Evolutionary Many-Objective Optimization Algorithm Based on Improved K-Dominance and M2M Population Decomposition

This paper improves K-dominated sorting by M2M population decomposition [14] to enhance the population convergence and diversity, and propose a new evolutionary many objective optimization algorithm based on the improved Kdominated sorting. The proposed K-dominated sorting can maintain the population diversity by M2M population decomposition. What´s more, we prove that the improved K-dominance can avoid circular dominance of the original K-dominance in theory. Compared with the Pareto-dominance, the improved K-dominance greatly increases the selection pressure of evolutionary algorithm when optimizing many-objective optimization problems. A new local density estimation method is used to improve population diversity. The contrast experiments of proposed algorithm and NSGA-II are conducted by optimizing the DTLZ series test functions. The simulation results show that the proposed algorithm has obvious advantages, which can not only improves the population convergence, but also ensure a good distribution along the Pareto-optimal Front.