A Many-Objective Practical Swarm Optimization Based on Mixture Uniform Design and Game Mechanism

1. Introduction

Multi-objective optimization problem(Kong et al., 2010) is one of the main forms of engineering practice and scientific research problems. There are often conflicts between objectives, and the optimal solution is a set of non dominant Pareto optimal solutions. Generally, when the number of objectives is 4 or more, it is called many-objective problem. As the number of conflicting objectives of many-objective problems increases, the number of non dominated solutions increases sharply, resulting in increased computational complexity and search difficulty. It is one of the most difficult problems in the field of intelligent optimization at home and abroad. In order to deal with this problem effectively, in recent years, many researchers have adopted different methods and technologies to solve the key problems in MaOPs from different angles. Here are four types of methods: (1) Pareto based methods; (2) Decomposition based method; (3) Index based approach; (4) Particle swarm optimization based method. Among them, the method based on particle swarm optimization will be the focus of this paper.