Genetic algorithm hybrid scheme with integer constraints

IntCon = 3;
nonlcon = @yueshu;
options = optimoptions('ga','PlotFcn',{@gaplotbestf,@gaplotstopping},'Display','diagnose',...
    'MaxStallGenerations',5,'FunctionTolerance',1e-3,'MaxGenerations',100,'PopulationSize',100,'EliteCount',10,'CrossoverFraction',0.8,...
    'UseVectorized',false);
fun = @youhua_ga0201;
lb = [0.0002 0.0002 5];
ub = [0.001 0.002 19];
rng default % For reproducibility
[x,fval,exitflag,output] = ga(fun,3,[],[],[],[],lb,ub,nonlcon,IntCon,options)

In this code, there is an integer constraint, I want to improve the genetic algorithm to use the mixing scheme in the optimization toolbox, but because of the integer constraint, the runtime is warned to ignore the mixing function, how to solve it？

[Alan Weiss]

This seems to be the same question as https://www.mathworks.com/matlabcentral/answers/1944489-when-the-genetic-algorithm-has-integer-constraint-issues-how-to-customize-the-variogram-to-eliminat#answer_1212834

As I told you there, you need a version of MATLAB of R2021b or later in order to have a custom crossover, mutation, or creation function. Is there something else that you are asking that is different?

Alan Weiss

MATLAB mathematical toolbox documentation

[杨杨 蒋]

No, this time I didn't use a custom variogram, but the "HybridFcn" function in the ga(option) structure, and my version is 2022b, but it still prompts to ignore the hybrid function,