The reason is different, but my advice is the same … if at all possible, use Gurobi or CPLEX as your solver.
MATPOWER’s miqps_matpower() <http://www.pserc.cornell.edu//matpower/docs/ref/matpower6.0b1/miqps_matpower.html> function, a wrapper around mixed-integer quadratic and linear programming solvers such as Matlab’s intlinprog(), first solves the mixed integer problem, then, by default, in order to solve for the constraint shadow prices, it solves the problem again with the integer variables fixed. In your case, the exit flag for the pricing run was different from the mixed integer run. In fact, linprog() seems to be claiming that the primal and dual problems are both infeasible (exitflag = -5, see linprog exitflag docs <http://www.mathworks.com/help/optim/ug/linprog.html#outputarg_exitflag>), which intlinprog() apparently didn’t find to be the case. Another option is to just disable the pricing re-run … mpopt = mpoption(mpopt, 'most.skip_prices', 1); … but then your results don’t include … you know … prices. Ray > On Aug 1, 2016, at 7:18 AM, Pouria Maghouli <magh...@gmail.com> wrote: > > Dear > I repeat your experience and received the following error which is different!: > Error using miqps_ot (line 346) > miqps_ot: EXITFLAG from price computation stage = -5 > > Error in miqps_matpower (line 219) > [x, f, eflag, output, lambda] = ... > > Error in most (line 2071) > [mdo.QP.x, mdo.QP.f, mdo.QP.exitflag, mdo.QP.output, ... > > On Mon, Jul 25, 2016 at 10:30 PM, Ray Zimmerman <r...@cornell.edu > <mailto:r...@cornell.edu>> wrote: > I believe this is a known bug in the current version of intlingprog(), part > of Matlab’s Optimization Toolbox. I strongly recommend using Gurobi or CPLEX > for running MOST if at all possible. > > Ray > > >> On Jul 24, 2016, at 10:56 AM, Jason Jett <jlj...@ieee.org >> <mailto:jlj...@ieee.org>> wrote: >> >> I have recently installed the latest MATPOWER version and am going through >> the setup of MOST. >> I ran the following commands to prepare and execute the example MOST case: >> >> >> mpc=loadcase('ex_case3b'); >> >> transmat=ex_transmat(12); >> >> xgd=loadxgendata('ex_xgd_uc',mpc); >> >> [iwind, mpc, xgd] = addwind(ex_wind_uc, mpc, xgd); >> >> [iess, mpc, xgd, sd] = addstorage(ex_storage, mpc, xgd); >> >> contab = ex_contab(); >> >> profiles = getprofiles(ex_load_profile); >> >> profiles = getprofiles(ex_wind_profile, profiles, iwind); >> >> mdi = loadmd(mpc, transmat, xgd, sd, contab, profiles); >> >> mpopt = mpoption('verbose', 0); >> >> tic;mdo = most(mdi, mpopt);toc >> >> >> I received the following error: >> Error using intlinprog (line 131) >> INTLINPROG encountered an internal error that caused the solution to lose >> integer and/or linear constraint feasibility. We are sorry for the >> inconvenience. >> >> Please contact technical support for assistance with your problem, quoting >> the code "-2_3". >> >> Error in miqps_ot (line 281) >> intlinprog(c, intcon, Ai, bi, Ae, be, xmin, xmax, ot_opt); >> >> Error in miqps_matpower (line 220) >> miqps_ot(H, c, A, l, u, xmin, xmax, x0, vtype, opt); >> >> Error in most (line 2072) >> mdo.QP.lambda ] = miqps_matpower( mdi.QP.H, mdi.QP.C, ... >> >> Following are the MATLAB and MATPOWER versions: >> >> mpver >> >> MATPOWER Version 6.0b1 01-Jun-2016 >> MATLAB Version 9.0 10-Feb-2016 Release: (R2016a) >> Optimization Toolbox Version 7.4 10-Feb-2016 Release: (R2016a) >> MIPS Version 1.2.1 01-Jun-2016 >> MOST Version 1.0b1 01-Jun-2016 >> SDP_PF -- not installed -- >> YALMIP -- not installed -- >> BPMPD_MEX -- not installed -- >> CLP -- not installed -- >> CPLEX -- not installed -- >> GLPK -- not installed -- >> Gurobi -- not installed -- >> IPOPT -- not installed -- >> KNITRO -- not installed -- >> MINOPF -- not installed -- >> MOSEK -- not installed -- >> PARDISO -- not installed -- >> PDIPMOPF -- not installed -- >> SCPDIPMOPF -- not installed -- >> SDPT3 -- not installed -- >> SeDuMi -- not installed -- >> TRALMOPF -- not installed -- >> Architecture: PCWIN64 >> >> Any help on resolving this error would be appreciated. >> >> Thanks >> Jason Jett > >
