[1]CHEN Qiang,WANG Yujia,LIANG Haina,et al.Multi-objective particle swarm optimization algorithm based on an objective space papping strategy[J].CAAI Transactions on Intelligent Systems,2021,16(2):362-370.[doi:10.11992/tis.202006042]
Copy
CAAI Transactions on Intelligent Systems[ISSN 1673-4785/CN 23-1538/TP] Volume:
16
Number of periods:
2021 2
Page number:
362-370
Column:
学术论文—人工智能基础
Public date:
2021-03-05
- Title:
-
Multi-objective particle swarm optimization algorithm based on an objective space papping strategy
- Author(s):
-
CHEN Qiang; WANG Yujia; LIANG Haina; SUN Xin
-
School of Electronic and Electrical Engineering, Shanghai University of Engineering Science, Shanghai 201620, China
-
- Keywords:
-
objective space mapping strategy; performance index; opposition learning; particle swarm optimization; high-dimensional multi-objective optimization; Pareto based criterion; convergence; diversity
- CLC:
-
TP18
- DOI:
-
10.11992/tis.202006042
- Abstract:
-
To balance the relationship between the convergence and diversity of the optimization algorithm in the multi-objective problem, the selection pressure of the algorithm is increased. A high-dimensional MOPSO-OSM (multi-objective particle swarm optimization algorithm based on objective space mapping strategy) is proposed in this paper. When solving high-dimensional multi-objective optimization problems, the Pareto based criterion cannot identify the best compromise solutions from many nondominated solutions. Therefore, the high-dimensional multi-objective optimization space is mapped into two-dimensional space based on indexes of convergence and diversity. Then, the two-dimensional space is divided into four regions according to the performance index. Simultaneously, the ability of the jumping local optimal solution is improved using the opposition learning strategy. The experimental results show that MOPSO-OSM can balance the relationship between convergence and diversity and solve complex problems.