Neuro-fuzzy genetic algorithm

Premature convergence is a classical problem in finding optimal solution in Genetic Algorithms (GAs). The population diversity is a way of avoiding the premature convergence in a GA. If the population diversity is low, the GA will converge very quickly. On the other hand, if the population diversity is high, the GA will takes a lot of time to converge and this may caused wastage in computational resources. This paper proposes a new variant of GA: neuro-fuzzy genetic algorithm with sexual selection. The motivation of this algorithm is to maintain the population diversity throughout the search procedure. To promote diversity, the proposed algorithm combines the concept of gender and age of individuals and the fuzzy logic during the selection of parents. The goal of this technique is to maintain suitable diversity in the population and to prevent GA from converge prematurely to local optimal. Computational experiments are conducted to compare the performance of this new technique with some commonly used mechanisms found in a standard GA from literature for solving the well known Generalised Rosenbrock’s Function.

Text ntc2009_proc-1.pdf Restricted to Repository staff only Download (167kB)

Actions (login required)

View Item