Adaptive Methods of Differential Evolution Multi-objective Optimization Algorithm Based on Decomposition

Decomposition-based algorithms e.g., multi-objective evolutionary algorithm based on decomposition (MOEA/D) has been proved as an effective and useful solution in a variety of multi-objective optimization problems (MOPs). On the basis of MOEA/D, the MOEA/D-DE replaces the simulated binary crossover (SBX) operator, which is used to enhance the diversity of the solutions, into differential evolution (DE) operator. However, the amplification factor and the crossover probability are fixed in MOEA/D-DE, which would lead to a low convergence rate and be more likely to fall into local optimum. To overcome such prematurity problem, this paper proposes three different adaptive operators in DE to adjust the parameter settings adaptively, including crossover probability and amplification factor. This paper also designs a changeable parameter η in the proposed algorithms. Several experiments are set to explore how the η would affect the convergence of the proposed algorithms. These adaptive algorithms are tested on many benchmark problems in comparison to MOEA/D-DE. The experimental results illustrate that the three proposed adaptive algorithms have better performance on the most benchmark problems.