Save values in a matrix using a non-integer index

Hello all! i have a major problem in matlab and although i read all the answers couldn't solve this. my code is:

%Moments calculation along beam
%Vertical forces
Oy = 1260.1;     %[KN]
N1 = 210.5;      %[KN]
N2 = 252.8;      %[KN]
N3 = 289;        %[KN]
N4 = 316.8;      %[KN]
N5 = 191;        %[KN]
%Position where the forces are implemented
x1 = 16.68;      %[m]
x2 = 16.78;      %[m]
x3 = 17.89;      %[m]
x4 = 16.93;      %[m]
x5 = 17.15;      %[m]
x6 = 0.936;      %[m]
%Moment at start position
Mo = -64758.212; %[KNm]
%Calculation
for x=0:0.1:85.43
    if x == 0
        M = Mo;
    elseif x <= x1
        M = Mo + Oy*x; 
    elseif x1 < x <= (x1+x2)
        M = Mo + Oy * x - N1*x1;
    elseif (x1+x2) < x <= (x1+x2+x3)
        M = Mo + Oy*x - N1*(x1+x2) - N2*x2;
    elseif (x1+x2+x3) < x <= (x1+x2+x3+x4)
        M = Mo + Oy*x - N1*(x1+x2+x3) - N2*(x2+x3) - N3*x3;
    elseif (x1+x2+x3+x4) < x <= (x1+x2+x3+x4+x5)
        M = Mo + Oy*x - N1*(x1+x2+x3+x4) - N2*(x2+x3+x4) - N3*(x3+x4) - N4*x4;
    end
end

i want to save all the moments (M) in a matrix in order to use it in another equation but M(x) is incorrect as x is a non-integer number. Any help would be really appreciated!!!!

[Theodoros Pardalakis]

%Moments calculation along beam
%Vertical forces
Oy = 1260.1;     %[KN]
N1 = 210.5;      %[KN]
N2 = 252.8;      %[KN]
N3 = 289;        %[KN]
N4 = 316.8;      %[KN]
N5 = 191;        %[KN]
%Position where the forces are implemented
x1 = 16.68;      %[m]
x2 = 16.78;      %[m]
x3 = 17.89;      %[m]
x4 = 16.93;      %[m]
x5 = 17.15;      %[m]
x6 = 0.936;      %[m]
%Moment at start position
Mo = -64758.212; %[KNm]
%Calculation
for x=0:0.1:85.43
  if x == 0
      M = Mo;
  elseif x <= x1
      M = Mo + Oy*x; 
  elseif x1 < x <= (x1+x2)
      M = Mo + Oy * x - N1*x1;
  elseif (x1+x2) < x <= (x1+x2+x3)
      M = Mo + Oy*x - N1*(x1+x2) - N2*x2;
  elseif (x1+x2+x3) < x <= (x1+x2+x3+x4)
      M = Mo + Oy*x - N1*(x1+x2+x3) - N2*(x2+x3) - N3*x3;
  elseif (x1+x2+x3+x4) < x <= (x1+x2+x3+x4+x5)
      M = Mo + Oy*x - N1*(x1+x2+x3+x4) - N2*(x2+x3+x4) - N3*(x3+x4) - N4*x4;
  end
end

[jgg]

Why not index it instead of with x but with the index of x?

 x_vals=0:0.1:85.43;
 M_out = zeros(length(x_vals),1)

Then at the end of your loop do

 M_out(find(x_vals==x,1)) = M;

Would something like that work? In this situations, row i of M_out corresponds to row i of x_vals so you can back out the relevant values.

As an aside, this is usually why when you write for loops it's a good idea to index with integers when possible.

[Steven Lord]

FYI this doesn't do what you think it does.

x1 < x <= (x1+x2)

• If x1+x2 is greater than or equal to 1 this is always true.
• If x1+x2 is greater than or equal to 0 but less than 1, this is ~(x1 < x).
• If x1+x2 is less than 0 this is always false.

You need to use something like this to do what you want:

(x1 < x) & (x <= (x1+x2))

[Theodoros Pardalakis]

Steven Lord THANK YOU!!!!!!!!!!!!!!! i hadn't see your answer and that was the problem for getting wrong values. YOU SAVED ME! i really appreciate this. THANK YOU

Answer 1

[the cyclist]

Here's one way:

%Calculation

x = 0:0.1:85.43;
NX = numel(x);

