I quickly tried on a very simple 3 buses system (pls check the attached figure) modified based on the t_case9_dcline system to test the islands.
Firstly I set the status of line 1-2 equal to 1, then both PF and OPF works fine. In this case, with the replacement of dummy generators for the DC lines, the whole system is still connected. Then, I set the status of line 1-2 equal to 0, then both PF and OPF do not converge. In this case, by using the dummy generators to model the DC line, the network was split into to 2 islands. By the way, I put a generator at bus 2 so each island should have a reference bus (suppose Matpower would use the first PV bus as the reference bus) Dr. TAO HUANG Senior Assistant Researcher SESAME -Executive vice coordinator Politecnico di Torino Dipartimento Energia Corso Duca degli Abruzzi, 24 10129 Torino - Italy tel. +39 011 090 7117 fax +39 011 090 7199 e-mail [email protected] From: [email protected] [mailto:[email protected]] On Behalf Of Ray Zimmerman Sent: Wednesday, April 17, 2013 5:52 PM To: MATPOWER discussion forum Subject: Re: Modelling transformer and phase shifter But as long as the dummy generators are dispatchable and appropriately coupled I don't see why you wouldn't get an optimal solution for the whole system (as with our DC line implementation described in Section 6.5.3 of the User's Manual <http://www.pserc.cornell.edu/matpower/manual.pdf> . -- Ray Zimmerman Senior Research Associate 419A Warren Hall, Cornell University, Ithaca, NY 14853 phone: (607) 255-9645 On Apr 17, 2013, at 10:35 AM, Tao HUANG <[email protected]> wrote: For power flow maybe yes (neglecting the angle differences and overall system losses), but for OPF, the optimal result for each island may not be the optimal solution for the whole system, as the islands are formed purely due to the replacement of a transformer (or a DC line) by two dummy generators. From: <mailto:[email protected]> [email protected] [mailto:bounce-82639400-8559090@ <http://list.cornell.edu/> list.cornell.edu] On Behalf Of Ray Zimmerman Sent: Wednesday, April 17, 2013 4:30 PM To: MATPOWER discussion forum Subject: Re: Modelling transformer and phase shifter As long as each island has a voltage angle reference I don't think the islands should present a problem. -- Ray Zimmerman Senior Research Associate 419A Warren Hall, Cornell University, Ithaca, NY 14853 phone: (607) 255-9645 On Apr 17, 2013, at 5:05 AM, Tao HUANG < <mailto:[email protected]> [email protected]> wrote: Dear Ray and Kartik, Thanks for your reply. I guess the easiest way to utilize the current version of matpower is to use dummy generators as Ray indicated. But what I concerned of using this method is the problem of split the network into many islands when replacing the transformer/phase shifter with 2 generators. If this happens, we cannot get the OPF/PF results. I would also skip using heuristic algorithms as I think they cannot give a "stable" and "coherent" solution. Thanks again for your suggestions. Tao From: <mailto:[email protected]> [email protected] [ <mailto:[email protected]> mailto:[email protected]] On Behalf Of Kartik Pandya Sent: Wednesday, April 17, 2013 6:35 AM To: MATPOWER discussion forum Subject: Re: Modelling transformer and phase shifter You may use AI method like PSO to consider TAPS as control variables. please refer "Optimal reactive power dispatch using Particle swarm optimization" file at matlab file exchange for more details. in that file i had optimized TAPS of transformers connected in line no. 11,12,15, and 36 for IEEE 30 bus test system. _____ From: Ray Zimmerman < <mailto:[email protected]> [email protected]> To: MATPOWER discussion forum < <mailto:[email protected]> [email protected]> Sent: Wednesday, 17 April 2013 12:49 AM Subject: Re: Modelling transformer and phase shifter It is correct that in the current MATPOWER OPF, the TAP and SHIFT parameters are fixed input parameters, not variables. I don't think there is a way to fake it with the AC OPF. That is, I think you really need to include them as variables in the flow equations and corresponding derivatives. There may be a way to fake it with dummy injections in the DC OPF, but I haven't really thought about it carefully. Please let us know if you come up with something that works. And of course, others can feel free to suggest ideas. -- Ray Zimmerman Senior Research Associate 419A Warren Hall, Cornell University, Ithaca, NY 14853 phone: (607) 255-9645 On Apr 16, 2013, at 9:04 AM, Tao HUANG < <mailto:[email protected]> [email protected]> wrote: Dear Dr. Ray, Speaking of the modeling transformer and phase shifter, I would like to see if the adjustment of the two parameters (tap ratios and shift angles) along with the generators and loads could get better results of OPF in terms of cost and convergence (due to the line limits in some cases). do you have any suggestions as to how to make the tap ratios and shift angles as variables in the OPF without using iterative calculation (like blindly or using some heuristic algorithms to modify them and redo the OPF)? If I understand it correctly, these two values are fixed before the PF & OPF calculation in terms of Y matrix at present in Matpower. I was considering to model the transformer/ phase shifter as two generators in respect to the flow limits, but it may split the network into two islands in some cases. So, do you have any suggestion on this as well? I guess it also applies to the DC line when it is the only connection between two parts of the network. Thanks a lot Best wishes, Tao From: <mailto:[email protected]> [email protected] [mailto:bounce-82280449-8559090@ <http://list.cornell.edu/> list.cornell.edu] On Behalf Of Ray Zimmerman Sent: Tuesday, April 16, 2013 2:31 PM To: MATPOWER discussion forum Subject: Re: Modelling transformer and phase shifter Yes, for a transformer you will typically have an off-nominal taps ratio, i.e. branch(b, TAP) ~= 1 (or 0, which signifies a normal transmission line). Similarly, if you have a phase shift, you will have branch(b, SHIFT) ~= 0. How the transformer TAP and SHIFT parameters affect the amount of load able to be dispatched depends on the network, so it could go either way. -- Ray Zimmerman Senior Research Associate 419A Warren Hall, Cornell University, Ithaca, NY 14853 phone: (607) 255-9645 On Apr 16, 2013, at 8:18 AM, Jiashen Teh < <mailto:[email protected]> [email protected]> wrote: Dear Dr Ray, According to your manual , in section '3.2 Branches' it is mentioned that: 'All transmission lines, transformers and phase shifters are modeled with a common branch model......' Does this mean, if I want to include modelling of transformer at a branch, I would set the column 9 and 10 of mpc.branch which govern the transformer turn ratio, N ? I notice that for values other than zero for both columns (in a manner to increase N or reduce N), the network (6 bus) will encounter increase of load not able to be dispatched. Isn't is load dispatch-able should increase? Yours sincerely, Jiashen Teh
function mpc = t_case3_dcline
%T_CASE9_DCLINE Same as T_CASE9_OPFV2 with addition of DC line data.
