Solve with several vectors

Hi, I have these vectors:

int_v_d
int_i_d
i_d
int_omega_i_q
int_v_q
int_i_q
i_q
int_omega_i_d
theta

All of them have the same size and each of them is a column vector.

Now, I have a set of two equations like this one:

eq_1 = int_v_d == -R_AFPM * int_i_d - (i_d - i_d(1)) .* L_d + L_q .* int_omega_i_q;
eq_2 = int_v_q == -R_AFPM * int_i_q - (i_q - i_q(1)) .* L_q - L_d .* int_omega_i_d + ...
    psi_f * theta;

The variables that are not vectors, and that I didn't mention before, are constants. I want to calculate the values L_d and L_q for each row.

I am using solve and subs functions for this, but I get this error:

Error using symengine
Unable to convert expression containing symbolic variables into double array. Apply 'subs' function
first to substitute values for variables.

Error in sym/double (line 709)
Xstr = mupadmex('symobj::double', S.s, 0);

Here is my complete code:

syms int_v_d int_i_d i_d int_omega_i_q int_v_q int_i_q i_q int_omega_i_d theta L_d L_q
eq_1 = int_v_d == -R_AFPM * int_i_d - (i_d - i_d(1)) .* L_d + L_q .* int_omega_i_q;
eq_2 = int_v_q == -R_AFPM * int_i_q - (i_q - i_q(1)) .* L_q - L_d .* int_omega_i_d + ...
    psi_f * theta;
sol = solve([eq_1;eq_2],[L_d;L_q]);
vec_L_d = double(subs(sol.L_d,{int_v_d;int_i_d;i_d;int_omega_i_q; ...
    int_v_q;int_i_q;i_q;int_omega_i_d;theta},{int_v_d;int_i_d;i_d;int_omega_i_q; ...
    int_v_q;int_i_q;i_q;int_omega_i_d;theta}));
vec_L_q = double(subs(sol.L_q,{int_v_d;int_i_d;i_d;int_omega_i_q; ...
    int_v_q;int_i_q;i_q;int_omega_i_d;theta},{int_v_d;int_i_d;i_d;int_omega_i_q; ...
    int_v_q;int_i_q;i_q;int_omega_i_d;theta}));

The result for L_d and L_q should be for each one a vector of the same size as the other vectors.

Thank you in advance.

[Walter Roberson]

syms int_i_d

does not declare a vector. You need a size after it

eq_1 = int_v_d == -R_AFPM * int_i_d - (i_d - i_d(1)) .* L_d + L_q .* int_omega_i_q;
eq_2 = int_v_q == -R_AFPM * int_i_q - (i_q - i_q(1)) .* L_q - L_d .* int_omega_i_d + ...
  psi_f * theta;

"The variables that are not vectors, and that I didn't mention before, are constants."

So L_q and L_d are constants, since you did not mention them before? Scalar constants?

"The result for L_d and L_q should be for each one a vector of the same size as the other vectors."

That would require that they are symbolic vectors since you are solving for them.

[Luciano Montanelli]

L_d and L_q should be vectors, since I should solve this set of equations with the vectors I specified at the beginning. R_AFPM and phi_f are the only constants (scalar constants).

I could solve this by calling every vector in syms with a different name, the same with every vector in the first {} in the subs arguments.

Thank you for your help, @Walter Roberson!

Answer 1

[Torsten]

syms int_v_d_sym int_i_d_sym i_d_sym i_d1_sym int_omega_i_q_sym int_v_q_sym int_i_q_sym i_q_sym i_q1_sym int_omega_i_d_sym theta_sym L_d_sym L_q_sym R_AFPM_sym psi_f_sym
eq_1 = int_v_d_sym == -R_AFPM_sym * int_i_d_sym - (i_d_sym - i_d1_sym) .* L_d_sym + L_q_sym .* int_omega_i_q_sym;
eq_2 = int_v_q_sym == -R_AFPM_sym * int_i_q_sym - (i_q_sym - i_q1_sym) .* L_q_sym - L_d_sym .* int_omega_i_d_sym + ...
    psi_f_sym * theta_sym;
sol = solve([eq_1;eq_2],[L_d_sym;L_q_sym]);
sol.L_d_sym

ans = 

sol.L_q_sym

ans = 

for i = 1:numel(int_v_d)
  L_d(i) = double(subs(L_d_sym,[int_v_d_sym int_i_d_sym i_d_sym i_d1_sym int_omega_i_q_sym int_v_q_sym int_i_q_sym i_q_sym i_q1_sym int_omega_i_d_sym theta_sym R_AFPM_sym psi_f_sym],[int_v_d(i) int_i_d(i) i_d(i) i_d(1) int_omega_i_q(i) int_v_q(i) int_i_q(i) i_q(i) i_q(1) int_omega_i_d(i) theta(i) R_AFPM psi_f]))
  L_q(i) = double(subs(L_q_sym,[int_v_d_sym int_i_d_sym i_d_sym i_d1_sym int_omega_i_q_sym int_v_q_sym int_i_q_sym i_q_sym i_q1_sym int_omega_i_d_sym theta_sym R_AFPM_sym psi_f_sym],[int_v_d(i) int_i_d(i) i_d(i) i_d(1) int_omega_i_q(i) int_v_q(i) int_i_q(i) i_q(i) i_q(1) int_omega_i_d(i) theta(i) R_AFPM psi_f]))
end

[Luciano Montanelli]

Thank you a lot!