How to multiply different size arrays and multiply each element by each element?

I have an equation with 4 arrays in, all of a different size. I want every combination of the arrays possible and multiply all by all because i'm going to filter out the outputs later. It's almost like a simplex algorithm but in one equation with the limits set by the peson inputting the data. Thanks for all your help!

Alex

Vh = ((Span * ((1 - Ct)+1)*MAC/2) * CoM)/(WingArea*WChord);

% Where, Span, Ct, MAC, and CoM are all differently sized arrays.

Answer 1

[Alex]

I did some searching and installed deep learning theory:

What does this do and would it work for the above?

% your arrays
A = rand(1,4);
B = rand(1,6);
C = rand(1,8);
D = rand(1,2);
% get 4xN matrix of all combinations of A,B,C,D
inputs = combvec(A,B,C,D)';
% get a 1XN vector of results
res = inputs(:,1).*inputs(:,2).*inputs(:,3).*inputs(:,4);

I believe this worked! I got loads of results so it looks like it did!

[Torsten]

I don't know if this is necessary for your application, but it might cost some effort to deduce which indices of A, B, C and D led to certain values of "res".

Answer 2

[Steven Lord]

I recommend using the combinations function to generate the combinations of values, then using variables from the table returned by combinations in your calculations.

Span = [1 2];
Ct = [3 4 5];
MAC = [6 7 8 9];
C = combinations(Span, Ct, MAC)

C = 24×3 table
    Span    Ct    MAC
    ____    __    ___

     1      3      6 
     1      3      7 
     1      3      8 
     1      3      9 
     1      4      6 
     1      4      7 
     1      4      8 
     1      4      9 
     1      5      6 
     1      5      7 
     1      5      8 
     1      5      9 
     2      3      6 
     2      3      7 
     2      3      8 
     2      3      9 

computation = C.Span.^2 + C.Ct.^3 - C.MAC.^4

computation = 24×1

-1268 -2373 -4068 -6533 -1231 -2336 -4031 -6496 -1170 -2275 -3970 -6435 -1265 -2370 -4065

<mw-icon class=""></mw-icon>

<mw-icon class=""></mw-icon>

We can spot check:

values = C{17, :} % Row 17; Span is 2, Ct is 4, MAC is 6

values = 1×3

2 4 6

<mw-icon class=""></mw-icon>

<mw-icon class=""></mw-icon>

calculationUsingIndividualValues = values(1).^2+values(2).^3-values(3).^4

calculationUsingIndividualValues = -1228

calculationsUsingCombinationsOutput = computation(17)

calculationsUsingCombinationsOutput = -1228

Answer 3

[Torsten]

If nothing helps, make a 4-fold nested loop.

for i=1:numel(Span)
    for j = 1:numel(MAC)
        for k = 1:numel(CoM)
            for l = 1:numel(WingArea)
                result(i,j,k,l) = ...
             end
        end
    end
end

[Dyuman Joshi]

Also, best to pre-allocate the output variable.

Answer 4

[Walter Roberson]

Vh = ((Span(:) .* ((1 - reshape(Ct,1,[]))+1).*reshape(MAC,1,1,[])/2) .* reshape(CoM,1,1,1,[]))./(WingArea.*WChord);

assuming that WingArea and WChord are scalars.

Answer 5

[Matt J]

Download ndgridVecs from the File Exhange,

https://www.mathworks.com/matlabcentral/fileexchange/74956-ndgridvecs?s_tid=srchtitle

Span = [1 2];
Ct = [3 4 5];
MAC = [6 7 8 9];
[Span, Ct, MAC] = ndgridVecs(Span, Ct, MAC);


Vh = ((Span .* ((1 - Ct)+1)*MAC/2) .* CoM)./(WingArea.*WChord);

[Walter Roberson]

The * operations are not going to work between 3D arrays.

[Matt J]

Yep. Fixed it.