Blind source separation of nonlinear mixture from correlated sources based on Kernel Canonical Correlation Analysis

PDF(1830 KB)

Journal of Vibration and Shock ›› 2015, Vol. 34 ›› Issue (5) : 154-158.

Based on the deficiency in the traditional blind separation method of statistically correlated sources, a new blind separation method of nonlinear mixture from correlated sources is proposed. In the proposed method, the nonlinear problem between the data can be processed by the kernel method, and the correlated sources can be effectively separated using the correlate of source signals. The proposed method is compared with traditional blind separation method of statistically correlated sources. The simulation results show that the proposed method is obviously superior to the traditional blind separation method of statistically correlated, and the performance index of separation can be reflected. Finally the proposed method is applied to blind separation of the misalignment of rotary and rotor rub-impact, the experiment results further validate the effectiveness of the proposed method.

Author information +

History +

Abstract

Based on the deficiency in the traditional blind separation method of statistically correlated sources, a new blind separation method of nonlinear mixture from correlated sources is proposed. In the proposed method, the nonlinear problem between the data can be processed by the kernel method, and the correlated sources can be effectively separated using the correlate of source signals. The proposed method is compared with traditional blind separation method of statistically correlated sources. The simulation results show that the proposed method is obviously superior to the traditional blind separation method of statistically correlated, and the performance index of separation can be reflected. Finally the proposed method is applied to blind separation of the misalignment of rotary and rotor rub-impact, the experiment results further validate the effectiveness of the proposed method.

Key words

kernel canonical correlation analysis / blind source separation / correlated sources / nonlinear mixture

Cite this article

EndNote

Ris (Procite)

Bibtex

Download Citations

LI Zhinong, ZHANG Fen, Xiao Yaoxian . Blind source separation of nonlinear mixture from correlated sources based on Kernel Canonical Correlation Analysis[J]. Journal of Vibration and Shock, 2015, 34(5): 154-158

References

1. 周晓峰，杨世锡，甘春标. 相关机械振源的盲源分离方法[J]. 振动与冲击, 2012, 31(14): 60-63.
ZHOU Xiaofeng, YANG Shixi, GAN Chunbiao. Blind source separation of statistically correlated sources[J]. Journal of Vibration and Shock, 2012, 31(14): 60-63.
2. 谢胜利, 何昭水, 傅予力. 基于稀疏元分析的欠定混叠自适应盲分离方法[J]. 中国科学E辑, 2007, 37(8): 1086-1098.
XIE Shengli, HE Zhaoshui, FU Yuli. Self-adaptive blind source separation method of underdetermined mixtures based on sparse component analysis[J]. Science in China(Series E), 2007, 37(8): 1086-1098.
3. 张延良, 楼顺天, 张伟涛. 用于统计相关源信号的盲分离方法[J]. 西安电子科技大学学报(自然科学版), 2009, 36(3): 401-405.
ZHANG Yan liang, LOU Shuntian, ZHANG Weitao. Blind source separation method applicable to dependent sources[J]. Journal of Xidian University,
4. 王惠刚, 梁红, 李志舜. 具有任意分布时间相关源的盲分离自适应算法[J]. 西北工业大学学报, 2003, 21(6): 735-739. 2009, 36(3): 401-405.
5. 杨世锡, 焦卫东, 吴昭同. 应用JADE盲分离算法分离统计相关源[J]. 振动工程学报, 2003, 16(4): 498-501.
YANG Shixi, JIAO Weidong, WU Zhaotong. Application of JADE to separation of statistically correlated sources[J]. Journal of Vibration Engineering, 2003, 16(4): 498-501.
6. 安静. 非线性盲源分离及其应用技术研究[D]. 电子科技大学. 2012.
AN Jing. Nonlinear blind source separation and its application[D]. University of Electronic Science and Technology of China, 2012.
7. 范涛, 李志农, 岳秀廷. 基于变分贝叶斯算法和MLP网络的后非线性混合盲源分离放假研究[J]. 振动与冲击. 2010, 29(6): 21-24.
FAN Tao, LI Zhinong, YUE Xiuting. Post-nonlinear blind separation of the source signals based on variational bayesian theory and MLP[J]. Journal of Vibration and Shock, 2010, 29(6): 21-24.
8. 范涛. 基于变分贝叶斯独立分量分析的机械故障诊断方法研究[D]. 郑州大学, 2009.
FAN Tao. The mechanical fault diagnosis method based on variational Bayesian independent component analysis[D]. Zhengzhou University, 2009.
9. Zhixiong Li, Xinping Yan, Zhe Tian, Chengqing Yuan, Zhongxiao Peng, Li Li. Blind vibration component separation and nonlinear feature extraction applied to the nonstationary vibration signals for the gearbox multi-fault diagnosis[J]. Measurement, 2013，46(1): 259-271.
10. Stefan Harmling, Andreas Ziehe, etc. Kernel Feature Spaces and Nonlinear Blind Source Separation[J]. The Fifteenth Annual Neural Information Processing Systems Conference. 2002,761-768.
11. 杨杰，郑海起，田昊等. 基于独立分量分析的欠定盲源分离方法[J]. 振动与冲击. 2013，32（7）：30-33.
YANG Jie, ZHENG Hai-qi, TIAN Hao, etc. Underdetermined blind source separation method based on independent component analysis[J]. Journal of Vibration and Shock, 2013，32（7）：30-33.
12. Magnus Borga，Hans Knutsson. A Canonical Correlation Approach to Blind Source Separation[D]. Linkopings uniwersitet. 2002.
13. 李志农. 时序模型盲辨识在旋转机械故障诊断中的应用研究[D]. 浙江大学. 2002.
Li Zhinong. Application and Research on Fault Diagnosis of Rotating Machinery Using Blind Identification of Time Series Model[D]. Zhejiang University. 2002.