• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>International Conference on Uncertainty in Structural Dynamics;International Conference on Noise and Vibration Engineering >Explicit-duration hidden Markov model inference and application to bearing fault diagnosis
【24h】

Explicit-duration hidden Markov model inference and application to bearing fault diagnosis

机译：明确持续时间隐马尔可夫模型推理和应用于承载故障诊断

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

The components in the vibration signals of faulty bearings mainly comprise damped and repetitive impulses on the top of stationary components, which can be modeled as a non-stationary signals whose statistical properties switch between two states. The duration of staying in each state turns out to be important information for presence of a fault and for identifying it. This work introduces a probabilistic model for vibration signals of faulty bearings that is able to capture the aforementioned characteristics. It is based on the explicit-duration Hidden Markov model (EDHMM), and joint force with short-time Fourier transform (STFT). The STFT coefficients are modeled as the explicit-duration Hidden Markov chain, and then extracting the duration time in the different states of the bearing vibration signal. A complete diagnostic framework is given out in this paper, including modeling, inference, estimation and application. Firstly, specific definition and formulation of the explicit-duration Hidden Markov process are addressed. Algorithm for estimating the EDHMM parameters are then introduced, and how the latter can be used for diagnosis is eventually illustrated. Finally, its effectiveness is validated with synthetic and experimental data.
机译：故障轴承的振动信号中的组件主要包括静止部件顶部的阻尼和重复脉冲，其可以作为非静止信号建模，其统计特性在两个状态之间切换。保持在每个状态的持续时间使其成为存在故障和识别它的重要信息。这项工作介绍了能够捕获上述特性的故障轴承的振动信号的概率模型。它基于显式持续时间隐马尔可夫模型（EDHMM），以及短时傅里叶变换（STFT）的联合力。 STFT系数被建模为显式隐藏马尔可夫链，然后在轴承振动信号的不同状态下提取持续时间。本文发出完整的诊断框架，包括建模，推理，估计和应用。首先，解决了明确持续时间隐马尔可夫进程的具体定义和制定。然后介绍用于估计EDHMM参数的算法，最终说明后者可以用于诊断。最后，其有效性与合成和实验数据验证。

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