Abstract (EN):
The development of the crossover operator is based on three different mechanisms:
mating selection mechanism, offspring generation mechanism and offspring selection mechanism.
Most of the crossover operators are able to get exploration or exploitation of the domain
depending on the way in which they handle the current diversity of the population. Each
crossover operator directs the search towards a different zone in the neighbourhood of the
parents. The quality of the elements that belong to the visited region depends on the particular
problem to be solved. This is confirmed by the well known no free lunch theorems. The
simultaneous use of diverse crossover operators on the population will induce more efficient
algorithms. The aiming of this paper is to analyze and to study the complementary properties
resulting from the synergetic effects using several crossover operators in genetic algorithms.
The improvements reached in structural optimization using hybrid crossover operators will
be analyzed through some standard examples.
Language:
English
Type (Professor's evaluation):
Scientific
No. of pages:
16