摘要
联合收获机滚动轴承的状态监测和故障诊断是目前保证其稳定运行的重要手段,而敏感故障特征的提取是轴承故障诊断的关键问题。为此,针对REB振动信号的非平稳、非线性特点及现有轴承故障诊断方法过于依赖监督学习算法等问题,提出了基于经验模态分解(EMD)能量熵的特征提取方法进行模型特征选取,并用于训练人工神经网络(ANN)对轴承故障进行分类。研究结果表明:提出的基于模态分解在联合收获机轴承故障诊断及分类中具有较强的可靠性,提出的健康指数(HI)可以系统表示不同的故障类型和严重程度,并可以成功地检测出联合收获机在运行过程中的故障类型,可为开发轴承自动诊断和预测系统提供参考。
The condition monitoring and fault diagnosis of the rolling bearing of the combine harvester is an important means to ensure the stable operation of agricultural machinery,and the extraction of sensitive fault features is the key issue of bearing fault diagnosis.Aiming at the non-stationary and nonlinear characteristics of REB vibration signals and the over-reliance on supervised learning algorithms in existing bearing fault diagnosis methods,a feature extraction method based on Empirical Mode Decomposition(EMD)energy entropy was proposed,and model features were selected for training.An artificial neural network(ANN)classifies bearing faults.The research results show that the modal decomposition proposed in this study has strong reliability in the fault diagnosis and classification of combine harvester bearings.The proposed health index(HI)can systematically represent different fault types and severities.The research results It shows that the failure types of the combine harvester during operation can be successfully detected based on the modal decomposition method,which provides a solid foundation for the development of automatic bearing diagnosis and prediction system.
作者
王晓燕
许栋刚
Wang Xiaoyan;Xu Donggang(Zhengzhou Vocational and Technical College,Zhengzhou 450121,China)
出处
《农机化研究》
北大核心
2023年第7期30-34,64,共6页
Journal of Agricultural Mechanization Research
基金
河南省科技厅科技攻关项目(192102210062)。
