欢迎访问机床与液压官方网站!

咨询热线:020-32385312 32385313 RSS EMAIL-ALERT
基于优化CNN的航空液压管路卡箍故障诊断
作者:
作者单位:

作者简介:

通讯作者:

中图分类号:

基金项目:

国家自然科学基金(51775257)


Fault Diagnosis of Air Hydraulic Pipe Clamp Based on Optimize CNN
Author:
Affiliation:

Fund Project:

  • 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:

    针对航空发动机液压卡箍-管路系统具有高度复杂性,导致卡箍振动信号存在非线性、非平稳性,从而难以提取出卡箍故障状态有效信息的问题,提出一种基于优化变分模态分解(VMD)与卷积神经网络(CNN)的卡箍智能故障诊断方法。基于优化的VMD将液压管路系统-卡箍振动信号分解成一系列固有模态函数;将含有卡箍故障信号明显的IMF输入到卷积神经网络训练模型,采用CNN进行自主特征学习和模式识别。并将该方法应用于实例中,结果表明:该方法不仅能有效地对信号进行分解,同时对不同类型的卡箍故障可达到精准识别和故障诊断。

    Abstract:

    Due to the high complexity of the aero-engine hydraulic clamp-piping system,the vibration signal of the clamp is nonlinear and nonstationary,which makes it difficult to extract effective information about the fault status of the clamp.An intelligent fault diagnosis method based on optimized variational mode decomposition (VMD) and convolutional neural network (CNN) was proposed.Based on optimized variational mode decomposition parameter factor\[k,α\],the vibration signal of the clamp in the hydraulic pipeline system was decomposed into a series of natural mode functions with the optimized VMD.IMF with obvious clamp fault signal was input into the training model of CNN,and CNN was used for autonomous feature learning and pattern recognition.Finally,the method was applied to an instance.The results show that this method can be used to not only decompose the signal effectively,but also accurately identify and diagnose different types of clamp faults.

    参考文献
    相似文献
    引证文献
引用本文

窦金鑫,薛政坤,于晓光,范玉鑫,刘忠鑫,杨同光.基于优化CNN的航空液压管路卡箍故障诊断[J].机床与液压,2020,48(16):188-194.
DOU Jinxin, XUE Zhengkun, YU Xiaoguang, FAN Yuxin, LIU Zhongxin, YANG Tongguang. Fault Diagnosis of Air Hydraulic Pipe Clamp Based on Optimize CNN[J]. Machine Tool & Hydraulics,2020,48(16):188-194

复制
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:
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期: 2020-09-08
  • 出版日期: