Optimization with ruler function

Hello everyone, I will make an optimization interface as a homework and I am using a few algorithms. Algorithms are algorithms provided by matlab itself (fmincon, genetic algorithm, particleswarm, pattern search). I would like these algorithms to set all the samples they select in the population they will select to be a multiple of 0.29. It is possible for me to write a function related to this, but how can I embed it in the options options of the ready algorithms

Answer 1

[Shubham]

Hey @Berke Çelik,

I understand that you want to select the samples from population in the multiple of 0.29. These samples are then passed to the optimization algorithms provided by MATLAB.

For this you can write a custom constraint function and pass it in the optimization algorithm itself.

The custom constraint function could be as follows:

% Define the custom constraint function
function [c, ceq] = customConstraint(x)
    % Calculate the remainder of each element in x divided by 0.29
    c = mod(x, 0.29);
    
    % No nonlinear equality constraints, so ceq is empty
    ceq = [];
end

You can then pass the objective function and the constraint function to the optimization algorithm.

You can refer to the example shown in the documentation :

https://www.mathworks.com/help/gads/constraints-in-patternsearch.html

You can also refer to the following :

https://stackoverflow.com/questions/36668656/how-to-pass-arguments-to-a-constraint-defining-function-in-matlab

https://www.mathworks.com/help/optim/ug/how-to-use-all-types-of-constraints.html

https://www.mathworks.com/help/optim/ug/passing-extra-parameters.html

Hope this helps!!