Hi Ray,This is very helpful. Thanks very much.I just read your presentation on SuperOPF project, is this going to be available soon?Cheers,Xiwen
--- On Wed, 3/3/10, Ray Zimmerman <[email protected]> wrote: From: Ray Zimmerman <[email protected]> Subject: Re: simple AC OPF problem To: "xiwen wang" <[email protected]> Cc: "MATPOWER discussion forum" <[email protected]> Date: Wednesday, March 3, 2010, 8:47 AM Hi Xiwen, I found a few issues with the data file, but was still not able to duplicate the results of the paper exactly. - your values of R and X for the branches are off by a factor of 100 (both the paper and MATPOWER use p.u.)- branch 2-3 X value was still off by factor of 10- bus 12 and the branch connecting it are not necessary, MATPOWER handles multiple gens per bus- technically, the loads should be entered as dispatchable loads (negative generators), but since none of the nodal prices are over 200, none will be curtailed and the solution will be identical With these changes I get the following for DC OPF ... ================================================================================| Generator Data |================================================================================ Gen Bus Status Pg Qg Lambda ($/MVA-hr) # # (MW) (MVAr) P Q ---- ----- ------ -------- -------- -------- -------- 1 1 1 1000.00 0.00 30.64 0.00 2 2 1 800.00 0.00 30.64 0.00 3 3 1 0.00 0.00 30.64 0.00 4 4 1 1864.41 0.00 30.64 0.00 5 5 1 1262.01 0.00 30.64 0.00 6 6 1 1081.72 0.00 30.64 0.00 7 7 1 571.51 0.00 30.64 0.00 8 8 1 1220.34 0.00 30.64 0.00 9 9 1 1500.00 0.00 30.64 0.00 10 10 1 1500.00 0.00 30.64 0.00 11 11 1 700.00 0.00 30.64 0.00 12 11 1 0.00 0.00 30.64 0.00 -------- -------- Total: 11500.00 0.00 ... and for AC OPF ... ================================================================================| Generator Data |================================================================================ Gen Bus Status Pg Qg Lambda ($/MVA-hr) # # (MW) (MVAr) P Q ---- ----- ------ -------- -------- -------- -------- 1 1 1 1000.00 5.61 31.55 0.00 2 2 1 800.00 9.10 30.98 0.00 3 3 1 0.00 55.56 32.72 0.00 4 4 1 1850.45 8.28 30.50 0.00 5 5 1 1258.95 3.38 30.61 0.00 6 6 1 1112.57 5.28 31.08 0.00 7 7 1 599.49 7.93 31.09 0.00 8 8 1 1195.87 21.78 30.35 0.00 9 9 1 1500.00 -30.41 29.68 -0.00 10 10 1 1500.00 -20.60 29.76 -0.00 11 11 1 700.00 15.98 31.93 0.00 12 11 1 51.45 15.98 31.93 0.00 -------- -------- Total: 11568.78 97.86 So, unfortunately, I'm not sure what is different between this case and the one in the paper. The base case solution in the paper looks like an AC OPF solution, since the lambdas are not identical, so it could be related to parameters not shown in the paper (voltage limits, generator VAr limits, etc.). However, my guess is that there is a typo somewhere, either in the author's MATPOWER file or in the paper ... unless, of course, there's still one in your data file that I missed. -- Ray ZimmermanSenior Research Associate211 Warren Hall, Cornell University, Ithaca, NY 14853phone: (607) 255-9645
