An Immune Clonal Algorithm for Dynamic Multi-Objective Optimization

Abstract

The difficulty of Dynamic Multi-Objective Optimization (DMO) problem lies in either the objective function and constraint or the associated problem parameters variation with time. In this paper, based on the immune clonal theory, a new DMO algorithm termed as Immune Clonal Algorithm for DMO (ICADMO) is proposed. In the algorithm, the entire cloning is adopted and the clonal selection based on the Pareto-dominance is adopted. The individuals in the antibody population are divided into two parts: Dominated ones and non-dominated ones, and the non-dominated ones are selected. Three operators are introduced into ICADMO, which guarantees the diversity, the uniformity and the convergence of the obtained solutions. ICADMO is tested on four DMO test problems and compared with the Direction-Based Method (DBM), and much better performance in both the convergence and diversity of the obtained solutions is observed.