operator +: nonconformant arguments (op1 is 10x10, op2 is 10x9)

[Beginner1]

Hi,

There is an error in how Octave first simplifies and then multiplies
matrixes. 
Here is my code:
/*pkg load control 
pkg load io  
pkg load signal 

%eig_A=eig(sys.A);
%plot(real(eig_A),imag(eig_A),"rx")
%grid on



%% se define la variable para hacer el barrido
%cycle_vector = logspace(log10(0.5),log10(2),100);
cycle_vector = linspace(0,15,15);
color_plot = colormap(jet(length(cycle_vector))); %los colores van variando
del azul al rojo


%all = axes('position',[.1,.1,.8,.8])

%% cycle (secalcula las nuevas matrices para cada valor del barrido)
for n_cycle = 1:length(cycle_vector)
  %hay que recalcular la matrz A para el barridod e la variable puede que
haya que Volver a linealizar
  kv=cycle_vector(n_cycle);
  %kv=15;
  Tv=0.05;%1;%Constante de tiempo del PI de la componente d 
  kq=0.4;%Constante proporcional del PI de la componente q 
  Tq=0.02;%Constante de tiempo del PI de la componente q 
  kdroop_AC=0;%-3;%Pendiente droop Q/U_AC 
  droop_FSM=0.05;%Pendiente droop P/f 
  FSM_slope=droop_FSM/0.002;%Pendiente droop P/f en p.u.
  
  %Inner Loop
  ki=5000;%1% Constante integral de los controladores proporcional
resonantes 
  kp=1.5;% Constante proporcional de los controladores proporcional
resonantes 
  w=2*pi*50;
  wg0=2*pi*50.01;
  w_r=2*pi*50;
  wr=2*pi*50;
  a23 = 2*(wg0^2+wr^2);
  da23 = 4*wg0;
  a03 = (wr^2-wg0^2)^2;
  da03 = -4*wg0*(wr^2-wg0^2);
  b33 = ki;
  db33 = 0;
  b13 = ki*(wg0^2+wr^2);
  db13 = ki*2*wg0;
  c23 = -ki*wg0;
  dc23 = -ki;
  c03 = ki*wg0*(wr^2-wg0^2);
  dc03 = ki*(wr^2-3*wg0^2);
  
  %2-level VSC&Normalization block
  Vdref_perm=1.002;%1.003;
  Vqref_perm=0.085;%0.077;
  Vdc_diff_perm=0.5;%0.492;
  kdiff=(0.7)*Vdc_diff_perm;%Según DIgilent debería ser multiplicada por 1.4
en vez de 0.7 
  K_0=sqrt(3)/(2*sqrt(2));%Factor de modulación sinusoidal del VSC
  Ub_DC=640/2;%Tensión base en DC del VSC-Según DIgilent debería ser 640 en
vez de 640/2 
  Ub_AC=275;%Tensión base en AC del VSC
  k0=(K_0)*Ub_DC/Ub_AC;
  Vdc_perm=1.001;%1.017;
  Pmd_perm=0.702;%0.702;%0.69;
  Pmq_perm=0;%0.053;
  Kvsc=k0*Vdc_perm;
  
  x1d_0= 0;
  x1q_0= 0;
  x2d_0=0;
  x2q_0=0;
  x3d_0=0;
  x3q_0=0;
  x4d_0=0;
  x4q_0=0;
  
%Función enable: para marcar la activación de la rama FSM
  enable=0;
  
%Definición de matrices A,B,C y D
  
  %Outer loop:Vdc-Q
  
  A_ol=[0 0;0 0];
  B_ol=[-kv/Tv kv/Tv 0 0 0 0 -kv*FSM_slope*enable/Tv kv*enable/Tv
-kv*enable/Tv kv*enable/Tv;0 0 kq/Tq -kq/Tq -(kq/Tq)*kdroop_AC
(kq/Tq)*kdroop_AC 0 0 0 0];
  C_ol=[1 0;0 1];
  D_ol=[-kv kv 0 0 0 0 -kv*FSM_slope*enable kv*enable -kv*enable kv*enable;0
0 kq -kq -kq*kdroop_AC kq*kdroop_AC 0 0 0 0];
  

