期刊文献+

基于OHF Elman-AdaBoost算法的滚动轴承故障多时期诊断方法 被引量:4

Multi-period fault diagnosis of rolling bearings based on the OHF Elman-AdaBoost algorithm
下载PDF
导出
摘要 针对随机噪声下滚动轴承多时期(初期、中期、晚期)故障诊断需求,提出OHF Elman-AdaBoost(output hidden feedback Elman-adaptive boosting)算法,以实现滚动轴承的精确故障诊断。采用集合经验模态分解(ensemble empirical mode decomposition,EEMD)对原始信号进行分解、降噪、信号重构。设计OHF Elman方法在Elman神经网络的基础上增加输出层对隐含层的反馈,提高了其对动态数据的记忆功能。选择OHF Elman神经网络作为弱回归器,结合AdaBoost算法集成出一种新的强回归器:OHF Elman-AdaBoost算法。实验结果表明,该算法不仅对滚动轴承不同故障时期具有很好的诊断效果,而且提高了对全样本数据的诊断准确度,为滚动轴承故障诊断提供了新型工具和有效方案。 In order to meet the needs of the multi-period fault diagnosis of rolling bearings under random noise,the OHF Elman-AdaBoost(output hidden feedback Elman-adaptive boosting)algorithm was proposed to achieve the accurate fault diagnosis of rolling bearings.The original signal was decomposed,denoised and reconstructed by the ensemble empirical mode decomposition(EEMD).The OHF Elman neural network improved the memory function for its dynamic data by adding feedback from the output layer to the hidden layer based on the Elman neural network.Then,a strong regressor(OHF Elman-AdaBoost algorithm)was integrated by selecting the OHF Elman neural network as the weak regressor and combining with the AdaBoost algorithm.The experimental results show that the OHF Elman-AdaBoost algorithm not only has a good diagnostic effect on different periods of rolling bearing faults,but also improves the diagnostic accuracy of the full sample data,providing a new tool and effective solution for the fault diagnosis.
作者 卓鹏程 夏唐斌 郑美妹 郑宇 奚立峰 ZHUO Pengcheng;XIA Tangbin;ZHENG Meimei;ZHENG Yu;XI Lifeng(School of Mechanical Engineering,Shanghai Jiao Tong University,Shanghai 200240,China;State Key Laboratory of Mechanical System and Vibration,Shanghai Jiao Tong University,Shanghai 200240,China)
出处 《振动与冲击》 EI CSCD 北大核心 2021年第6期71-78,共8页 Journal of Vibration and Shock
基金 国家自然科学基金(51875359,51535007) 机械系统与振动国家重点实验室重点课题(MSVZD201909)。
关键词 滚动轴承 OHF Elman-AdaBoost 神经网络 集合经验模态分解(EEMD) 故障多时期诊断 rolling bearing OHF Elman-AdaBoost neural network ensemble empirical mode decomposition(EEMD) multi-period fault diagnosis
  • 相关文献

参考文献1

二级参考文献4

共引文献285

同被引文献34

引证文献4

二级引证文献2

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部