计算机工程与应用 ›› 2019, Vol. 55 ›› Issue (13): 145-150.DOI: 10.3778/j.issn.1002-8331.1803-0444

• 模式识别与人工智能 • 上一篇    下一篇

结合高斯分布的改进二进制灰狼优化算法

陈长倩,慕晓冬,牛  犇,王立志   

  1. 火箭军工程大学 作战保障学院,西安 710025
  • 出版日期:2019-07-01 发布日期:2019-07-01

Improved Binary Grey Wolves Optimization Algorithm Combined with Gaussian Distribution

CHEN Changqian, MU Xiaodong, NIU Ben, WANG Lizhi   

  1. College of War Support, Rocket Force University of Engineering, Xi’an 710025, China
  • Online:2019-07-01 Published:2019-07-01

摘要: 针对灰狼优化算法(GWO)解决离散问题应用较少,发展不成熟的现状,提出一种用于解决二进制问题的离散灰狼优化算法(BGWO)。针对混沌搜索在解决二进制问题时,产生的初始种群较为集中的问题,引入高斯分布曲线对种群初始化,使初始种群地空间分布更加均匀;提出一种转换函数,对GWO进行二进制化处理;通过典型测试函数对该算法性能进行验证,实验表明该算法收敛精度明显优于其他算法。将该算法用于实际背包问题的求解,结论表明该算法迭代次数更少,求解精度更高。

关键词: 二进制灰狼优化(BGWO), 高斯分布, 背包问题, 最优化选择

Abstract: In order to solve the problem that the Gray Wolf Optimization(GWO) algorithm is less utilized and developes immature on discrete issues, a Binary Gary Wolf Optimization(BGWO) algorithm is proposed. Firstly, aiming at the problem of chaos search that the initial population is more concentrated in solving binary problems, the Gaussian distribution curve is introduced, which makes the spatial distribution of initial population more uniform. Secondly, a transfer function is proposed to binarize the GWO. Then the performance of the algorithm is tested by the typical test function. The simulation results show that the proposed BGWO algorithm has better performance in precision. Finally, the BGWO is used to solve the knapsack problem. The conclusion shows that the BGWO has fewer iterations and higher solution accuracy.

Key words: Binary Gary Wolf Optimization(BGWO) algorithm, Gaussian distribution, knapsack problem, optimization choice