2.765

2022影响因子

(CJCR)

  • 中文核心
  • EI
  • 中国科技核心
  • Scopus
  • CSCD
  • 英国科学文摘

留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

面向混合属性数据集的改进半监督FCM聚类方法

李晓庆 唐昊 司加胜 苗刚中

李晓庆, 唐昊, 司加胜, 苗刚中. 面向混合属性数据集的改进半监督FCM聚类方法. 自动化学报, 2018, 44(12): 2259-2268. doi: 10.16383/j.aas.2018.c170510
引用本文: 李晓庆, 唐昊, 司加胜, 苗刚中. 面向混合属性数据集的改进半监督FCM聚类方法. 自动化学报, 2018, 44(12): 2259-2268. doi: 10.16383/j.aas.2018.c170510
LI Xiao-Qing, TANG Hao, SI Jia-Sheng, MIAO Gang-Zhong. An Improved Semi-supervised FCM Clustering Method for Mixed Data Sets. ACTA AUTOMATICA SINICA, 2018, 44(12): 2259-2268. doi: 10.16383/j.aas.2018.c170510
Citation: LI Xiao-Qing, TANG Hao, SI Jia-Sheng, MIAO Gang-Zhong. An Improved Semi-supervised FCM Clustering Method for Mixed Data Sets. ACTA AUTOMATICA SINICA, 2018, 44(12): 2259-2268. doi: 10.16383/j.aas.2018.c170510

面向混合属性数据集的改进半监督FCM聚类方法

doi: 10.16383/j.aas.2018.c170510
基金项目: 

国家重点研发计划 2017YFB0902600

国家自然科学基金 61573126

详细信息
    作者简介:

    李晓庆  合肥工业大学电气与自动化工程学院博士研究生.2013年获得合肥工业大学学士学位.主要研究方向为故障预测及健康管理.E-mail:lixiaoqing@mail.hfut.edu.cn

    司加胜  合肥工业大学智能制造技术研究院硕士研究生.2015年获得东北大学学士学位.主要研究方向为故障预测与健康管理.E-mail:jasenchn@hotmail.com

    苗刚中  合肥工业大学电气与自动化工程学院副教授.1991年获合肥工业大学工程硕士学位.主要研究方向为电工与电子技术, 物联网相关技术, 数据挖掘, 移动手机软件开发.E-mail:miaogzh@126.com

    通讯作者:

    唐昊  合肥工业大学电气与自动化工程学院教授.2002年获得中国科学技术大学博士学位.主要研究方向为离散事件动态系统, 随机决策与优化理论, 智能优化与控制方法.本文通信作者.E-mail:htang@hfut.edu.cn

An Improved Semi-supervised FCM Clustering Method for Mixed Data Sets

Funds: 

National Key Research and Development Program of China 2017YFB0902600

National Natural Science Foundation of China 61573126

More Information
    Author Bio:

       Ph. D. candidate at the School of Electrical Engineering and Automation, Hefei University of Technology. She received her bachelor degree from Hefei Unive- \noindent rsity of Technology in 2013. Her research interest covers prognostic and health management

      Master student at the Intelligent Manufacturing Institute, Hefei University of Technology. He received his bachelor degree from Northeastern University in 2015. His research interest covers prognostic and health management

      Associate professor at the School of Electrical Engineering and Automation, Hefei University of Technology. He received his master degree from Hefei University of Technology in 1991. His research interest covers electrical and electronic, the internet of things, data mining, and software development about mobile phone

    Corresponding author: TANG Hao   Professor at the School of Electrical Engineering and Automation, Hefei University of Technology. He received his Ph. D. degree from University of Science and Technology of China in 2002. His research interest covers discrete event dynamic system, stochastic decision and optimization theory, intelligent optimization and control method. Corresponding author of this paper
  • 摘要: 针对混合属性数据集聚类精度低的问题,本文提出一种基于改进距离度量的半监督模糊均值聚类(Fuzzy C-means,FCM)算法.首先,在数据集中针对类别属性进行预处理,并设置相应的相异度阈值;将传统聚类距离度量与改进的Jaccard距离度量结合,确定混合属性数据集的距离度量函数;最后,将所得距离度量函数与传统半监督FCM算法相结合,并在滚动轴承的不同复合故障数据的特征集中进行聚类.实验表明,该算法能在含无序属性的混合属性数据集的聚类中取得更好的聚类效果.
    1)  本文责任编委 刘艳军
  • 图  1  复合振动信号$EMD$分解

