摘要
针对异步电机单一故障信号的局限性和故障特征存在较强非线性关系的特点,提出一种基于异类信息特征融合的故障诊断方法。以采集的振动信号和电流信号为原始信源,分别提取它们的时域特征和小波包熵特征,采用核主元分析对原始特征的组合进行降维融合,得到信息互补的特征量,将融合特征通过支持向量机进行模式识别。异步电机转子和轴承故障诊断实例表明,基于核主元分析的异类信息特征融合方法,可充分利用异类信源的冗余互补信息和特征数据之间的非线性关系,更全面地表征设备运行状态,相比单参数法及同类信息特征融合法具有更高的诊断精度。
Aiming at the limitation of single fault signal and strong nonlinear relation of fault features in asynchronous motor,a fault diagnosis method based on heterogeneous information feature fusion is proposed. Taking the acquired vibration signal and current signal as original signals ,the time domain features and wavelet packet entropy features are respectively extracted. Kernel principle component analysis （KPCA） is used to cant out dimensionality reduction fusion of the combi- nation of original features and obtain information complementary features. The fused features are sent to support vector ma- chines （SVM） for fault pattern recognition. The fault diagnosis examples of asynchronous motor rotor and bearing show that the heterogeneous information feature fusion method based on KPCA can make full use of redundant and complementary information from different information sources and nonlinear relation among feature data, and characterize equipment operation state more comprehensively. Compared with single parameter method and homogeneous information feature fusion method,the method proposed in this paper can obtain higher diagnosis accuracy.
出处
《仪器仪表学报》
EI
CAS
CSCD
北大核心
2013年第1期227-233,共7页
Chinese Journal of Scientific Instrument
基金
国家自然科学基金(51105138)
武器装备预研项目(813040302)
湖南省教育厅资助项目(11A034)
湖南省产学研结合创新平台(2010XK6066)
湖南省高校科技创新团队支持计划资助项目
关键词
异类信息
特征融合
异步电机
故障诊断
核主元分析
heterogeneous information
feature fusion
asynchronous motor
fault diagnosis
kernel principle component analysis （KPCA）