M = zeros(NX,1);

for xi = 1:NX
  if x(xi) == 0
      M(xi) = Mo;
  elseif x(xi) <= x1
      M(xi) = Mo + Oy*x(xi); 
  elseif x1 < x(xi) <= (x1+x2)
      M(xi) = Mo + Oy * x(xi) - N1*x1;
  elseif (x1+x2) < x(xi) <= (x1+x2+x3)
      M(xi) = Mo + Oy*x(xi) - N1*(x1+x2) - N2*x2;
  elseif (x1+x2+x3) < x(xi) <= (x1+x2+x3+x4)
      M(xi) = Mo + Oy*x(xi) - N1*(x1+x2+x3) - N2*(x2+x3) - N3*x3;
  elseif (x1+x2+x3+x4) < x(xi) <= (x1+x2+x3+x4+x5)
      M(xi) = Mo + Oy*x(xi) - N1*(x1+x2+x3+x4) - N2*(x2+x3+x4) - N3*(x3+x4) - N4*x4;
  end
end

[Theodoros Pardalakis]

Thank you for the reply.. i tried something similar with this and works fine :)

[the cyclist]

Be sure to see Steven Lord's comment. I didn't notice that potential problem with your code.

Answer 2

[the cyclist]

You can also do the whole calculation in a vectorized fashion, by replacing the conditional statement like this:

MM = Mo + (x<=x1).*Oy.*x - (x<=(x1+x2)).*N1*x1 ... % and so on

Answer 3

[Theodoros Pardalakis]

Hello again! After i run the code several times with various changes and trying all the solutions you all proposed i have another isue. Matlab gives me wrong results. By wrong i mean that if i solve each equation of the code by hand i get another result(the correct one). For example at this code the diagram of the moments shouldn't have positive values but Matlab gives me also positive values after a specific loop. can anyone help me please? Do i use the elseif correct? I have to progress my report and i am stuck at this so long....

%Moments calculation along beam %Vertical forces Oy = 1260.1; %[KN] N1 = 210.5; %[KN] N2 = 252.8; %[KN] N3 = 289; %[KN] N4 = 316.8; %[KN] N5 = 191; %[KN] %Position where the forces are implemented x1 = 16.68; %[m] x2 = 16.78; %[m] x3 = 17.89; %[m] x4 = 16.93; %[m] x5 = 17.15; %[m] x6 = 0.936; %[m] R1 = 16.68; R2 = 33.46; R3 = 51.35; R4 = 68.28; R5 = 85.43; %Moment at start position Mo = -64758.212; %[KNm] %Calculation x=0:0.01:85.43; for l=1:length(x) if x(l) == 0 M(l) = Mo; elseif 0 < x(l) <= R1 M(l) = Mo + Oy*x(l); elseif R1 < x(l) <= R2 M(l) = Mo + Oy*x(l) - N1*(x(l)-R1); elseif R2 < x(l) <= R3 M(l) = Mo + Oy*x(l) - N1*(x(l)-R1) - N2*(x(l)-R2); elseif R3 < x(l) <= R4 M(l) = Mo + Oy*x(l) - N1*(x(l)-R1) - N2*(x(l)-R2) - N3*(x(l)-R3); elseif R4 < x(l) <= R5 M(l) = Mo + Oy*x(l) - N1*(x(l)-R1) - N2*(x(l)-R2) - N3*(x(l)-R3) - N4*(x(l)-R4); end end

[Theodoros Pardalakis]

    %Moments calculation along beam
%Vertical forces
Oy = 1260.1;     %[KN]
N1 = 210.5;      %[KN]
N2 = 252.8;      %[KN]
N3 = 289;        %[KN]
N4 = 316.8;      %[KN]
N5 = 191;        %[KN]
%Position where the forces are implemented
x1 = 16.68;      %[m]
x2 = 16.78;      %[m]
x3 = 17.89;      %[m]
x4 = 16.93;      %[m]
x5 = 17.15;      %[m]
x6 = 0.936;      %[m]
R1 = 16.68;
R2 = 33.46;
R3 = 51.35;
R4 = 68.28;
R5 = 85.43;
%Moment at start position
Mo = -64758.212; %[KNm]
%Calculation
x=0:0.01:85.43;
for l=1:length(x)
    if x(l) == 0
        M(l) = Mo;
    elseif 0 < x(l) <= R1
        M(l) = Mo + Oy*x(l);
    elseif R1 < x(l) <= R2
        M(l) = Mo + Oy*x(l) - N1*(x(l)-R1);
    elseif R2 < x(l) <= R3
        M(l) = Mo + Oy*x(l) - N1*(x(l)-R1) - N2*(x(l)-R2);
   elseif R3 < x(l) <= R4
        M(l) = Mo + Oy*x(l) - N1*(x(l)-R1) - N2*(x(l)-R2) - N3*(x(l)-R3);
   elseif R4 < x(l) <= R5
        M(l) = Mo + Oy*x(l) - N1*(x(l)-R1) - N2*(x(l)-R2) - N3*(x(l)-R3) - N4*(x(l)-R4);
   end
end

[the cyclist]

Did you carefully read and understand Steven Lord's comment? It is almost certainly the case that your if-else statements are not doing what you think.