    Fig.  1  The EMD decomposition of complex vibration signals

    图  2  有标签数据预聚类

    Fig.  2  Pre-clustering of the label data

    图  3  重聚类结果

    Fig.  3  Re-clustering result

    图  4  重聚类结果柱状统计图

    Fig.  4  Bar chart of re-clustering result

    图  5  重聚类结果散点图

    Fig.  5  Scatter diagram of re-clustering result

    图  6  重聚类结果柱状统计图

    Fig.  6  Bar chart of re-clustering result

    图  7  改进FCM自适应阈值调整后重聚类结果

    Fig.  7  Re-clustering result by improved FCM algorithm after adaptive threshold

    图  8  改进FCM自适应阈值调整后重聚类结果柱状统计图

    Fig.  8  Bar chart of re-clustering result by improved FCM algorithm after adaptive threshold

    表  1  轴承各部件故障特征频率(Hz)

    Table  1  Characteristic frequency of rolling bearings (Hz)

    内圈外圈保持架滚动体
    163.2107.411.9141.2
    下载: 导出CSV

    表  2  聚类精度对比表

    Table  2  Comparison table of clustering accuracy

    单故障耦合故障
    传统FCM聚类精度0.980.65
    改进FCM聚类精度1.000.87
    下载: 导出CSV

    表  3  三种算法聚类精度对比表

    Table  3  Comparison table of clustering accuracy by three algorithms

    传统FCMK-prototypes改进FCM
    聚类精度0.7860.8420.902
    下载: 导出CSV

    表  4  不同$\varepsilon$值下聚类精度对比表

    Table  4  Comparison table of clustering accuracy by different $\varepsilon$

    $\varepsilon$0.090.100.110.120.130.14
    聚类精度0.7960.8680.8980.9020.880.822
    下载: 导出CSV
  • [1] 徐明亮, 王士同, 杭文龙.一种基于同类约束的半监督近邻反射传播聚类方法.自动化学报, 2016, 42(2):255-269 http://www.aas.net.cn/CN/abstract/abstract18815.shtml

    Xu Ming-Liang, Wang Shi-Tong, Hang Wen-Long. A semi-supervised affinity propagation clustering method with homogeneity constraint. Acta Automatica Sinica, 2016, 42(2):255-269 http://www.aas.net.cn/CN/abstract/abstract18815.shtml
    [2] 赵慧珍, 刘付显, 李龙跃. K-近邻估计协同系数的协同模糊C均值算法.计算机工程与应用, 2016, 52(19):19-24 doi: 10.3778/j.issn.1002-8331.1601-0312

    Zhao Hui-Zhen, Liu Fu-Xian, Li Long-Yue. Novel collaboration fuzzy C-means algorithm with K-nearest neighbor method determined Collaboration Coefficient. Computer Engineering and Applications, 2016, 52(19):19-24 doi: 10.3778/j.issn.1002-8331.1601-0312
    [3] Huang Z X. Clustering large data sets with mixed numeric and categorical values. In: Proceedings of the 1st Pacific-Asia Conference on Knowledge Discovery and Data Mining. Singapore, Singapore: PAKDD, 1997. 21-34
    [4] 陈晋音, 何辉豪.基于密度和混合距离度量方法的混合属性数据聚类研究.控制理论与应用, 2015, 32(8):993-1002 http://d.old.wanfangdata.com.cn/Periodical/kzllyyy201508001

    Chen Jin-Yin, He Hui-Hao. Density-based clustering algorithm for numerical and categorical data with mixed distance measure methods. Control Theory and Applications, 2015, 32(8):993-1002 http://d.old.wanfangdata.com.cn/Periodical/kzllyyy201508001
    [5] 黄德才, 李晓畅.基于相对密度的混合属性数据增量聚类算法.控制与决策, 2013, 28(6):815-822 http://d.old.wanfangdata.com.cn/Periodical/kzyjc201306005

    Huang De-Cai, Li Xiao-Chang. Incremental relative density-based clustering algorithm for mixture data sets. Control and Decision, 2013, 28(6):815-822 http://d.old.wanfangdata.com.cn/Periodical/kzyjc201306005
    [6] 陈晋音, 何辉豪.基于密度的聚类中心自动确定的混合属性数据聚类算法研究.自动化学报, 2015, 41(10):1798-1813 http://www.aas.net.cn/CN/abstract/abstract18754.shtml

