facat
/
ipmthreephasese
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

1.修复了三相有电容计算错误的bug。 

2.找到一个电压还算合理的算例，不过迭代次数太多。

Signed-off-by: dmy@lab <dmy@lab.lab>
This commit is contained in:
dmy@lab 2015-04-03 20:29:29 +08:00
parent e05d1fea1e
commit 315f3509d3
10 changed files with 85 additions and 46 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
FormG.m
View File

@ -1,7 +1,8 @@
function Mat_G=FormG(PD,QD,Loadi)
function Mat_G=FormG(Volt,PD,QD,Loadi)
Mat_G=[
sparse(PD);
sparse(QD);
Volt;
];
end

9
FormLw.m
View File

@ -1,4 +1,4 @@
function Lw=FormLw(Mat_G,Init_U,Loadi,mPD3P,mQD3P)
function Lw=FormLw(Mat_G,Init_U,Loadi,mPD3P,mQD3P,mVoltABCV)
% KK=999;
% %PU=GenU(:,2);%发电机有功上界
%
@ -26,9 +26,10 @@ function Lw=FormLw(Mat_G,Init_U,Loadi,mPD3P,mQD3P)
% QDU=1.0*PD(Loadi).*sqrt(1 -PF.^2)./PF;
% PDU=1.8*ones(length(Loadi),1);
% QDU=1.8*ones(length(Loadi),1);
PDU=1.2*mPD3P;
QDU=1.2*mQD3P;
t1=([PDU',QDU'])';
PDU=1.05*mPD3P;
QDU=1.05*mQD3P;
% t1=([PDU',QDU',1*ones(length(Volt),1)'])';
t1=([PDU',QDU',1.1*mVoltABCV'])';
t2=Mat_G+Init_U'-t1;
Lw=t2;

9
FormLz.m
View File

@ -1,4 +1,4 @@
function Lz=FormLz(Mat_G,Init_L,Loadi,mPD3P,mQD3P)
function Lz=FormLz(Mat_G,Init_L,Loadi,mPD3P,mQD3P,mVoltABCV)
% KK=999;
%
% VoltL=(0.9)*ones(1,Busnum);
@ -23,9 +23,10 @@ function Lz=FormLz(Mat_G,Init_L,Loadi,mPD3P,mQD3P)
% QDL=0*PD(Loadi).*sqrt((1-PF.^2))./PF;
% PDL=0*ones(length(Loadi),1);
% QDL=0*ones(length(Loadi),1);
PDL=0.8*mPD3P;
QDL=0.8*mQD3P;
t1=([PDL',QDL'])';
PDL=0.95*mPD3P;
QDL=0.95*mQD3P;
% t1=([PDL',QDL',0.88*ones(length(Volt),1)'])';
t1=([PDL',QDL',0.9*mVoltABCV'])';
t2=Mat_G-Init_L'-t1;
Lz=t2;

