基于Fisher Score 与最大互信息系数的齿轮箱故障特征选择方法
作者:
作者单位:

1.重庆交通大学信息科学与工程学院;2.重庆微标科技股份有限公司

作者简介:

通讯作者:

中图分类号:

TP391.4

基金项目:

国家自然科学基金项目(61703063、61663008 和61573076);重庆市技术创新与应用专项重点项目 (cstc2019jscx-mbdxX0015);重庆市教委重点项目(KJZD-K20190070);重庆市教委科学技术研究项 目(KJZD-K201800701);重庆市研究生科研创新项目基金(CYS19232);桥梁工程结构动力学国家重点实验室开放基金(2019-01)


Fault Feature Selection Method of Gearbox Based on Fisher Score and Maximum Mutual Information Coefficient
Author:
Affiliation:

1.Chongqing Jiaotong University;2.Chongqing Micro Standard Technology Co., Ltd,

Fund Project:

National Natural Science Foundation of China (61703063, 61663008 and 61573076); Chongqing Special Key Project of Technology Innovation and Application (cstc2019jscx-mbdxX0015); Key Project of Chongqing Education Commission (KJZD-K20190070); Science and Technology Research Project of Chongqing Education Commission (KJZD-K201800701);Chongqing Postgraduate Research and Innovation Project Fund (CYS19232); Open Fund of State Key Laboratory of Structural Dynamics of Bridge Engineering (2019-01)

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    针对工业环境中齿轮箱多故障特征难以选择的问题,结合Fisher Score与最大信息系数(MIC)构建了一种新的故障特征优化选择方法。首先,考虑到多故障特征分布不均匀和重叠性问题,采用Fisher Score计算方法构建了特征指标重要度排序规则;之后,在考虑冗余特征对有效特征表征的影响基础上,利用最大信息系数构建了特征间关联性评价方法,对冗余特征实现更新排序;在此基础上,以分类准确率为判断依据,基于支持向量机理论(SVM)对排序模型进行修正,建立基于Fisher Score与最大信息系数的故障特征优化选择方法;最后,利用UCI标准数据集和实验仿真的齿轮箱故障数据进行实验,验证算法的有效性和工程实用性,仿真实验对比分析表明,与传统的mRMR、reliefF方法相比,本文所提出的方法特征子集数量适中,准确率更高。

    Abstract:

    Aiming at the problem that it is difficult to select multiple fault features of gearboxes in industrial environments, a new fault feature optimization selection method combining Fisher Score and maximum information coefficient (MIC) is proposed in this paper. First, considering about uneven distribution and overlapping of multi-fault features, the Fisher Score calculation method is used to construct the ranking rules of the importance of the feature indicators. Second, based on the impact of redundant features on the effective feature representation, the maximum information coefficient is used to update and rank redundant features. Then, taking classification accuracy as the judgement basis, by using support vector machine theory (SVM), a fault feature optimization selection method combining Fisher Score and maximum information coefficient was established. Finally, the UCI standard data set and gear failure simulation data set are used to verify the effectiveness and engineering practicability of this algorithm. Comparative analysis of simulation experiments shows that compared with the traditional mRMR and reliefF methods,the number of feature subsets proposed by this method is moderate and the accuracy is higher.

    参考文献
    相似文献
    引证文献
引用本文
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2019-12-19
  • 最后修改日期:2021-02-22
  • 录用日期:2020-06-12
  • 在线发布日期:
  • 出版日期: