A special point-based transfer component analysis for dynamic multi-objective optimization

Abstract To solve dynamic multi-objective optimization problems better, the key is to adapt quickly environmental changes and track possible changing optimal solutions in time. In this paper, we propose a special point-based transfer component analysis for algorithm (SPTr-RM-MEDA). be specific, when change occurs, neighbors of some points are selected from set at previous time, makes use minimizing distance between mapped optima current optima. Accordingly, purpose predict part next initial population neighborhoods by analysis. well, SPTr-RM-MEDA also reevaluates set. addition, an adaptive diversity introduction strategy adopted maintain size. performed on 12 test under 8 kinds changes, experimental results show that it superior other five state-of-the-art algorithms most problems.