42
OPF.m
View File

@ -41,7 +41,7 @@ PD=Pabc/3;
QD=Qabc/3;
Loadi=find(PD~=0);
maxD=100000;% 最大不平衡量
EPS=1e-5;
EPS=1e-6;
k=0;
kmax=20;
fsY11=fsY1;
@ -68,12 +68,12 @@ tic
VoltpA=sparse(ones(busNum,1));
VoltpB=sparse(ones(busNum,1)).*exp(1j*-120/180*pi);
VoltpC=sparse(ones(busNum,1)).*exp(1j*+120/180*pi);
while(k<=kmax && maxD> EPS)
while(k<=kmax+10 && maxD> EPS)
k=k+1;
%把补偿电容看作负荷
SA=VoltpA.*conj(VoltpA.*sparse(cap.capNode,1,1j*cap.capB(:,1),busNum,1));
SB=VoltpB.*conj(VoltpA.*sparse(cap.capNode,1,1j*cap.capB(:,2),busNum,1));
SC=VoltpC.*conj(VoltpA.*sparse(cap.capNode,1,1j*cap.capB(:,3),busNum,1));
SB=VoltpB.*conj(VoltpB.*sparse(cap.capNode,1,1j*cap.capB(:,2),busNum,1));
SC=VoltpC.*conj(VoltpC.*sparse(cap.capNode,1,1j*cap.capB(:,3),busNum,1));
%先不要电容
% SA=0;
% SB=0;
@ -180,6 +180,7 @@ Balance3P(3:3:end)=(Balance-1)*3+3;
% QGC=sum(QD3P(3:3:end))*1.03;
% PG3P=sparse( (Balance-1)*3+1:(Balance-1)*3+3,1,[PGA,PGB,PGC],Busnum*3,1);
% QG3P=sparse( (Balance-1)*3+1:(Balance-1)*3+3,1,[QGA,QGB,QGC],Busnum*3,1);
PG3P=real(PGQG);
QG3P=imag(PGQG);
@ -210,7 +211,7 @@ rQD3P(3:3:end)=phaseCSpotLoadQ;
rPD3P=rPD3P(Loadi);
rQD3P=rQD3P(Loadi);
%量测量
sigma=0.01;
sigma=0.03;
VoltSigma=(1+normrnd(0,sigma,length(rVoltABCV),1));
mVoltABCV=rVoltABCV.*VoltSigma;
PD3PSigma=(1+normrnd(0,sigma,length(rPD3P),1));
@ -221,11 +222,28 @@ mQD3P=rQD3P.*QD3PSigma;
wVolt=1./(abs(mVoltABCV*sigma).^2);
wPD=1./(abs(mPD3P*sigma).^2);
wQD=1./(abs(mQD3P*sigma).^2);
%% 没有量测量
noLoadi=[1,5,6,10,11];
noLoadi=Loadi;
% noLoadi=[1,11];
% noLoadi=[];
noPQi3P=zeros(length(noLoadi)*3,1);
noPQi3P(1:3:end)=(noLoadi-1)*3+1;
noPQi3P(2:3:end)=(noLoadi-1)*3+2;
noPQi3P(3:3:end)=(noLoadi-1)*3+3;
% noLoadi=noPQi3P;
wVolt( ismember( Loadi,noLoadi) )=0;
wPD(ismember( Loadi,noLoadi))=0;
wQD(ismember( Loadi,noLoadi))=0;
% wVolt(5)=0;
% wPD(5)=0;
% wQD(5)=0;
%%
RestraintCount=size(Loadi,1)*2; %Ô¼ÊøÌõ¼þÊý
RestraintCount=size(Loadi,1)*2+length(rVoltABCV); %约束条件数
% RestraintCount=size(Loadi,1)*2; %约束条件数
Init_Z=sparse(ones(1,RestraintCount));
Init_W=sparse(-1*ones(1,RestraintCount));
Init_L=1*sparse(ones(1,RestraintCount));
@ -234,12 +252,12 @@ Init_Y=sparse(1,2*Busnum*3);%
Gap=(Init_L*Init_Z'-Init_U*Init_W');
KK=0;
ContrlCount=size(Loadi,1)*2+Busnum*6;
kmax=20;
kmax=2000;
%%
Precision=1e-5;
CenterA=0.1;
%% 加误差
Volt=Vp3m;
Volt=1*Vp3m;
UAngel=Vp3a;
while(abs(Gap)>Precision)
if KK>kmax
@ -256,7 +274,7 @@ while(abs(Gap)>Precision)
% L_1Z=diag(Init_Z./Init_L);
% U_1W=diag(Init_W./Init_U);
%% 形成海森阵
deltdeltF=func_deltdeltF(wVolt,wPD,wQD,ContrlCount);
deltdeltF=func_deltdeltF(wVolt,wPD,wQD,ContrlCount,Loadi);
% deltdeltF=0;
%% 形成ddHy
ddh=func_ddh(Volt,Init_Y,Busnum,Y,UAngel,r,c,Yangle,Loadi,ContrlCount);
@ -269,11 +287,11 @@ while(abs(Gap)>Precision)
%% 形成方程矩阵
Luu=Init_U'.*Init_W'+Init_u*ones(RestraintCount,1);
Lul=Init_L'.*Init_Z'-Init_u*ones(RestraintCount,1);
Mat_G=FormG(PD3P,QD3P,Loadi);
Mat_G=FormG(Volt,PD3P,QD3P,Loadi);
Mat_H=FormH(Busnum,Volt,PG3P,PD3P,QG3P,QD3P,Y,UAngel,r,c,Yangle,Loadi);
Ly=Mat_H;
Lz=FormLz(Mat_G,Init_L,Loadi,mPD3P,mQD3P);
Lw=FormLw(Mat_G,Init_U,Loadi,mPD3P,mQD3P);
Lz=FormLz(Mat_G,Init_L,Loadi,rPD3P,rQD3P,rVoltABCV);
Lw=FormLw(Mat_G,Init_U,Loadi,rPD3P,rQD3P,rVoltABCV);
Lx=FormLx(deltF,deltH,Init_Y,deltG,Init_Z,Init_W);
% YY=FormYY(Lul,Lz,Ly,Luu,Lw,Lx);
%% 开始解方程

