Toward efficient global optimization in large dynamic systems-the adaptive complex method

The efficient solution of global optimization problems in large dynamic systems requires methods that are robust, do not require analytical expressions for the objective function or its derivatives and require function evaluations only on the order O(n), where n is the dimension of the search-variable space. The adaptive complex method operated in a staged feed-ahead mode has been found to possess these qualities. The method is described along with tests using classical multimodal functions. The tests indicate that the method is efficient relative to other methods. Tests with higher-dimension multimodal problems (with spaces of dimension 22 and 36) are described, and the results indicate that the method can be employed to solve efficiently much larger problems of the class described