It doesn't seem possible that the new smartmarket.m could be the cause of the error below. Have you done a clean install of MATPOWER 4.1 and then replaced smartmarket.m with the new version? Earlier you reported errors that I could not reproduce, leading me to believe you may have a modified version of MATPOWER. Do all of the MATPOWER tests run correctly?
-- Ray Zimmerman Senior Research Associate 419A Warren Hall, Cornell University, Ithaca, NY 14853 phone: (607) 255-9645 On Mar 25, 2012, at 12:08 AM, Dailan Xu wrote: > Dear Dr. Zimmerman, > > After replacing the new code of smartmkt.m I get SMARTMARKET: non-convergent > UOPF??? Error using ==> mtimes in some cases > while with the previous one I had done runmarket and it was working well. > Please help me. > > Best Regards > > D. Xu > > > > On Fri, Feb 24, 2012 at 19:56, Ray Zimmerman <[email protected]> wrote: > I don't know what you mean by "generation capacity" being negative. I > normally think of PMAX as the generation capacity. If you mean the dispatch, > then you obviously have a case where PMIN is negative, which doesn't make > sense for a generator, but does for a dispatchable load. > > I suggest you read carefully the MATPOWER manual, especially the sections > describing the OPF and Appendix F. Make sure you understand exactly what > problem the OPF is solving and how runmarket is using the bids and offers you > provide to set up the inputs to the OPF. When all else fails, read the code. > It *is* the documentation in some cases. Aside from that, find a good > textbook on optimal power flow. > > By looking at the constraints on the dispatch in question (PMIN, PMAX and any > user constraints you've imposed, such as power factor) and looking at the > generator cost function and comparing it to the nodal price at the bus > (LAM_P), it should be clear why the dispatch is negative. > > > -- > Ray Zimmerman > Senior Research Associate > 419A Warren Hall, Cornell University, Ithaca, NY 14853 > phone: (607) 255-9645 > > > > > On Feb 24, 2012, at 1:26 PM, Dailan Xu wrote: > >> Dear Dr. Zimmerman, >> >> I know you don't have time. I am so sorry I disturbed you again. At least, >> please introduce me some references to read and understand when and in what >> situations a runmarket 's results (generation capacity) will be negative? >> >> Best Regards >> >> Dailan >> >> >> >> On Thu, Feb 23, 2012 at 17:10, Ray Zimmerman <[email protected]> wrote: >> There could be many reasons, unfortunately, I don't have the time to >> investigate for you. If you think it is incorrect, I suggest you check the >> inputs to the underlying OPF to make sure that when the OPF is called it has >> the expected gencost and PMIN, PMIN parameters. >> >> -- >> Ray Zimmerman >> Senior Research Associate >> 419A Warren Hall, Cornell University, Ithaca, NY 14853 >> phone: (607) 255-9645 >> >> >> >> >> On Feb 23, 2012, at 10:55 AM, Dailan Xu wrote: >> >>> Thank you very much. I know you don't have time. I only want you ask why I >>> get negative active capacities? When It is happened in MATPOWER? >>> Best Regards >>> >>> >>> >>> >>> >>> On Mon, Feb 20, 2012 at 16:48, Dailan Xu <[email protected]> wrote: >>> Dear Dr. Zimmerman, >>> >>> I want to do runmarket with active and reactive offers and bids. I receive >>> the following message: >>> >>> How can I solve the problem? I have attached the case files. >>> By the way, I do not want to distribute my case file because of this reason >>> I only sent to your email. >>> >>> Reference to non-existent field 'Q'. >>> >>> Error in smartmkt (line 139) >>> nQo = size(offers.Q.qty, 2); >>> >>> Error in runmarket (line 141) >>> [co, cb, r, dispatch, success] = smartmkt(mpc, offers, bids, mkt, mpopt); >>> >>> Best Regards >>> >>> Dailan >>> >>> >>> >> >> > >
