Constraint dependent in mutatuon

Can anyone help me with description about "constraint dependent" for mutation in GA optimization?

Answer 1

[Alan Weiss]

I'm not sure what you want to know. The built-in mutationadaptfeasible function is briefly described here:

"Adaptive Feasible (mutationadaptfeasible), the default mutation function when there are constraints, randomly generates directions that are adaptive with respect to the last successful or unsuccessful generation. The mutation chooses a direction and step length that satisfies bounds and linear constraints."

In other words, the mutation strictly satisfies bounds and linear constraints. It does not attempt to satisfy nonlinear constraints. It satisfies the bounds and linear constraints by internally using a QR algorithm to generate trial points along the active set boundary and then checking for feasibility of the resulting steps to ensure that all constraints remain satisfied. You can read the code by entering

edit mutationadaptfeasible

Alan Weiss

MATLAB mathematical toolbox documentation

[Muhammad Tayyab]

then what about nonlinear constraints, as you said that it does not attempt to satisfy nonlinear constraints. In my case, all of my constraints are nonlinear, then what should I understand about the constraint dependent mutation in this case? how can I define in my article that how mutation operator worked in my case?

Answer 2

[hoofar hemmatabady]

Hi,

So the mutation process is completely random and it is not like ga (Vary Mutation and Crossover - MATLAB & Simulink (mathworks.com)), in which 'MaxStallGenerations', shrink and scale define the mutation procedure?

Does 'MaxStallGenerations' have an effect on the mutation prcocess of Gamutiobj as well?

Thank you in advance

Best regards