Abstract:

This paper presents a filled function algorithm based on filter technique for the nonconvex global optimization problems with boxed constrains. The filled function method is one of effective methods for solving global optimization problems. And the filter technique is widely used in local optimization algorithm because of its good numerical results. In order to optimize the filled function method, we try to use the filter set to supervise the iteration. In the paper, a new filled function is formulated first and then its necessary characteristics are discussed. Based on that, the algorithm dominated by the filter is proposed and its properties are proved. The numerical results are list at last to show the effectiveness of the algorithm.

Key words: boxed constrains, global optimization, filter technique, filled function