• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Mathematical Problems in Engineering >Bispectrum Texture Feature Manifold for Feature Extraction in Rolling Bear Fault Diagnosis
【24h】

Bispectrum Texture Feature Manifold for Feature Extraction in Rolling Bear Fault Diagnosis

机译：双谱纹理特征流形用于滚动轴承故障诊断中的特征提取

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Effectively classify the fault types and the degradation degree of a rolling bearing is an important basis for accurate malfunction detection. A novel feature extract method - bispectrum image texture features manifold (BTM) of the rolling bearing vibration signal is proposed in this paper. The BTM method is realized by three main steps: bispectrum image analysis, texture feature construction and manifold feature dimensionality reduction. In this method, bispectrum analysis is employed to convert the mass vibration signals into bispectrum contour map, the typical texture features were extracted from the contour map by gray level co-occurrence matrix (GLCM), then the manifold dimensionality reduction method liner local tangent space alignment (LLTSA) is used to remove redundant information and reduce the dimension from the extracted texture features and obtain more meaningful low-dimensional information. Furthermore, the low-dimensional texture features were identified by support vector machine (SVM) which was optimized by genetic optimization algorithm (GA). The validity of BTM is confirmed by rolling bear experiments, the result show that the proposed feature extraction method can accurately distinguish different fault types and have a good performance to classify the degradation degree of inner race fault, outer race fault and rolling ball fault.
机译：有效地分类滚动轴承的故障类型和退化程度是准确检测故障的重要基础。提出了一种新颖的特征提取方法-滚动轴承振动信号的双谱图像纹理特征流形（BTM）。 BTM方法通过三个主要步骤实现：双光谱图像分析，纹理特征构建和流形特征维降。该方法利用双谱分析将质量振动信号转换为双谱等值线图，利用灰度共生矩阵（GLCM）从等值线图上提取出典型的纹理特征，然后利用流形降维方法进行局部切线空间化。对齐（LLTSA）用于删除冗余信息并从提取的纹理特征中缩小尺寸，并获得更有意义的低尺寸信息。此外，通过支持向量机（SVM）识别低维纹理特征，并通过遗传优化算法（GA）对其进行优化。通过滚动轴承实验验证了BTM的有效性，结果表明所提出的特征提取方法能够准确地区分不同的故障类型，并对内圈故障，外圈故障和滚动球故障的退化程度进行分类具有良好的性能。

著录项

  • 来源
    《Mathematical Problems in Engineering》 |2019年第5期|3805729.1-3805729.11|共11页
  • 作者

    Wang Fei; Fang Liqing;

  • 作者单位

    Army Engn Univ, Dept Artillery Engn, Shijiazhuang Campus,Heping West Rd 97, Shijiazhuang 050003, Hebei, Peoples R China;

    Army Engn Univ, Dept Artillery Engn, Shijiazhuang Campus,Heping West Rd 97, Shijiazhuang 050003, Hebei, Peoples R China;

  • 收录信息
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号