strategy when diversity is weaker and weaker till zero ?

[nicolas baurin]

Hello to all the galib people,

i run a steadystate GA with a population of 100 (30%replacement,
Pmutation=0.001,Pcrossover0.90). 200 cycles later (i 'd like a 10000), i
got convergence since i check a diversity that started at 0.5 to drop
till 0.

What could be done to improve the diversity during the search (i.e.
convergence will come later, a better exploration will be achieved) ?

Thanks in advance,

--
Nicolas Baurin

Doctorant
Laboratoire de Chimiométrie - Modélisation Moléculaire
Institut de Chimie Organique et Analytique, UMR 6005
http://www.univ-orleans.fr/SCIENCES/ICOA/
Université d'Orléans, BP 6759
45067 ORLEANS Cedex 2, France
Tel: (33+) 2 38 49 45 77