科研之友
微信
新浪微博
Facebook
分享链接
摘要
In this paper a Modified Particle Swarm Optimization (PSO) method for multi-objective (MO) problems with a discrete decision space is proposed. In the PSO method the procedure to determine inertia weight, learning factor and social factor is modified. In addition, both an elitism strategy and innovative deceleration mechanism preventing the particles from going beyond the limits of decision space are introduced. The proposed approach has been applied to a series of currently used test functions as well as to optimization problems connected with finish hard turning operation, where the obtained results have been compared with those obtained by means of Genetic Algorithms (GA). The results indicate that the proposed approach is relatively quick, and thus it is highly competitive with other optimization methods. The authors have obtained a very good diversity, convergence and a maximum range of the Pareto front in the criteria space. In order to assess the quality of the generated Pareto set for each of presented examples, a rating has been determined based on the entropy measurement and inverted generational distance (IGD).
- 出版日期2014
