机电工程技术2025，Vol.54Issue(7)：23-27,39,6.DOI:10.3969/j.issn.1009-9492.2025.07.005
基于WMPE-NPE的风电机组轴承在线早期故障预警与诊断方法
Online Early Fault Warning and Diagnosis Method for Wind Turbine Bearings Based on WMPE-NPE
摘要
Abstract
In order to solve the problem of difficult extraction of early fault features and low diagnostic accuracy in the online operation of wind turbine bearings,the online early fault warning and diagnosis method of WMPE-NPE of wind turbine bearing which considering global-local feature extraction is proposed.This method first uses multi-scale weighted permutation entropy(MWPE)to comprehensively extract different scale features from real-time online monitoring vibration signals of bearings,fully mining the fault information existing in the high and low frequencies of vibration signals.Secondly,the neighborhood preserving embedded method(NPE)is used to reduce the dimensionality of multi-scale features and preserve local structural information in multi-scale information,thereby obtaining global-local structural features that can reflect key degradation indicators of wind turbine bearings.Finally,the early fault samples detected based on key indicators are subjected to variational mode decomposition(VMD)decomposition for noise reduction,followed by envelope spectrum analysis to achieve early fault diagnosis.The results indicate that the proposed method performs better than the diagnostic methods of RMS,WPE,MWPE+PCA,and direct envelope spectroscopy.关键词
轴承/MWPE/NPE/VMD/早期故障预警与诊断Key words
bearings/MWPE/NPE/VMD/early fault warning and diagnosis分类
机械工程引用本文复制引用
陈鹏,张亚洲..基于WMPE-NPE的风电机组轴承在线早期故障预警与诊断方法[J].机电工程技术,2025,54(7):23-27,39,6.基金项目
甘肃省陇原青年创新人才(团队)项目(310100296012) （团队）
