How to solve transfer function inside summation?

I tried this method:

kr=600;
kp=9;
s=tf('s');
Gc_r= (0.866*s - n*50*pi)/((s^2) + 6*s + (n*100*pi)^2 );
sol= subs(Gc_r, n, 1:2:9);
f=sum(sol)
r_term=f*kr
Gc=kp+r_term

But, it is giving this error : Incorrect dimensions for raising a matrix to a power. Check that the matrix is square and the power is a scalar. To

perform elementwise matrix powers, use '.^'.

I corrected above mentioned error but it didn't work.

error : Undefined operator '.^' for input arguments of type 'tf'.

__________________________________________________________

I also tried this method:

kr=600;
kp=9;
syms s n 
Gc_r= (0.866*s - n*50*pi)/((s^2) + 6*s + (n*100*pi)^2 );
sol= subs(Gc_r, n, 1:2:9);
f=sum(sol)
r_term=f*kr
Gc=kp+r_term

It didn't simplified answer either. returned in the form of sum of products.

Answer 1

[Walter Roberson]

Symbolic variables do not mix with tf. You cannot use subs() with a tf

You need to use your second approach. Then

[N, D] = numdem(Gc);
collect(N,s) / D

or just

collect(Gc, s)

which will collect the denominator too when the first approach leaves it factored.

[Walter Roberson]

If you want to use tf then use a loop or arrayfun to build the total instead of putting in symbolic n, which is incompatible with tf

[Uzair Amin]

thankyou @walter ... it worked!

can you please develop code using tf ? it would be helpful for me in my future works.

[Walter Roberson]

kr=600;
kp=9;
s=tf('s');
n = 1;
f = (0.866*s - n*50*pi)/((s^2) + 6*s + (n*100*pi).^2 );
for n = 3:2:9
    f = f + (0.866*s - n*50*pi)/((s^2) + 6*s + (n*100*pi).^2 );
end

r_term = f*kr;
Gc = kp+r_term;
Gc

Gc =
 
  9 s^10 + 2868 s^9 + 1.443e08 s^8 + 3.731e10 s^7 + 7.432e14 s^6 + 1.467e17 s^5 + 1.408e21 s^4 + 1.895e23 s^3 + 8.176e26 s^2 + 5.702e28 s + 6.099e31
  --------------------------------------------------------------------------------------------------------------------------------------------------
    s^10 + 30 s^9 + 1.629e07 s^8 + 3.908e08 s^7 + 8.551e13 s^6 + 1.539e15 s^5 + 1.661e20 s^4 + 1.994e21 s^3 + 1.003e26 s^2 + 6.019e26 s + 8.363e30
 
Continuous-time transfer function.

[Paul]

Though it may not matter for this particular example, you should consider using ss objects instead of tf objects for this type of computation.