why population size give bad result if it more than 50

Dear all,

I'm trying to apply the genetic algorithm, when I choose population size more than 50 in 3 dimension it works and give good result where my individual size is 7, but when I rise the dimension more than 3 I can't get result if the size more than 50 for the same code, in 4 dimension my individual size is 9,and in 5 dimension my individual size is 11 and so on.

Regrads,

[Geoff Hayes]

Imola - please clarify what you mean by individual size. How does this value relate to the genetic algorithm? Are you using the Genetic Algorithm software from the Global Optimization Toolbox or are you using your own version of the algorithm?

[imola]

Dear Geoff,

sorry for late, I'm using the Genetic Algorithm software from the Global Optimization Toolbox and i mean by individual size the number of variables.

regards

Answer 1

[Walter Roberson]

The more dimensions you have, the larger your search space, and the less of that space any given number of iterations can cover. You may need to increase your iterations; increase Generations in your options structure.

[imola]

Dear Walter,

Thanks for your advice, I will try it and see what happen.

Regards