Computer Engineering and Applications ›› 2021, Vol. 57 ›› Issue (22): 102-109.DOI: 10.3778/j.issn.1002-8331.2011-0468

• Theory, Research and Development • Previous Articles     Next Articles

Multi-objective Sparrow Search Algorithm Based on New Crowding Distance

WEN Zeyu, XIE Jun, XIE Gang, XU Xinying   

  1. 1.College of Electrical and Power Engineering, Taiyuan University of Technology, Taiyuan 030024, China
    2.Shanxi Key Laboratory of Advanced Control and Equipment Intelligence, College of Electronic Information Engineering, Taiyuan University of Science and Technology, Taiyuan 030024, China
    3.College of Information and Computer, Taiyuan University of Technology, Jinzhong, Shanxi 030600, China
  • Online:2021-11-15 Published:2021-11-16

基于新型拥挤度距离的多目标麻雀搜索算法

温泽宇,谢珺,谢刚,续欣莹   

  1. 1.太原理工大学 电气与动力工程学院,太原 030024
    2.太原科技大学 电子信息工程学院 先进控制与装备智能化山西省重点实验室,太原 030024
    3.太原理工大学 信息与计算机学院,山西 晋中 030600

Abstract:

More and more complex multi-objective optimization problems have emerged in the real world, and the solution of such complex problems requires efficient optimization algorithms. Based on sparrow search algorithm, this paper proposes Multi-objective Sparrow Search Algorithm(MSSA). Firstly, the sparrow population scale factor is dynamically adjusted according to the external archive convergence to achieve the optimal balance between global exploration and local development. Secondly, the population of sparrow is sorted by non-dominance. Then, polynomial mutation is introduced to the discoverer of sparrow population to enhance the ability of the algorithm to jump out of local optimum. Last, a new crowding distance calculation strategy is proposed, which uses the crowding degree of the solution to eliminate similar individuals, the population is pruned so that the individual does not exceed the upper limit of the archive while maintaining the diversity of the population. The performance of the proposed algorithm is tested through multi-objective functions and the design of a disc brake. MSSA is compared with MOPSO, MOGWO, NSGA-II and SPEA2 on multi-objective test problems, and the experimental results show that MSSA has significant performance advantages in terms of convergence and homogeneity. The disc brake design results show that MSSA can quickly find the non-dominant solution of the problem, which proves the effectiveness of MSSA.

Key words: Sparrow Search Algorithm(SSA), scale factor, external archive, polynomial mutation, crowding distance

摘要:

现实中的多目标问题日益复杂,解决这类问题需要高效的优化算法。基于麻雀搜索算法,提出多目标麻雀搜索算法(Multi-objective Sparrow Search Algorithm,MSSA),对多目标优化问题进行求解。依据外部存档收敛性动态调整麻雀种群比例因子,以达到全局探索能力和局部开发能力的最佳平衡,确保收敛性;对麻雀种群进行非支配排序;对麻雀种群的发现者引入多项式变异因子,增强算法跳出局部最优的能力;设计一种新型拥挤度距离计算策略,利用外部存档解的拥挤度大小剔除相似个体的方法对种群进行裁剪,使个体不超过存档上限的同时维持种群的多样性。分别使用多目标函数和盘式制动器设计测试算法性能。MSSA与MOPSO、MOGWO、NSGA-II和SPEA2在多目标测试函数上进行对比实验,结果表明MSSA算法在收敛性和均匀性两项指标上有显著的优势。盘式制动器仿真结果表明,MSSA可以快速地找到问题的非支配解,证明了该方法的有效性。

关键词: 麻雀搜索算法(SSA), 比例因子, 外部存档, 多项式变异, 拥挤度距离