| As one of the most important research areas in optimization method, Multiobjective optimization problem (MOP), is very common in the real-world and it has great value in solving MOP with large practical significance. Different with single objective optimizations, the results of MOP are a set of solutions which cannot be compared with each other. All of these solutions constitute the set of Pareto optimal solution set. The traditional method for solving multi-objective optimization problem has many defects, so how to design efficient optimization algorithms to solve multi-objective optimization problem are very urgent. Since the 80s of last century, the development of optimization algorithm provides a new way of thinking and methods for the development of the optimization theory.The particle swarm optimization (PSO) is a newly emerged swarm intelligence method. Each particle get its valid information to guide the search according to its own experience and the experience of other particles. Since the algorithm has the virtue of simplicity and quick convergence speed, it has been widely used in function optimization, neural networks, fuzzy system control, pattern recognition and other fields, which shows strong vitality.In this paper, a multi-objective particle swarm optimization algorithm based onε-dominant is proposed. The concept ofε-dominant was originally proposed by Deb, which uses parameters setεto obtain the required number of Pareto optimal solutions and divide the entire Pareto optimal surface into several super-cubes, each cube has only one non-dominated individual in order to maintain the distribution of the obtained solution. Meanwhile, the orthogonal design method is used to generate initial population to improve the utilization of initial population. Finally, the proposed algorithm is compared with several classic test functions, and the results of the experiments show that our new approach has good performances in terms of time efficiency, distribution, and convergence. |
PDF Full Text Request