摘要
针对通风机轴承信号非平稳和故障样本稀少的问题,提出基于完备集合经验模态分解(CEEMDAN)和极限学习机(ELM)的轴承故障诊断方法。首先,利用CEEMDAN分解故障信号得到本征模态分量(IMF);然后,使用极限学习机学习IMF能量特征;最后,将极限学习机用于故障诊断。
As the non-stationary of fan bearing signal and scarce of fault samples, bearing fault diagnosis method based on complete ensemble empirical mode decomposition with adaptive noise(CEEMDAN) and extreme learning machine(ELM) is proposed. First, CEEMDAN is used to decompose fault signal to obtain intrinsic mode component(IMF). Then, IMF energy features is learned through ELM. Finally, ELM is applied for fault diagnosis.
出处
《煤炭技术》
北大核心
2017年第8期211-213,共3页
Coal Technology
基金
江苏省重点研发计划项目(BE2016046)
江苏省煤矿电气与自动化工程实验室建设项目(2014KJZX05)