confusion regarding Matrix dimension.

Hello,

I have a B matrix in my model having a size of 5x5. I want to multiply B matrix with S matrix which is 5x6. So I am assuming an additional raw in matrix B as [ 0 0 0 0 0]. This way my Matrix B will be 6x5 and I am doing S*B.

Is this the right method? Am I doing right or wrong? Can I assume a null raw? Please help me.

[Ced]

EDIT: Sorry, initially misread your question. To do S*B, the number of columns in S must be the same as the number of rows in B. But this is just basic matrix multiplication, and not really matlab-related. I need more information about your calculation to know whether a zero row makes sense or not.

[Bhavik]

Thank you Ced, I really appreciate your interest.

Smatrix = [ S11 S12 0 0 0; S21 S22 0 0 0; S31 -S31 0 0 0; 0 0 1 0 0; 0 0 0 1 0; 0 0 0 0 1];

Bmatrix = [B11 B12 0 0 0; B21 B22 0 0 0; 0 0 1 0 0; 0 0 0 1 0; 0 0 0 0 1];

---------------------- Mbar = (Smatrix.'*Mmatrix)*Smatrix;

Vbar = Smatrix.'*(Vvector + Mmatrix*Sdotmatrix*v);

Gbar = Smatrix.'*Ggravity;

Bbar = Smatrix.'*Bmatrix;

I am finding vdot by the following equation:

vdot = (Mbar)/(Bbar*u - Vbar - Gbar);

Where, Mbar is 6x6, Vbar and G bar are 5x1 and Bbar should be 5x1. I have B matrix 5x5 to get Bbar 5x1, B matrix must be 6x5. But I have data in which B matrix is 5x5. So I assumed B matrix as 6x5 (Taking the last raw as [0 0 0 0 0]). Is this the right way I am doing?

[Ced]

I'm sorry, I have no idea what the actual question is. But if Smatrix and Bmatrix are 6x5, then Smatrix'.*Bmatrix is not going to work, no matter if Bmatrix is 5x5 or 6x5, since Smatrix' and Bmatrix need to have the same number of elements.

I think Stephen Cobeldick is correct: You need to have a look at the difference between matrix multiplication S*B and element-wise multiplication S.*B.

Answer 1

[Stephen23]

You might want to read about the difference between matrix and array multiplication.

Try this:

S.*B