Self-Adaptive Differential Evolution Algorithm With Alpha-Constrained-Domination Principle For Constrained Multi-Objective Optimization

Feng Qian,Bin Xu,Rongbin Qi,Huaglory Tianfield

DOI: https://doi.org/10.1007/s00500-012-0816-6

IF: 3.732

2012-01-01

Soft Computing

Abstract:Real-world problems are inherently constrained optimization problems often with multiple conflicting objectives. To solve such constrained multi-objective problems effectively, in this paper, we put forward a new approach which integrates self-adaptive differential evolution algorithm with alpha-constrained-domination principle, named SADE-alpha CD. In SADE-alpha CD, the trial vector generation strategies and the DE parameters are gradually self-adjusted adaptively based on the knowledge learnt from the previous searches in generating improved solutions. Furthermore, by incorporating domination principle into alpha-constrained method, alpha-constrained-domination principle is proposed to handle constraints in multi-objective problems. The advantageous performance of SADE-alpha CD is validated by comparisons with non-dominated sorting genetic algorithm-II, a representative of state-of-the-art in multi-objective evolutionary algorithms, and constrained multi-objective differential evolution, over fourteen test problems and four well-known constrained multi-objective engineering design problems. The performance indicators show that SADE-alpha CD is an effective approach to solving constrained multi-objective problems, which is basically enabled by the integration of self-adaptive strategies and alpha-constrained-domination principle.