电信科学 ›› 2020, Vol. 36 ›› Issue (3): 27-33.doi: 10.11959/j.issn.1000-0801.2020068

• 专题:工业互联网平台及安全 • 上一篇    下一篇

混沌改进鱼群算法及其在工业控制网络异常检测中的应用

吴亚萍1,2,董红召1,林盈盈1,柳菁2   

  1. 1 浙江工业大学机械工程学院,浙江 杭州 310032
    2 绍兴文理学院机械与电气工程学院,浙江 绍兴 312000
  • 修回日期:2020-03-13 出版日期:2020-03-20 发布日期:2020-03-26
  • 作者简介:吴亚萍(1979- ),女,浙江工业大学机械工程学院博士生,绍兴文理学院机械与电气工程学院实验师,主要研究方向为智能算法优化、数据挖掘、计算机网络通信等|董红召(1969- ),男,博士,浙江工业大学机械工程学院教授,主要研究方向为智能系统、数据挖掘等|林盈盈(1979- ),女,浙江工业大学机械工程学院博士生,主要研究方向为信息通信等|柳菁(1979- ),女,绍兴文理学院机械与电气工程学院实验师,主要研究方向为计算机应用、算法优化等
  • 基金资助:
    国家自然科学基金资助项目(61773347);浙江省教育厅科学基金资助项目(LY17F030017)

Improved fish swarm algorithm based on chaos and its application in abnormal detection of industrial control network

Yaping WU1,2,Hongzhao DONG1,Yingying LIN1,Jing LIU2   

  1. 1 College of Mechanical Engineering,Zhejiang University of Technology,Hangzhou 310032,China
    2 Mechanical and Electrical Engineering College,Shaoxing University,Shaoxing 312000,China
  • Revised:2020-03-13 Online:2020-03-20 Published:2020-03-26
  • Supported by:
    The National Natural Science Foundation of China(61773347);The Education Department Foundation of Zhejiang Province of China(LY17F030017)

摘要:

人工鱼群算法作为一种新型的仿生群体智能优化算法已成功地应用于各种优化问题及实际工程领域。然而,在复杂优化问题方面,受目标函数存在相对较大数量的局部极值的影响,收敛速度慢、早熟等缺陷难以避免地存在。在基本人工鱼群算法中引入具有随机性、遍历性的混沌理论,构建了一种改进算法,以使得人工鱼群搜索能够规避可能存在的局部极值状况。在工业控制网络异常检测中进行仿真应用,验证了算法的有效性。与基本的人工鱼群算法相比,混沌改进人工鱼群算法可使得算法在局部极值附近长时间搜索的状况得到有效避免,在全局收敛方面算法有更佳表现,且搜索效率更为突出。

关键词: 混沌, Logistic映射, 人工鱼群算法, 优化, 工业控制网络异常检测

Abstract:

Artificial fish swarm algorithm as a new type of bionic swarm intelligence optimization algorithm has been successfully used in a variety of optimization problems and practical engineering field,but when faced with the complex optimization problems,especially the multiple extreme value of peak and multimodal function optimization problems,due to the objective function has many local minima,inevitably there are defects such as premature and slow convergence speed.The random and ergodic theory was introduced into the basic artificial fish swarm algorithm,and an improved algorithm was constructed to make artificial fish swarm search avoid possible local extremum.The effectiveness of the algorithm was validated in the simulation application in industrial control network anomaly detection.Compared with the basic artificial fish school algorithm,the chaos improved artificial fish swarm algorithm can effectively avoid the long-term search of the algorithm near the local extreme value.The algorithm has better performance in terms of global convergence and the search efficiency is more prominent.

Key words: chaos, Logistic map, artificial fish swarm algorithm, optimization

中图分类号: 

No Suggested Reading articles found!