点击次数：

论文类型：会议论文

发表时间：2017-01-01

收录刊物：CPCI-S

卷号：2017-January

页面范围：8716-8720

关键字：multi-objective; PSO; active learning; mutation opterator; sampling

摘要：A multi-objective particle swarm algorithm based on the active learning (MOPSAL) approach is proposed that combines a Multi-Objective particle swarm optimization (MOPSO) with an Pareto Active Learning (PAL) approach. In MOPSAL, the candidate solution set is produced by a sampling method based on mutation operator and preselected by the PAL approach. Then, the best Pareto solution from the candidate solution set is used to guide the search of MOPSO. To validate the performance of MOPSAL, the proposed algorithm compares with the standard multi-objective particle swarm algorithm (MOPSO) and the improved non-dominated sorting genetic algorithm (NSGA-II) for five widely used benchmark problems. The results show the effectiveness of the proposed MOPSAL algorithm.