Ray, Thanks for this suggestion. It does reduce the number of instances where the OPF cant find a solution. I was working with Matpower 3.2, now I am using version 4.0 and it is much faster.
I have another brief question, it might have been asked before, is the normal OPF formulation (cost minimisation) able to handle tap-changers? Is there a way to modify the problem, or add additional variables, so that tap-changers (or the Vf/Vt ratio) is changed to keep the system within limits? Many thanks for your help! Regards Arturo ________________________________________ From: [email protected] [[email protected]] On Behalf Of Ray Zimmerman [[email protected]] Sent: 01 March 2010 16:05 To: MATPOWER discussion forum Subject: Re: feasible problem - unfeasible OPF solution? I'm not completely sure what is going on, but I suspect it has to do with non-convexities in the problem and the combination of algorithm and starting point. I was able to solve the attached problem using the MINOPF solver, by running a simple power flow first. load testdata; mpc = struct('baseMVA', baseMVA, 'bus', bus, 'branch', branch, 'gen', gen, ... 'areas', areas, 'gencost', gencost, 'A', Au, 'l', ll, 'u', ul); mpc2 = runpf(mpc); r = runopf(mpc2, mpoption(mpopt, 'OPF_ALG', 500)) -- Ray Zimmerman Senior Research Associate 211 Warren Hall, Cornell University, Ithaca, NY 14853 phone: (607) 255-9645 On Mar 1, 2010, at 8:53 AM, Arturo Daniel Alarcón Rodríguez wrote: Dear All, I have formulated a generation capacity maximization problem, in which I want to maximize the amount of generation that can be connected without violating network constraints. To do this, I defined the cost of generation to be negative (C=-Co*P), hence the minimization of total cost (negative), becomes maximization of benefits (positive). I have also created some additional linear constraints to keep the power factor of the added generator constant. It works in most cases, however, in some cases it gives me a warning message and the optimal power flow doesn’t converge, even if I know there is a feasible solution (the feasible solution is not to connect any generation, so all X=0). I am not sure why this happens, I suspect it has something to do with the network topology. I attached one test case, with all the variables I have before running the OPF. Any help to solve this problem will be greatly appreciated. Regards Arturo ------------------------------------------------------------------------------------ Dr. Arturo D. Alarcon-Rodriguez Research Assistant Institute for Energy and Environment Department of Electronic and Electrical Engineering University of Strathclyde Royal College Building 204 George Street Glasgow G1 1XW Mobile: 07910 490 402 Email: [email protected]<mailto:[email protected]> ------------------------------------------------------------------------------------ <testdata.zip>
