'Index exceeds array dimensions. Index value 0 exceeds valid range [1-12] for array 'A'.' I get this error when I run the code? Can you please help?

function cx_value = cx( alpha, el )

A = [ -0.099 -0.081 -0.081 -0.063 -0.025 0.044 0.097 0.113 0.145 0.167 0.174 0.166;

-0.048 -0.038 -0.040 -0.021 0.016 0.083 0.127 0.137 0.162 0.177 0.179 0.167;

-0.022 -0.020 -0.021 -0.004 0.032 0.094 0.128 0.130 0.154 0.161 0.155 0.138;

-0.040 -0.038 -0.039 -0.025 0.006 0.062 0.087 0.085 0.100 0.110 0.104 0.091;

-0.083 -0.073 -0.076 -0.072 -0.046 0.012 0.024 0.025 0.043 0.053 0.047 0.040];

A = A';

row =3 ; %3

col =3; %3

s = 0.2 * alpha;

k = fix ( s );

if ( k <= -2 )

k = -1;

end

if ( k >= 9 )

k = 8;

end

da = s - k;

l = k + fix ( sign ( da ) * 1.1 );

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

% added as boundary condition

if l < -2

l = -2;

elseif l > 9

l = 9;

end

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

s = el / 12.0;

m = fix ( s );

if ( m <= -2 )

m = -1;

end

if ( m >= 2)

m = 1;

end

de = s - m;

n = m + fix ( sign ( de ) * 1.1 );

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

% added as boundary condition

if n < -2

n = -2;

elseif n > 2

n = 2;

end

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

t = A( nansum(k+row), nansum(m+col) ); %numsum ı ben ekledim.

u = A( nansum(k+row), nansum(n+col) );

v = t + abs( da ) * ( A( l+row, m+col ) - t );

w = u + abs( da ) * ( A( l+row, n+col ) - u );

cx_value = v + ( w - v ) * abs( de );

[VBBV]

if you pass alpha & el as vectors , then function errs due to dimension inconsistency.

You might want to avoid such errors by calling function within a loop as below

for k = 1: length(alpha)
   cx_value(k) = cx( alpha(k), el(k) );
end
% do something next 

[Alperen Köse]

Where do I add this in the code?

[VBBV]

in the m-file that calls the simulink model you posted

[Alperen Köse]

Thank you, I will try.

Answer 1

[Mathieu NOE]

hello

I replaced nansum which is outdatted by sum with omitnan parameter

seems to work, at least for the input data I picked (you did not provide any test data)

% test function with data ?? 
cx_value = cx( 10, 10 )

cx_value = 0.0103

function cx_value = cx( alpha, el )
A = [ -0.099  -0.081  -0.081  -0.063  -0.025  0.044  0.097  0.113  0.145  0.167  0.174  0.166;
      -0.048  -0.038  -0.040  -0.021   0.016  0.083  0.127  0.137  0.162  0.177  0.179  0.167;
      -0.022  -0.020  -0.021  -0.004   0.032  0.094  0.128  0.130  0.154  0.161  0.155  0.138;
      -0.040  -0.038  -0.039  -0.025   0.006  0.062  0.087  0.085  0.100  0.110  0.104  0.091;
      -0.083  -0.073  -0.076  -0.072  -0.046  0.012  0.024  0.025  0.043  0.053  0.047  0.040];
  
A = A';  
row =3 ;  %3
col =3;   %3
s = 0.2 * alpha;
k = fix ( s );
if ( k <= -2 )
    k = -1;
end
if ( k >= 9 )
    k = 8;
end
da = s - k;
l = k + fix ( sign ( da ) * 1.1 );
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% added as boundary condition
if l < -2
    l = -2;
elseif l > 9
        l = 9;
end
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
s = el / 12.0;
m = fix ( s );
if ( m <= -2 )
   m = -1;
end
if ( m >= 2)
   m = 1;
end
de = s - m;
n = m + fix ( sign ( de ) * 1.1 );
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% added as boundary condition
if n < -2
    n = -2;
elseif n > 2
        n = 2;
end
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% t = A( nansum(k+row), nansum(m+col) );  %numsum ı ben ekledim.
% u = A( nansum(k+row), nansum(n+col) );
t = A( sum(k+row,"omitnan"), sum(m+col,"omitnan") );  %numsum ı ben ekledim.
u = A( sum(k+row,"omitnan"), sum(n+col,"omitnan") );
v = t + abs( da ) * ( A( l+row, m+col ) - t );
w = u + abs( da ) * ( A( l+row, n+col ) - u );
cx_value = v + ( w - v ) * abs( de );
end

[Alperen Köse]

I'm trying to extract the F16 model. The cx_value changes with the 'alpha' and 'el' entries. But it gives error. If you want to take a closer look, I can send you the whole model.

[Alperen Köse]

[Mathieu NOE]

so that m code is inside a simulink block ... and what are the dimensions of alpha and elv in your simulink file ?

could you test the m code block alone in a separate simplified simulink file ?

if you want me to test your simulink file I need it in R2020b format (yes I know I'm not super up to date)

[Alperen Köse]

I have attached the model, thank you for your interest.

[Alperen Köse]

Any news? :)

[Mathieu NOE]

trying to do my official work too ...

[Alperen Köse]

Sorry, I was hasty. Have a nice day.