welfare usually means both generation and demand are free to change in a way that maximizes welfare. When you are minimizing loss, you usually assume that something (typically demand) is fixed and you are minimizing losses subject to that.
In your case, if generation costs are equal and demand benefits are high enough that all available demand is served, I suppose your maximizing of social welfare does implicitly also minimize cost, but that’s only because minimizing cost is the only option the optimizer has to maximize welfare. But this is a consequence of the input parameters, not inherent to the problem formulation. So, to reiterate, maximizing social welfare does not always minimize loss … only when using very special (and unrealistic) values of demand benefits and generator costs. -- Ray Zimmerman Senior Research Associate B30 Warren Hall, Cornell University, Ithaca, NY 14853 phone: (607) 255-9645 On Jan 15, 2014, at 6:43 AM, Silvio Miceli <[email protected]> wrote: > Dear Prof. Zimmerman, > > Thank you. Can I say total power losses are minimized If I maximize social > welfare considering equal cost of generators according to the following > formula > > <image.png> > > Best regards, > > S.M > > > On Tue, Jan 14, 2014 at 3:23 PM, Ray Zimmerman <[email protected]> wrote: > Check out the FAQ (just posted) … > > http://www.pserc.cornell.edu/matpower/#lossminimization > > -- > Ray Zimmerman > Senior Research Associate > B30 Warren Hall, Cornell University, Ithaca, NY 14853 > phone: (607) 255-9645 > > > > On Jan 14, 2014, at 6:36 AM, Silvio Miceli <[email protected]> wrote: > >> Dear Dr. Zimmerman, >> >> I want to minimize loss in MATPOWER. I have written the following code to >> obtain losses. How can I minimize it with the MATPOWER solvers? >> >> >> >> define_constants; >> mpc=loadcase('case9') >> [r, co, cb, f, dispatch, success, et] = runmarket (mpc); >> loss=r.branch(:,14)+r.branch(:,16); >> >> >> Best regards, >> >> > >
