摘要
针对齿轮故障时振动信号复杂、特征提取困难的问题,提出采用局部特征尺度分解(Local Characteristicscale Decomposition,LCD)与双谱分析相结合的故障特征提取方法。首先,利用LCD分解法对振动信号进行分解,并结合贝叶斯信息准则(Bayesian Information Criterion,BIC)和峭度时间序列互相关系数2个指标对内禀尺度分量(Intrinsic Scale Component,ISC)进行筛选;其次,利用双谱分析法对所选取的ISC分量进行融合,提取双谱熵作为特征量;最后,运用该方法实现齿轮振动信号故障特征的提取,并通过齿轮预置故障试验验证了该方法的有效性。
Aiming at the problem that the vibration signal is complex and its feature extraction is difficult when the gear is faulty,a fault feature extraction method combining Local Characteristic-scale Decomposition(LCD)and Bispectrum analysis is proposed.Firstly,the LCD method is used to decompose the vibration signal,and the Instrinsic Scale Component (ISC)is selected by combining Bayesian Information Criterion (BIC)and sliding kurtosis time series correlation coefficients.Secondly,the bispectrum analysis method is used to fuse the selected ISC components,and extract the Bispectrum entropy as the eigenvector.Finally,this method is used to extract the fault characteristics of the gear vibration signal,and the validity of the method is verified by the gear presetting fault experiment.
作者
仝蕊
康建设
李宝晨
钟文
TONG Rui;KANG Jian-she;LI Bao-chen;ZHONG Wen(Equipment Command and Management Department,Shijiazhuang Campus of Army Engineering University,Shijiazhuang 050003,China;Scientific Research and Academies Division,Army Engineering University,Nanjing 210007,China;Troop No.95685of PLA,Kunming 650000,China)
出处
《装甲兵工程学院学报》
2018年第5期42-48,共7页
Journal of Academy of Armored Force Engineering
基金
河北省自然科学基金资助项目(E2015506012)
关键词
特征提取
局部特征尺度分解(LCD)
双谱分析
互相关系数
feature extraction
Local Characteristic-scale Decomposition (LCD )
Bispectrum analysis
correlation coefficients
