• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>International Conference on Mechanical Engineering and Mechanics vol.2; 20051026-28; Nanjing(CN) >Noise Reduction Method for Nonlinear Time Series Based on Principal Manifold Learning and Its Application to Fault Diagnosis
【24h】

Noise Reduction Method for Nonlinear Time Series Based on Principal Manifold Learning and Its Application to Fault Diagnosis

机译：基于主流形学习的非线性时间序列降噪方法及其在故障诊断中的应用

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

A new noise reduction method for nonlinear time series based on principal manifold learning is proposed. After embedding the time series into a high reconstructed phase space, the principal manifold of the dynamical system attractors, which is represented in a low dimensional, global and orthogonal coordinate system, is obtained by manifold learning method such as Local Tangent Space Alignment algorithm. The final denoised result in the form of one dimensional time series is achieved after averaging of the data in phase space which are regenerated according to the principal manifold. Being different from the current noise reduction methods for nonlinear time series, for example, local projective method, the method based on principal manifold learning emphasizes more on the holistic structure of nonlinear deterministic characters hidden in the time series. Experimental results show that the proposed method has a better performance of removing Gaussian white noise in chaotic time series, compared with local projective noise reduction method. Also the method was applied to the analysis of a vibration signal sampled from a defective gear box with a broken tooth. The impact features, which are difficult to be identified from noisy data directly, can be extracted successfully after noise reduction.
机译：提出了一种基于主流形学习的非线性时间序列降噪新方法。将时间序列嵌入到高重构相空间中之后，通过诸如局部切线空间对准算法的流形学习方法，获得了以低维，全局和正交坐标系表示的动力学系统吸引子的主要流形。在一维时间序列形式的最终去噪结果是在相空间中的数据平均后获得的，这些数据是根据主流形重新生成的。与当前用于非线性时间序列的降噪方法（例如局部投影方法）不同，基于主流形学习的方法更加强调隐藏在时间序列中的非线性确定性字符的整体结构。实验结果表明，与局部投影降噪方法相比，该方法在混沌时间序列上具有更好的去除高斯白噪声的性能。该方法还用于分析从有断齿的故障齿轮箱采样的振动信号。难以直接从噪声数据中识别出的影响特征，可以在降噪后成功提取出来。

著录项

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