2
checkSSatisfied.m
View File

@ -17,7 +17,7 @@ pLoadABC(2:3:end)=phaseBSpotLoadP+1j*phaseBSpotLoadQ;
pLoadABC(3:3:end)=phaseCSpotLoadP+1j*phaseCSpotLoadQ;
ck=Sp3+pLoadABC;
ck(3*(Balance-1)+1:3*(Balance-1)+3)=0;
if any(abs(ck)>1e-5)
if any(real(ck)>1e-5) ||any(imag(ck)>1e-5)
fprintf('反推回的功率不匹配。潮流方程约束不满足。\n');
end
output_args=ck;

6
dataRead.m
View File

@ -46,7 +46,7 @@ phaseCSpotLoadQ( ismember(setIJ,spotloads(:,1) ) )=spotloads(:,7);
%% 补偿电容
cap=data(zeroEntries(2)+1:zeroEntries(3)-1,:);
if ~isempty(cap)
capNode=nodeNum(cap(:,1)==setIJ);
capNode=nodeNum(ismember(setIJ,cap(:,1)));
%数据文件中的补偿电容给的是补偿容量，要将其转换为导纳。
capB=cap(:,2:4)/1000;
cap=struct();
@ -55,9 +55,9 @@ cap.capB=capB;
%把补偿电容加到矩阵中。
offSet=(capNode-1)*3+1;
phaseABCY=phaseABCY+sparse(offSet,offSet,1j*capB(:,1),busNum*3,busNum*3);
offSet=(capNode-1)*3+1;
offSet=(capNode-1)*3+2;
phaseABCY=phaseABCY+sparse(offSet,offSet,1j*capB(:,2),busNum*3,busNum*3);
offSet=(capNode-1)*3+1;
offSet=(capNode-1)*3+3;
phaseABCY=phaseABCY+sparse(offSet,offSet,1j*capB(:,3),busNum*3,busNum*3);
else
cap=struct();

26
feeder13/data1.txt
View File

@ -13,14 +13,18 @@
602 684 611 300
601 692 675 500
0
632 0 0 0
684 200 200 200
671 200 320 320
652 150 300 200
0
634 160 110 120 90 120 90
645 123 102 170 125 133 110
646 172 150 230 132 211 160
652 128 86 30 10 80 70
675 85 10 68 60 20 12
611 52 42 85 60 70 40
634 120 90 120 90 160 110
645 40 30 50 40 60 50
646 100 50 70 40 60 40
652 20 10 60 30 40 40
671 385 220 385 220 385 220
675 68 60 290 212 285 190
692 60 50 40 40 70 40
611 80 20 60 40 30 20
0
600 632 645 500
602 632 633 500
@ -46,5 +50,11 @@
692 142 140 153 121 170 151
611 156 142 185 60 170 80
0
634 160 110 120 90 120 90
645 123 102 170 125 133 110
646 17 15 23 13 21 16
652 128 86 30 10 80 70
675 85 10 68 60 20 12
611 52 42 85 60 70 40
680 12 4 10 6 7 4

