Adding smaller matrices into a larger matrix using a loop

I asked this question before too, and I used the solutions given, however I still got several errors so I'm going to give this another shot since my code has too many errors

V9Split = mat2cell(V9,repmat(5,20,1),5)  
V9_eff = zeros(20,5);        
for i=1:20
V9_eff(i) = V9Split{i}+(((P9-Po).*A9(i))./mdot_core)   
end

However the coding did not work. I want the code to produce something like:

V9_eff1 = V9Split{1}+(((P9-Po).*A9(1))./mdot_core)
V9_eff2 = V9Split{2}+(((P9-Po).*A92)./mdot_core);
V9_eff3 = V9Split{3}+(((P9-Po).*A93)./mdot_core);
V9_eff4 = V9Split{4}+(((P9-Po).*A94)./mdot_core);
V9_eff5 = V9Split{5}+(((P9-Po).*A95)./mdot_core);
V9_eff6 = V9Split{6}+(((P9-Po).*A96)./mdot_core);
V9_eff7 = V9Split{7}+(((P9-Po).*A97)./mdot_core);
V9_eff8 = V9Split{8}+(((P9-Po).*A98)./mdot_core);
V9_eff9 = V9Split{9}+(((P9-Po).*A99)./mdot_core);
V9_eff10 = V9Split{10}+(((P9-Po).*A910)./mdot_core);
V9_eff11 = V9Split{11}+(((P9-Po).*A911)./mdot_core);
V9_eff12 = V9Split{12}+(((P9-Po).*A912)./mdot_core);
V9_eff13 = V9Split{13}+(((P9-Po).*A913)./mdot_core);
V9_eff14 = V9Split{14}+(((P9-Po).*A914)./mdot_core);
V9_eff15 = V9Split{15}+(((P9-Po).*A915)./mdot_core);
V9_eff16 = V9Split{16}+(((P9-Po).*A916)./mdot_core);
V9_eff17 = V9Split{17}+(((P9-Po).*A917)./mdot_core);
V9_eff18 = V9Split{18}+(((P9-Po).*A918)./mdot_core);
V9_eff19 = V9Split{19}+(((P9-Po).*A919)./mdot_core);
V9_eff20 = V9Split{20}+(((P9-Po).*A920)./mdot_core);
V9_eff = [V9_eff1; V9_eff2; V9_eff3; V9_eff4; V9_eff5; V9_eff6; V9_eff7; V9_eff8; V9_eff9; V9_eff10; V9_eff11; V9_eff12; V9_eff13; V9_eff14; V9_eff15; V9_eff16; V9_eff17; V9_eff18; V9_eff19; V9_eff20]

The naming isn't a problem because I could just recall using mat2cell I just want the code to add as shown with as few lines as possible Thanks again for help

[Stephen23]

What exactly is V9? Is it a numeric array, or a cell array?

Answer 1

[Stephen23]

Try this:

V9Split = mat2cell(V9,repmat(5,20,1),5)  
V9_eff = zeros(20,5);        
for k = 1:20
    V9_eff(k,:) = V9Split{k}+(((P9-Po).*A9(k))./mdot_core)   
end

I made two changes:

• replaced V9_eff(k) with V9_eff(k,:), because you cannot allocate a vector of values into one scalar location of V9_eff. The syntax |V9_eff(k) is only one matrix element, yet you are trying to put multiple elements into this location! This is always going to be an error. The syntax V9_eff(k,:) instead fills the entire row with all of the values from the calculation.
• replaced the loop variable i with k, because i and j are the inbuilt imaginary unit and should not be used for variable names.

[Sundus Awan]

I tried it with the changes you made but it still gives me an error

Subscripted assignment dimension mismatch.

I tried editing the code but it still gave me the same error. The size of all the variables:

V9Split{#} - 20 matrices of size 5x5
V9_eff - 100x5
A9(#) - 20 matrices of size 5x5
mdot_core - 5x5

I don't know if this makes it easier to code, but thank you for your help so far!

[Guillaume]

Is A9 a cell array then, if it contains 20 matrices?

Answer 2

[Guillaume]

Assuming A9 is a cell array (with the same number of cells as V9split):

V9Split = mat2cell(V9,repmat(5,20,1),5)
V9_eff = cellfun(@(V9, A9) V9+((P9-Po).*A9)./mdot_core, V9Split, A9, 'UniformOutput', false);
V9_eff = vertcat (V9_eff{:})