% Please see CASEFORMAT for details on the case file format.
%
% See also: TOGGLE_DCLINE, IDX_DCLINE.
% MATPOWER
% $Id: t_case9_dcline.m,v 1.1 2011/12/08 20:34:20 cvs Exp $
%% MATPOWER Case Format : Version 2
mpc.version = '2';
%%----- Power Flow Data -----%%
%% system MVA base
mpc.baseMVA = 100;
%% bus data
% bus_i type Pd Qd Gs Bs area Vm Va
baseKV zone Vmax Vmin
mpc.bus = [
1 3 10 5 0 0 1 1 0
345 1 1.1 0.9;
2 2 0 0 0 0 1 1 0
345 1 1.1 0.9;
% 30 2 0 0 0 0 1 1 0
345 1 1.1 0.9;
% 4 1 0 0 0 0 1 1 0
345 1 1.1 0.9;
% 5 1 90 30 0 0 1 1 0
345 1 1.1 0.9;
% 6 1 0 0 0 0 1 1 0
345 1 1.1 0.9;
% 7 1 100 35 0 0 1 1 0
345 1 1.1 0.9;
% 8 1 0 0 0 0 1 1 0
345 1 1.1 0.9;
9 1 125 100 0 0 1 1 0
345 1 1.1 0.9;
];
%% generator data
% bus Pg Qg Qmax Qmin Vg mBase status Pmax
Pmin Pc1 Pc2 Qc1min Qc1max Qc2min Qc2max ramp_agc ramp_10
ramp_30 ramp_q apf
mpc.gen = [
1 0 0 300 -300 1 100 1 250
0 0 0 0 0 0 0 0 0 0
0 0;
2 100 30 300 -300 1 100 1 300
0 0 0 0 0 0 0 0 0 0
0 0;
% 30 85 0 300 -300 1 100 1 270
10 0 200 -30 30 -15 15 0 0 0
0 0;
];
%% branch data
% fbus tbus r x b rateA rateB rateC ratio
angle status angmin angmax
mpc.branch = [
2 9 0 0.0576 0 0 550 250 0
0 1 -360 2.48;
1 2 0.017 0.092 0.158 0 550 250 0
0 0 -360 360;
% % 5 6 0.039 0.17 0.358 150 150 150 0
0 1 -360 360;
% % 30 6 0 0.0586 0 0 300 300 0
0 1 -360 360;
% % 6 7 0.0119 0.1008 0.209 40 150 150 0
0 1 -360 360;
% % 7 8 0.0085 0.072 0.149 250 250 250 0
0 1 -360 360;
% % 8 2 0 0.0625 0 250 250 250 0
0 1 -360 360;
% % 8 9 0.032 0.161 0.306 250 250 250 0
0 1 -360 360;
% % 9 4 0.01 0.085 0.176 250 250 250 0
0 1 -2 360;
];
%%----- OPF Data -----%%
%% area data
% area refbus
mpc.areas = [
1 1;
];
%% generator cost data
% 1 startup shutdown n x1 y1 ... xn
yn
% 2 startup shutdown n c(n-1) ... c0
mpc.gencost = [
1 0 0 4 0 0 100 2500 200
5500 250 7250;
2 0 0 2 24.035 -403.5 0 0 0
0 0 0;
% 1 0 0 3 0 0 200 3000 300
5000 0 0;
];
%%----- DC Line Data -----
% fbus tbus status Pf Pt Qf Qt Vf Vt
Pmin Pmax QminF QmaxF QminT QmaxT loss0 loss1
mpc.dcline = [
1 9 1 20 18.9 0 0 1.01 1
1 300 -300 300 -300 300 1 0.01;
% 1 9 1 2 1.96 0 0 1 1
2 10 0 0 0 0 0 0;
% 5 8 0 0 0 0 0 1 1
1 10 -10 10 -10 10 0 0;
% 5 9 1 10 9.5 0 0 1 0.98
0 10 -10 10 -10 10 0 0.05;
];
%% DC line cost data
% 1 startup shutdown n x1 y1 ... xn
yn
% 2 startup shutdown n c(n-1) ... c0
mpc.dclinecost = [
% 2 0 0 2 0 0 0 0 0
0 0 0 0 0;
% 2 0 0 2 0 0 0 0 0
0 0 0 0 0;
% 2 0 0 2 0 0 0 0 0
0 0 0 0 0;
2 0 0 2 7.3 0 0 0 0
0 0 0 0 0;
];
<<attachment: DCline.png>>
