Divide transfer function get "Model I/O dimensions must agree." Please help

clc;
clear;
num1 = {1, 1, 1, 1, 1, 1, 1};
num2 = {-4, -4, -4, -4, -4, -4, -4};
den1 = {[10 1], [10 1], [10 1], [10 1], [10 1], [10 1], [10 1]};
den2 = {[15 1], [15 1], [15 1], [15 1], [15 1], [15 1], [15 1]};
Gp= tf(num1,den1, 'IODelay', [3 4 5 6 7 8 9]);
Gp = minreal(Gp);
Gd= tf(num2,den2, 'IODelay', [3 4 5 6 7 8 9]);
Gd = minreal(Gd);
a=(3:1:9);
kc=(5./(0.8.*a));
Ti=10;
s = tf('s');
Gc = kc*(1/(1/10*s));
Gc=minreal(Gc);
Gc_a = (Gc*Gp)/(1+(Gc*Gp)); %y/SP

Hello all, this is my code. However when I run it every single * or / gets an error message of Model I/O dimensions must agree. Any help here?

[Paul]

I only see one error on the last line

num1 = {1, 1, 1, 1, 1, 1, 1};
num2 = {-4, -4, -4, -4, -4, -4, -4};
den1 = {[10 1], [10 1], [10 1], [10 1], [10 1], [10 1], [10 1]};
den2 = {[15 1], [15 1], [15 1], [15 1], [15 1], [15 1], [15 1]};
Gp= tf(num1,den1, 'IODelay', [3 4 5 6 7 8 9]);
Gp = minreal(Gp);
Gd= tf(num2,den2, 'IODelay', [3 4 5 6 7 8 9]);
Gd = minreal(Gd);
a=(3:1:9);
kc=(5./(0.8.*a));
Ti=10;
s = tf('s');
Gc = kc*(1/(1/10*s));
Gc=minreal(Gc);
Gc_a = (Gc*Gp)/(1+(Gc*Gp)); %y/SP

Error using * (line 80)
Model I/O dimensions must agree.

This error arises becsue Gc and Gp are both 1x7 tf matrices. Can't mutiply a 1 x 7 with another 1 x 7.

What exactly are you trying to accomplish?

[Ioannis Saravanos]

Gc_a is the response of the system. So im trying to define everything meaning control, disturbance and process to get the plot

Answer 1

[VBBV]

clc;
clear;
num1 = {1, 1, 1, 1, 1, 1, 1};
num2 = {-4, -4, -4, -4, -4, -4, -4};
den1 = {[10 1], [10 1], [10 1], [10 1], [10 1], [10 1], [10 1]};
den2 = {[15 1], [15 1], [15 1], [15 1], [15 1], [15 1], [15 1]};
Gp= tf(num1,den1, 'IODelay', [3 4 5 6 7 8 9]);
Gp = minreal(Gp);
Gd= tf(num2,den2, 'IODelay', [3 4 5 6 7 8 9]);
Gd = minreal(Gd);
a=(3:1:9);
kc=(5./(0.8.*a));
Ti=10;
s = tf('s');
Gc = kc*(1/(1/10*s));
Gc=minreal(Gc);
K = Gc.*Gp;
KK = ((1+(Gc.*Gp))); % auxiliary variable 
Gc_a = tf(Gc.*Gp,KK) %apply the  transfer function to convert coefficients from statespace 

Gc_a =
 
  From input 1 to output:
                 2.083
  exp(-3*s) * -----------
              s^2 + 0.1 s
 
  From input 2 to output:
                 1.562
  exp(-4*s) * -----------
              s^2 + 0.1 s
 
  From input 3 to output:
                 1.25
  exp(-5*s) * -----------
              s^2 + 0.1 s
 
  From input 4 to output:
                 1.042
  exp(-6*s) * -----------
              s^2 + 0.1 s
 
  From input 5 to output:
                0.8929
  exp(-7*s) * -----------
              s^2 + 0.1 s
 
  From input 6 to output:
                0.7812
  exp(-8*s) * -----------
              s^2 + 0.1 s
 
  From input 7 to output:
                0.6944
  exp(-9*s) * -----------
              s^2 + 0.1 s
 
Continuous-time transfer function.

apply the transfer function to convert coefficients from statespace

[Ioannis Saravanos]

Gc_a = tf(1,KK); Is this correct to create a transfer function with 1 in the numerator divided by KK factor in the denominator?

Answer 2

[KSSV]

May be you can run a loop for each tf..

for i = 1:7
    Gc_a(i) = (Gc(i)*Gp(i))/(1+(Gc(i)*Gp(i))); %y/SP
end

[Ioannis Saravanos]

This produces ss