    Chen Jin-Yin, He Hui-Hao. Research on density-based clustering algorithm for mixed data with determine cluster centers automatically. Acta Automatica Sinica, 2015, 41(10):1798-1813 http://www.aas.net.cn/CN/abstract/abstract18754.shtml
    [7] 陈新泉.面向混合属性数据集的双重聚类方法.计算机工程与科学, 2013, 35(2):127-132 doi: 10.3969/j.issn.1007-130X.2013.02.022

    Chen Xin-Quan. Dual clustering method of mixed data set. Computer Engineering and Science, 2013, 35(2):127-132 doi: 10.3969/j.issn.1007-130X.2013.02.022
    [8] Gardner A, Kanno J, Duncan C A, Selmic R. Measuring distance between unordered sets of different sizes. In: Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Columbus, OH, USA: IEEE, 2014. 137-143
    [9] 李城梁, 马芸, 张锐, 魏伟.基于半监督谱核聚类的转子系统故障诊断.振动、测试与诊断, 2016, 36(3):562-567 http://d.old.wanfangdata.com.cn/Periodical/zdcsyzd201603026

    Li Cheng-Liang, Ma Yun, Zhang Rui, Wei Wei. Rotor system fault diagnosis based on semi-supervised spectrum kernel clustering. Journal of Vibration, Measurement and Diagnosis, 2016, 36(3):562-567 http://d.old.wanfangdata.com.cn/Periodical/zdcsyzd201603026
    [10] 嵇威华, 吕国芳.基于广义Jaccard系数处理冲突证据方法.控制工程, 2015, 22(1):98-101 http://d.old.wanfangdata.com.cn/Periodical/jczdh201501017

    Ji Wei-Hua, Lv Guo-Fang. Conflicting evidence combination method based on generalized Jaccard coefficient. Control Engineering of China, 2015, 22(1):98-101 http://d.old.wanfangdata.com.cn/Periodical/jczdh201501017
    [11] 周晨曦, 梁循, 齐金山.基于约束动态更新的半监督层次聚类算法.自动化学报, 2015, 41(7):1253-1263 http://www.aas.net.cn/CN/abstract/abstract18699.shtml

    Zhou Chen-Xi, Liang Xun, Qi Jin-Shan. A semi-supervised agglomerative hierarchical clustering method based on dynamically updating constraints. Acta Automatica Sinica, 2015, 41(7):1253-1263 http://www.aas.net.cn/CN/abstract/abstract18699.shtml
    [12] 袁杰, 王福利, 王姝, 赵露平.基于D-S融合的混合专家知识系统故障诊断方法.自动化学报, 2017, 43(9):1580-1587 http://www.aas.net.cn/CN/abstract/abstract19134.shtml

    Yuan Jie, Wang Fu-Li, Wang Shu, Zhao Lu-Ping. A fault diagnosis approach by D-S fusion theory and hybrid expert knowledge system. Acta Automatica Sinica, 2017, 43(9):1580-1587 http://www.aas.net.cn/CN/abstract/abstract19134.shtml
    [13] 张超, 陈建军, 郭迅.基于第2代小波和EMMD的转子系统复合故障诊断.振动、测试与诊断, 2011, 31(1):98-103 doi: 10.3969/j.issn.1004-6801.2011.01.022

    Zhang Chao, Chen Jian-Jun, Guo Xun. Complex fault diagnosis for rotor systems using the second generation wavelet and extremum field mean mode decomposition. Journal of Vibration, Measurement and Diagnosis, 2011, 31(1):98-103 doi: 10.3969/j.issn.1004-6801.2011.01.022
    [14] Hao H, Wang H L, Rehman N U. A joint framework for multivariate signal denoising using multivariate empirical mode decomposition. Signal Processing, 2017, 135:263-273 doi: 10.1016/j.sigpro.2017.01.022
  • 加载中
图(8) / 表(4)
计量
  • 文章访问数:  2059
  • HTML全文浏览量:  198
  • PDF下载量:  526
  • 被引次数: 0
出版历程
  • 收稿日期:  2017-09-06
  • 录用日期:  2017-12-06
  • 刊出日期:  2018-12-20

目录

    /

    返回文章
    返回