  %Inner loop:PR current controller

  A_il=[0 1 0 0 0 0 0 0; 0 0 1 0 0 0 0 0;0 0 0 1 0 0 0 0; -(w^2-wg0^2) 0
-4*w^2 0 0 0 0 0;0 0 0 0 0 1 0 0;0 0 0 0 0 0 1 0;0 0 0 0 0 0 0 1;0 0 0 0
-(w^2-wg0^2) 0 -4*w^2 0];
  B_il=[0 0 0 0 0 0 0;0 0 0 0 0 0 0;0 0 0 0 0 0 0;0 0 -1 0 1 0 -4*w*x3d_0;0
0 0 0 0 0 0;0 0 0 0 0 0 0;0 0 0 0 0 0 0;0 0 0 -1 0 1 -4*w*x3q_0];
  C_il=[0 2*ki*w^2 0 ki 0 0 ki*w 0; 0 0 -ki*w 0 0 2*ki*w^2 0 ki];
  D_il=[1 0 -kp 0 kp 0 ki*(2*w*x2d_0+x3q_0+2*w^2*x1q_0); 0 1 0 -kp 0 kp
ki*(2*w*x2q_0-2*w^2*x1d_0-x3d_0)];

  %Normalization block 
  D_nb=[kdiff 0 (0.7)*Vdref_perm;0 kdiff (0.7)*Vqref_perm];%Según DIgilent
debería ser multiplicada por 1.4 en vez de 0.7 

  %2-level VSC:controlled voltage source-average model
  D_vsc=[k0*Vdc_perm 0 k0*Pmd_perm;0 k0*Vdc_perm k0*Pmq_perm];
  
  %Outer loop
  stname_ol = {'xv','xq'};
 
inname_ol={'vdcref','vdc','qacref','qac','vacref','vac','fmeas','pdcref','paux','FSMcoordpos'};
  outname_ol={'idref','iqref'};
  sys_ol = ss
(A_ol,B_ol,C_ol,D_ol,'stname',stname_ol,'inname',inname_ol,'outname',outname_ol);
  
  %Inner loop
  stname_il = {'xd1','xd2','xd3','xd4','xq1','xq2','xq3','xq4'};
  inname_il={'Vd','Vq','Id','Iq','idref','iqref','omegagiro'};
  outname_il={'Vdref','Vqref'};
  sys_il = ss
(A_il,B_il,C_il,D_il,'stname',stname_il,'inname',inname_il,'outname',outname_il);
  
  %Normalization block 
  inname_nb={'Vdref','Vqref','Vdcdiff'};
  %inname_nb={'Vdref','Vqref'};
  outname_nb={'Pmd','Pmq'};
  sys_nb = ss (D_nb,'inname',inname_nb,'outname',outname_nb);
  
  %2-level VSC
  inname_vsc={'Pmd','Pmq','vdc'};
  outname_vsc={'Vd_AC','Vq_AC'};
  sys_vsc = ss (D_vsc,'inname',inname_vsc,'outname',outname_vsc);
  
  %Assembled system1
 
inname_group={'vdcref','vdc','qacref','qac','vacref','vac','fmeas','pdcref','paux','FSMcoordpos','Vd','Vq','Id','Iq','omegagiro','Vdcdiff'};
 
%inname_group={'vdcref','vdc','qacref','qac','vacref','vac','fmeas','pdcref','paux','FSMcoordpos','Vd','Vq','Id','Iq'};
 
outname_group={'idref','iqref','Vdref','Vqref','Pmd','Pmq','Vd_AC','Vq_AC'};
  sys=connect(sys_ol,sys_il,sys_nb,sys_vsc,inname_group,outname_group);
  B=sys.B;
  K_cl=ones(16,9);
  Aol=sys.A;
  Acl=Aol+B*K_cl;
    
    end*/

You can copy and paste this code and see what hapens afterwads. The error I
get is
error: rrr: operator +: nonconformant arguments (op1 is 10x10, op2 is 10x9)
error: called from
    rrr at line 131 column 6

But what is important in this message is the last part of the code, that
refers to the sum of matrixes:
/*  B=sys.B;
  K_cl=ones(16,9);
  Aol=sys.A;
  Acl=Aol+B*K_cl;*/

The dimension problem should be solved by making K_cl equal to ones(16,10).
But when I do this, and run the code, the console reports again a new
dimension problem:
error: rrr: operator +: nonconformant arguments (op1 is 9x9, op2 is 9x10)
error: called from
    rrr at line 131 column 6

So Octave must be simplifying matrixes since they are plenty of zeros and
this does not allow to perform the sum of them as they cannot coincide in
dimension.

Any suggestion to avoid the simplification of matrixes in Octave?





--
Sent from: http://octave.1599824.n4.nabble.com/Octave-General-f1599825.html