23
func_deltG.m
View File

@ -8,26 +8,25 @@ sizeLoadi=size(Loadi,1);
%%
dg3_dPD=sparse(1:size(Loadi,1),1:size(Loadi,1),ones(size(Loadi,1),1),size(Loadi,1),size(Loadi,1));
dg4_dPD=sparse(size(Loadi,1),length(Loadi));
% dg5_dPD=sparse(size(Loadi,1),Busnum*3);
dg5_dPD=sparse(size(Loadi,1),Busnum*3);
%%
dg3_dQD=sparse(length(Loadi),length(Loadi));
dg4_dQD=sparse(1:size(Loadi,1),1:size(Loadi,1),ones(size(Loadi,1),1),size(Loadi,1),size(Loadi,1));
% dg5_dQD=sparse(size(Loadi,1),Busnum*3);
dg5_dQD=sparse(size(Loadi,1),Busnum*3);
%%
dg3_dx=sparse(2*Busnum*3,sizeLoadi);
dg4_dx=sparse(2*Busnum*3,length(Loadi));
% dg5_dx=[sparse(1:Busnum*3,1:Busnum*3,zeros(Busnum*3,1),Busnum*3,Busnum*3);
% sparse(Busnum*3,Busnum*3);
% ];
dg5_dx=[sparse(1:Busnum*3,1:Busnum*3,ones(Busnum*3,1),Busnum*3,Busnum*3);
sparse(Busnum*3,Busnum*3);
];
%%
% deltG=[dg3_dPD,dg4_dPD,dg5_dPD;
% dg3_dQD,dg4_dQD,dg5_dQD;
% dg3_dx,dg4_dx,dg5_dx;
% ];
deltG=[dg3_dPD,dg4_dPD;
dg3_dQD,dg4_dQD;
dg3_dx,dg4_dx;
deltG=[dg3_dPD,dg4_dPD,dg5_dPD;
dg3_dQD,dg4_dQD,dg5_dQD;
dg3_dx,dg4_dx,dg5_dx;
];
% deltG=[dg3_dPD,dg4_dPD;
% dg3_dQD,dg4_dQD;
% dg3_dx,dg4_dx;];
end

2
func_deltdeltF.m
View File

@ -1,4 +1,4 @@
function deltdeltF=func_deltdeltF(wVolt,wPD,wQD,ContrlCount)
function deltdeltF=func_deltdeltF(wVolt,wPD,wQD,ContrlCount,Loadi)
%ContrlCount=size(PVi,1)+size(PGi,1)+size(wD,1)+Busnum*2; %P,Q,Volt theta这些控制变量数
C=[wPD' wQD'];

9
test.m
View File

@ -6,6 +6,9 @@ AME_VAngleSum=0;
JMeasurementSum=0;
AME_PDSum=0;
AME_QDSum=0;
AME_mVoltSum=0;
AME_mPDSum=0;
AME_mQDSum=0;
N=1000;
for I=1:N
[JMeasurement,AME_Volt,AME_VAngle,AME_PD,AME_QD,AME_mVolt,AME_mPD,AME_mQD,isConverged]=OPF();
@ -18,9 +21,15 @@ for I=1:N
AME_VAngleSum=AME_VAngleSum+AME_VAngle;
AME_PDSum=AME_PDSum+AME_PD;
AME_QDSum=AME_QDSum+AME_QD;
AME_mVoltSum=AME_mVoltSum+AME_mVolt;
AME_mPDSum=AME_mPDSum+AME_mPD;
AME_mQDSum=AME_mQDSum+AME_mQD;
end
JMeasurementSum=JMeasurementSum/N;
AME_VoltSum=AME_VoltSum/N;
AME_VAngleSum=AME_VAngleSum/N;
AME_PDSum=AME_PDSum/N;
AME_QDSum=AME_QDSum/N;
AME_mVoltSum=AME_mVoltSum/N;
AME_mPDSum=AME_mPDSum/N;
AME_mQDSum=AME_mQDSum/N;