A novel differential evolution for dynamic multiobjective optimization with adaptive immigration scheme

Abstract

Dynamic Multi-objective Optimization(DMO) is a challenge research topic because the problems it involves are multi-objective and constantly change in time. Maintaining the diversity of the population in the evolutionary process is very important for DMO. When the diversity loss is too fast, tracing the Pareto Optimal Front(POF) will become very difficult. In this paper, an adaptive immigration scheme is proposed and integrated into a the DE algorithm for Dynamic Multi-objective Optimization. The immigration scheme can improve the diversity of the population in the evolutionary process so the proposed algorithm can work well in the dynamic multi-objective optimization environment. The proposed algorithm is tested against a variety of benchmark function types and its performance is compared to the NSGAII-B algorithm.

Dynamic Multi-Objective Evolutionary Algorithms: An Overview

“…Immigrants are generated by mutating randomly selected elite individuals. In each generation, HIDMODE [75] adaptively generates immigrants using a hybrid immigration scheme. The degree of diversity of the current population is calculated.…”