Improved differential evolution algorithm and its applications to orbit design

A double self-adaptive differential evolution algorithm with a random mutant was proposed. When a random mutant and a double self-adaptive scaling factor are introduced into the traditional differential evolution algorithm, the scaling factors of the proposed algorithm can adjust with the optimization procedure, and the algorithm can jump out of the local optimal. When the algorithms are applied to several test function studies including low dimension and high dimension and compared with the other algorithms, the simulations demonstrated that the advanced algorithm can give a better performance in solution accuracy, convergence, and the results' standard deviation. The case studies prove that the novel self-adaptive algorithm with random mutant can provide a better performance on multitarget, maneuver optimal problems than others.