计算机工程与应用 ›› 2021, Vol. 57 ›› Issue (24): 83-89.DOI: 10.3778/j.issn.1002-8331.2012-0080

• 理论与研发 • 上一篇    下一篇

基于正弦因子和量子局部搜索的灰狼优化算法

徐辰华,骆珠光,吴冠宏,刘斌   

  1. 1.广西大学 电气工程学院,南宁 530004
    2.广东技术师范大学 自动化学院,广州 510630
  • 出版日期:2021-12-15 发布日期:2021-12-13

Grey Wolf Optimization Algorithm Based on Sine Factor and Quantum Local Search

XU Chenhua, LUO Zhuguang, WU Guanhong, LIU Bin   

  1. 1.School of Electrical Engineering, Guangxi University, Nanning 530004, China
    2.School of Automation, Guangdong Normal University of Technology, Guangzhou 510630, China
  • Online:2021-12-15 Published:2021-12-13

摘要:

针对基本灰狼优化算法在求解复杂问题时,存在依赖初始种群、过早收敛和易陷入局部最优等缺点,提出一种融合正弦控制因子和量子局部搜索的灰狼优化算法(QGWO)。通过对灰狼算法中的控制因子按照具有正弦变化的曲线变化,使改进后的算法在迭代前期加快收敛速度以快速完成全局搜索,并且在迭代后期减缓收敛速度以提高算法精度。引入量子局部搜索降低算法陷入局部最优的概率。选用12个标准测试函数对QGWO算法性能进行验证,分别从单峰、多峰和固定维测试函数对比分析。实验结果表明,与GWO、WOA、SCA和CGWO相比,QGWO对测试函数的求解有更高的精度和稳定性。通过工程实例优化KELM进行分类实验验证,QGWO表现出更好的寻优性能。

关键词: 改进灰狼优化算法, 正弦因子, 量子局部搜索, 测试函数

Abstract:

In order to solve complex problems, the gray wolf optimization algorithm has some shortcomings, such as relying on the initial population, low convergence accuracy and getting easily trapped into local optima. An improved grey wolf optimization algorithm (Quantum Gray Wolf Optimization Algorithm, QGWO) combining sinusoidal control factor and quantum local search is proposed. The control factors of Gray Wolf algorithm are changed according to the curve with sine change. The improved algorithm accelerates the convergence speed in the early stage of iteration to complete the global exploration quickly, and slows down the convergence speed in the late iteration to improve the accuracy of the algorithm. At the same time, quantum local search is introduced to reduce the probability of the algorithm falling into local optimum. Then, twelve standard test functions are selected to verify the performance of QGWO algorithm, and the single peak, multi peak and fixed dimension test functions are compared. The experimental results show that compared with GWO, WOA, SCA and CGWO, QGWO has higher accuracy and stability in solving test functions. Finally, an engineering example is used to optimize KELM for classification experiments. The results show that QGWO has better optimization performance.

Key words: improved grey wolf optimization algorithm, sine factor, quantum local search, test function