中国机械工程 ›› 2016, Vol. 27 ›› Issue (02): 160-167.

• 机械基础工程 • 上一篇    下一篇

旋转机械运行稳定性劣化的高阶累积量特征提取方法

蒋章雷;徐小力   

  1. 北京信息科技大学现代测控技术教育部重点实验室,北京,100192
  • 出版日期:2016-01-25 发布日期:2016-12-15
  • 基金资助:
    国家自然科学基金资助项目(51275052);北京市自然科学基金资助重点项目(3131002);北京市教委科研计划资助重点项目(KZ201311232036);现代测控技术教育部重点实验室开放课题资助项目(KF20141123202)  

Feature Extraction Method Based on High Order Cumulant Faced to Running Stability  Deterioration  of  Rotating  Machinery

Jiang Zhanglei;Xu Xiaoli   

  1. Key Laboratory of  Modern Measurement  &  Control  Technology,Ministry  of  Education,Beijing Information Science and Technology University,Beijing,100192
  • Online:2016-01-25 Published:2016-12-15
  • Supported by:

摘要:

研究了基于高阶累积量理论的1.5维谱、四阶累积量对角切片谱方法在提取运行稳定性劣化特征方面的性能;从敏感性、趋势性、差异性、一致性角度,讨论了两种特征提取方法的趋势预测适用性。基于构建的转子实验台进行转子系统多种劣化类型下不同劣化程度状态的特征提取实验,检验两种方法作为特征提取手段的性能,解决了风电机组传动系统运行稳定性劣化的状态诊断、劣化趋势预测中特征提取方法的选择缺少理论依据的问题。

关键词: 风电机组, 稳定性劣化, 1.5维谱, 四阶累积量对角切片谱, 特征提取

Abstract:

This paper focused  on  feature extraction method based on high order cumulant like 1.5-dimension spectrum and fourth-order cumulant diagonal slice spectrum,and also its  performance on running stability deterioration.Experimental data of varying degrees of deterioration under various types of deterioration were collected,and  sensitivity  property,trend property,difference property,and consistency property were proposed to evaluate  different methods.It can provide theoretical basis for selecting feature extraction method for fault diagnosis  and trend prediction of running stability deterioration.

Key words: wind , turbine;stability deterioration;1.5-dimension spectrum;fourth-order , cumulant diagonal slice spectrum;feature extraction

中图分类号: