多工况直升机附件齿轮箱振动故障诊断
Vibration Fault Diagnosis of Helicopter Accessory Gearbox Under Multi-operating Conditions
万安平 1龚志鹏 1王景霖 2单添敏 2何家波1
作者信息
- 1. 浙大城市学院机械系 杭州,310015
- 2. 故障诊断与健康管理技术航空科技重点实验室 上海,201601
- 折叠
摘要
针对直升机附件齿轮箱在有限多工况条件下故障特征提取难度大、识别准确率低等问题,提出一种结合变分模态分解(variational mode decomposition,简称VMD)与多尺度卷积神经网络(multi-scale convolutional neural netwo,简称MCNN)的故障诊断方法.首先,对直升机附件齿轮箱进行地面实验和信号采集,对原始信号进行滤波、降噪等预处理;其次,利用VMD将信号分解为若干个固有模态(intrinsic mode functions,简称IMF),依据齿轮副频率特性对分解模态进行重构与归一化,增强微弱的高频故障特征;最后,将重构信号的每个分量视作不同尺度,经多尺度卷积神经网络进行多尺度特征提取并融合,由指数归一化分类器给出识别的故障类别.实验结果表明,所提方法能够有效增强信号故障特征,挖掘多工况条件下信号的差异性与同一性,在直升机附件齿轮箱振动故障诊断中平均准确率为97.25%.
Abstract
Aiming at the problems of difficulty in fault feature extraction and low recognition accuracy of heli-copter accessory gearbox under limited variable working conditions,a fault diagnosis method is proposed com-bining variational mode decomposition(VMD)and multi-scale convolution neural network(MCNN).Firstly,the helicopter accessory gearbox is tested on the ground and sampled,and the original signal is preprocessed by filtering and noise reduction.Secondly,the VMD decomposition signal is used as several intrinsic mode func-tions(IMF)to reconstruct and normalize the decomposition modes according to the frequency characteristics of the gear meshing ground,so as to enhance the weak high-frequency fault characteristics.Finally,each compo-nent of the reconstructed signal is regarded as a different scale,and multi-scale features are extracted and fused by MCNN.The identified fault category is given by softmax classifier.The test results show that the proposed method can effectively enhance the signal fault characteristics,excavate the difference and identity of signals un-der multiple working conditions.In the vibration fault diagnosis of helicopter accessory gearbox,the average ac-curacy rate is 97.25%.
关键词
变分模态分解/多尺度卷积网络/振动故障诊断/附件齿轮箱Key words
variational modal decomposition/multi-scale convolution network/vibration fault diagnosis/acces-sory gear box引用本文复制引用
基金项目
航空科学基金(20183333001)
国家自然科学基金(52372420)
中国博士后科学基金特别资助项目(2018T110587)
浙大城市学院科研培育基金(J-202220)
出版年
2024
NETL
NSTL
维普
万方数据