摘要
独立成分分析(independent component analysis,ICA)作为盲信号分离算法中的一种有效方法,能够快速、准确地实现源信号分离和恢复。为将变压器有载分接开关(on-loadtapchanger,OLTC)振动信号从变压器油箱表面振动信号中有效地分离出来,采用ICA经典算法中的二阶盲辨识算法来分离变压器OLTC信号,并结合冲击信号的端点检测法以及互相关函数相位对准法来实现信号分离的自适应。实验结果证明,该方法能够准确、有效地从变压器油箱表面振动信号中分离出OLTC的振动信号。
As an effective manner of blind source separation, independent component analysis （ICA） can rapidly and accurately implement the separation and recovery of source signals. In order to effectively separate the vibration signal of on-load tap changer （OLTC） of power transformer from the vibration signals of tank surface of power transformer, the second order blind identification in classical ICA algorithm is adopted to separate vibration signal of OLTC, and combining with endpoint detection algorithm of impact signal and phase difference detection algorithm of cross-correlation function the adaptive signal separation is implemented. Experimental results show that the proposed method can accurately and effectively extract OLTC vibration signal from tank surface vibration signals of power transformer.
出处
《电网技术》
EI
CSCD
北大核心
2010年第11期208-213,共6页
Power System Technology
基金
国家自然科学基金项目(60875002)~~
关键词
二阶盲辨识算法
端点检测
有载分接开关
自适
应分离
变压器
独立成分分析
second order blind identification （SOBI）algorithm
endpoint detection
on-load tap changer （OLTC）
adaptive separation
transformer
independent componentanalysis （ICA）
