This function keeps returning the error"ttempted to access wf(2); index out of bounds because numel(wf)=1. "

This function keeps returning this error and I cannot figure out how to fix it. Can anyone help?

function Powerval=Analyse_Grid(wf)
global wind_farm;
global N;
global velocity_farm;
N = 30;
wind_farm = zeros(N,2);
velocity_farm = zeros(N,1);
wf = wf * 1000;
wf = round(wf);
for a = 1:1:N
 b = (2 * a) - 1;
 wind_farm(a,1) = wf(b);
 wind_farm(a,2) = wf(b + 1);
end
power = 0;
total_power = 0;
wind_farm = sortrows(wind_farm,[2 1]);
for i = 1:1:N

   x = wind_farm(i,1);
   y = wind_farm(i,2);

   velocity = check_wake(x,y,i);
   velocity_farm(i) = velocity;
   power = 0.3 * (velocity ^ 3);

   total_power = total_power + power;

end

Answer 1

[Roger Stafford]

Matlab states that the number of elements it finds in 'wf' is 1. It must mean that you passed a scalar 'wf' to 'Powerval' when you called on it. You had better check on how you called on it, because your code within the first for-loop calls for it to have at least 60 elements in it.

Answer 2

[Star Strider]

You have defined ‘wf’ as an input argument. What are you supplying to it in your call to ‘Analyse_Grid’?

It wants a vector of length (2*N-1), that since you defined ‘N=30’, would be 59.

At least that’s how I interpret your code.

[Lorcan O'Toole]

I supply 60 to f but it still returns an error, I cannot see how I can call it differently! Thanks for the help all the same much appreciated!

[Star Strider]

My pleasure!

Supplying a scalar value of 60 and a 60-element vector (that the function wants) are two different situations. The scalar value will quite simply not work.

When I strip the code out of the function (and delete the global calls), I get:

wf = linspace(100, 1, 60);

N = 30;
wind_farm = zeros(N,2);
velocity_farm = zeros(N,1);
wf = wf * 1000;
wf = round(wf);
for a = 1:1:N
 b = (2 * a) - 1;
 wind_farm(a,1) = wf(b);
 wind_farm(a,2) = wf(b + 1);
end
power = 0;
total_power = 0;
wind_farm = sortrows(wind_farm,[2 1]);
for i = 1:1:N

     x = wind_farm(i,1);
     y = wind_farm(i,2);

     velocity = check_wake(x,y,i);
     velocity_farm(i) = velocity;
     power = 0.3 * (velocity ^ 3);

     total_power = total_power + power;

end

This works until it throws the error:

Undefined function 'check_wake' for input arguments of type 'double'.

Error in AnswersTestScript (line 20263)
   velocity = check_wake(x,y,i);

So give it a 60-element vector and see how it goes.

[Lorcan O'Toole]

Thanks dude ! It works without the function but I was trying to use this function along with 2 constraint functions in a G.A . The purpose is to optimize the layout of a Wind farm. Do you think Ill have to start again? Thanks Again man !

[Star Strider]

My pleasure!

I don’t know how you’re using it in GA. If ‘wf’ is a ‘chromosome’ (or ‘individual’), then it needs to be a vector, and possibly a vector of parameters. The GA algorithm then optimises the parameters according to your designated ‘fitness function’.

I don’t have significant recent experience with the MATLAB GA functions (I wrote a number of my own in FORTRAN ‘way back when), so I would be reluctant to advise you specifically on the rest of your code, but consider that if ‘Analyse_Grid’ is your fitness function (or called by it), ‘wf’ nevertheless needs to be a vector.