Abstract: This paper proposes a gravitational search algorithm of adaptive chaos mutation. By introducing the average distance and the chaotic search variation, this paper improves local search ability of gravity algorithm, increases the diversity of substance population and treats the unfeasible materials after mutation by boundary variation constraint. The results show that the new algorithm has the advantages of high precision and fast rate in convergence which can effectively avoid premature convergence. And the variation of the material after not feasible material treated by boundary variationconstraint.

Key words: gravitational search algorithm; chaos; average particle distance; boundary mutation