I I'm new to matlab and Ikeep getting "Index in position 1 is invalid. Array indices must be positive integers or logical values. Error in "THRUST CURVE FUNCTION LINE 3"

I I'm new to matlab and Ikeep getting "Index in position 1 is invalid. Array indices must be positive integers or logical values. Error in "THRUST CURVE FUNCTION LINE 3"

function output_value = thrust_curve_function(t)
thrust_curve_values = readmatrix('thrust_curve_data.csv');
thrust_curve_t_values = thrust_curve_values(0.01 , 0.5 , 0.1 , 2 , 2.2 , 2.45 , 2.72);
thrust_curve_f_values = thrust_curve_values(50 , 56 , .48 , 24 , 19 , 5 , 0);   
output_value = interp1(thrust_curve_t_values,thrust_curve_f_values,t);
end

[Stephen23]

Regarding 'thrust_curve_data.csv':

• what size is the matrix?
• what do the columns/rows represent?

Regarding input argument t:

• what size is it?

[Ryan]

It has 2 columns and 7 rows , the 2 comlumns represent time and thrust, and the 7 rows represent the 7 seperate values obtained,

And for the last thing I'm not sure what you mean by that, it represents time (secs) if you mean that.

[Ryan]

forgot to post second part of code

start_t_value = 0;

end_t_value = 2.75;

integral@thrust_curve_function(thrust_curve_function), start_t_value, end_t_value;

[Walter Roberson]

thrust_curve_f_values = thrust_curve_values(50 , 56 , .48 , 24 , 19 , 5 , 0);   

Should that be 48 instead of .48 ?

Answer 1

[Sulaymon Eshkabilov]

From your post, what I understood is that you're trying to make linear interpolation of the imported data. I am not 100% sure if this is the case or not :). Anyhow, if this is what I guessed, here is a simple example code how to solve this exercise, e.g.:

FName = 'F_time.csv';  % This is a data file name. The data will be read from this file
t = 0:.1:2.75;         % Input variable t values to be used for interp1()
% NOTE: Input and output variable names can be altered
[thrust_curve_values, output_value] = thrust_curve_function(t, FName); 
plot(thrust_curve_values(:,1), thrust_curve_values(:,2), 'ro'), hold on
plot(t, output_value, 'k-')
legend('thrust_curve_values', 'inperpolated values')
xlabel('t')
ylabel('Output')
grid on

function [thrust_curve_values, output_value] = thrust_curve_function(t, FName)
    thrust_curve_values = readmatrix(FName);
    t_values = thrust_curve_values(:,1);              % Column 1 of the imported data is t
    F_values = thrust_curve_values(:,2);              % Column 2 of the imported data is F(t)
    output_value = interp1(t_values,F_values,t);
end

All the best

[Ryan]

Your right about what I'm trying to do, but I'm still getting errors sadly

[Sulaymon Eshkabilov]

Your error on Line 3 is linked with the index values that must be integers, e.g. 1, 2, 3, ...

Here is a demo example with error on Line 3:

t = 0:.1:2.75;         % Input variable t values to be used for interp1()
[thrust_curve_values, output_value] = thrust_curve_function(t)
function [thrust_curve_values, output_value] = thrust_curve_function(t)
    thrust_curve_values = readmatrix('DATA_file.csv');
    t_values = thrust_curve_values(0.01 , 0.5 , 0.1 , 2 , 2.2 , 2.45 , 2.72);  % Index related error
    F_values = thrust_curve_values(:,2);              % Column 2 of the imported data is F(t)
    output_value = interp1(t_values,F_values,t);
end

Here is a demo example with the fixed index related error on Line 3, etc. Note that we are finding the indexes by using logical index operations - see IDX1 and IDX2

t = 0:.1:2.75;         % Input variable t values to be used for interp1()
[thrust_curve_values, output_value] = thrust_curve_function(t)
function [thrust_curve_values, output_value] = thrust_curve_function(t)
    thrust_curve_values = readmatrix('DATA_file.csv');
    % Note here how indexes are found with respect to the logic condition values.
    % % Index related error fixed
    IDX1=find(thrust_curve_values(:,1)==0.01 | thrust_curve_values(:,1)==0.5...
        | thrust_curve_values(:,1)==0.1 | thrust_curve_values(:,1)==2 | thrust_curve_values(:,1)==2.2 ...
        |thrust_curve_values(:,1)==2.45 | thrust_curve_values(:,1)==2.72);
    % See the values of IDX1
    disp(IDX1)
    t_values = thrust_curve_values(IDX1,1);              % Index related error fixed
    % Note here how indexes are found with respect to the logic condition values.
    % % Index related error fixed
    IDX2=find(thrust_curve_values(:,2)==50 | thrust_curve_values(:,2)==56 ...
        | thrust_curve_values(:,2)==48 | thrust_curve_values(:,2)==24 | thrust_curve_values(:,1)==19 ...
        |thrust_curve_values(:,2)==5 | thrust_curve_values(:,2)==0); % Index related error fixed
    F_values = thrust_curve_values(IDX2,2);              % Column 2 of the imported data is F(t)
    output_value = interp1(t_values,F_values,t);
end

Note that IDX1 need to be sufficient for your planned data take out. Presumable t vs. F(t) from your imported data. because t is independent variable and F(t) is a dependent. Unless you have some special conditions for F(t).

Answer 2

[Walter Roberson]

You should not be using integral for that kind of situation.

You have a table of numeric values, and you are trying to integrate that table. Unless you have an additional thrust model then the best you can justify is to use trapz or cumtrapz

In order to go beyond that you would need to have a function that models the transition between the points.

For example you might choose to interpolate a polynomial of degree 6 over your 7 points, and then proceed to integrate that polynomial of degree 6. This would probably not be a good idea, but it is an example of one way to model the data. Another way would be to use a cubic spline interpolation.