基于粒子群算法优化的独立分量分析算法

李良敏,任景岩

振动与冲击 ›› 2015, Vol. 34 ›› Issue (8) : 7-11.

PDF(2376 KB)
PDF(2376 KB)
振动与冲击 ›› 2015, Vol. 34 ›› Issue (8) : 7-11.
论文

基于粒子群算法优化的独立分量分析算法

  • 李良敏,任景岩
作者信息 +

An Universal Particle Swarm Optimized Independent Component Analysis Algorithm

  • Li Liangmin, Ren Jingyan
Author information +
文章历史 +

摘要

通过两组模拟信号对三种主流独立分量分析算法—JADE、FastICA、扩展Infomax算法的性能进行了对比分析,结果表明三种算法均无法完全分离超高斯源与亚高斯源形成的混合信号,FastICA算法对能量强弱差别大的混合信号失效。基于这一现象,提出了一种新的独立分量分析算法,以粒子群算法为优化工具,以分离矩阵为优化变量,最小化分离信号联合概率与边缘概率乘积的差值,并给出了具体的计算流程。仿真实验结果表明,该算法的性能显著优于上述三种独立分量分析算法。同时,新提出算法实施过程中不需要任何先验知识,相比其他三种ICA算法,更适合解决工程实际问题。最后,将该算法应用于对滚动轴承实验台实测信号的处理,通过对分离信号的分析实现了对滚动轴承故障类型的准确识别,进一步证明了算法的有效性。

Abstract

Two sets of synthetic signals are created to test the separation ability of three popular independent component analysis algorithms—JADE, FastICA, and extended-Infomax. The conclusion is drawn that the above independent component analysis can’t recover source signals from the mixtures of super-Gaussian and sub-Gaussian precisely, and FastICA fails in solving the separation problem of strong sources mixed with weak sources. A particle swarm optimized independent component analysis algorithm, which chooses the difference between joint probabilities and products of marginal probabilities of separated signals as the objective function, is proposed. The implementation procedure is described in detail. Simulation tests show that, compared with the above three independent component analysis algorithms, the proposed algorithm performs the best. Furthermore, the implementation of the proposed algorithm relies on no prior knowledge, thus more suits for solving practical engineering problem. Finally, the proposed algorithm is used to process sound signals sampled from a rolling bearing test rig. Analysis of the separated signal reveals the cause of bearing failure, indicating the validity of the proposed algorithm.   

 

关键词

独立分量分析 / FastICA / JADE / 扩展Infomax算法 / 粒子群算法 / 滚动轴承

Key words

Independent component analysis / FastICA / JADE / extended-Infomax / particle swarm optimization / rolling bearing

引用本文

导出引用
李良敏,任景岩. 基于粒子群算法优化的独立分量分析算法[J]. 振动与冲击, 2015, 34(8): 7-11
Li Liangmin, Ren Jingyan. An Universal Particle Swarm Optimized Independent Component Analysis Algorithm[J]. Journal of Vibration and Shock, 2015, 34(8): 7-11

参考文献

[1] Comon, P and Jutten C. Handbook of Blind Source Separation, Independent Component Analysis and Applications[M]. Oxford UK: Academic Press, 2010: 155-178.
[2] 李欣,梅德庆,陈子辰.基于ICA的镗削过程颤振征兆信号分离方法研究[J].振动与冲击,2013,32(9):5-9.
LI Xin, MEI De-qing, CHEN Zi-chen. ICA based separation of chatter symptom signals for precision hole boring processing[J].Journal of Vibration and Shock,2013,32(9):5-9.
[3] Lei X., Caldes Sosa P., Yao D.. EGG/fMRI fusion based on independent component analysis: integration of data-driven and model-driven methods[J]. Journal of Integrative Neuroscience, 2012, 11(3): 313-337.
[4] Jie Yu, Jingyan Chen, Mudassir M. Rashid. Multiway independent component analysis mixture model and mutual information based fault detection and diagnosis approach of multiphase batch processes[J]. AiChE Journal, 2013, 59(8): 2761–2779.
[5] Hyvarien A. Fast and roubust fixed-point algorithms for independent component analysis[J]. IEEE Transactions on Neural Networks, 1999, 10(3): 626-634.
[6] Nima Reyhani, Jarkko Ylipaavalniemi, Ricardo Vigario, et al. Consistency and asymptotic normality of FastICA and bootstrap FastICA[J]. Signal processing, 2012, 92(8):1767-1778.
[7] J.F. Cardoso. High-order contrasts for independent component analysis[J]. Neural Computation, 1999, 11(1):157-192.
[8] 虞海强,王平.基于JADE算法的变压器振动信号分离的研究[J].现代电力,2012,29(1):42-46.
YU Hai-qiang, WANG Ping. Research on vibration signal separation of transformer based on JADE algorithm[J]. Modern Electric Power, 2012, 29(1): 42-46.
[9] T. Lee, M. Girolami and T. Sejnowski. Independent component analysis using an extended infomax algorithm for mixed sub-Gaussian and super-Gaussian sources[J]. Neural Computation, 1999, 11(2): 417-441.
[10] 徐明彪,朱维彰.关于信号盲分离分离效果评判指标的分析[J].杭州电子工业学院学报,2002,2(3):63-66.
XU Ming-biao, ZHU Wei-zhang. Analyses of the index used for evaluating the effect of BSS[J]. Journal of Hangzhou Institute of Electronic Engineering, 2002, 2(3): 63-66.
[11] 张贤达,保铮. 盲信号分离[J]. 电子学报,2001,29(12):1766-1771.
ZHANG Xian-da, BAO Zheng. Blind source separation[J]. Acta Electornica Sinica, 2001, 29(12): 1766-1771.
[12] 李士勇,李研. 智能优化算法与应用[M]. 哈尔滨:哈尔滨工业大学出版社,2012.
LI Shi-yong, LI Yan. Intelligent optimization algorithm theory and applications[M]. Harbin: Harbin Institute of Technology Press, 2012.
[13] 莫思敏,曾建潮,徐卫滨. 具有自组织种群结构的微粒群算法[J]. 系统仿真学报,2013,25(3):445-450.
MO Si-min, ZENG Jian-chao, XU Wei-bin. Particle swarm optimization based on self-organization topology[J]. Journal of System Simulation, 2013, 25(3): 445-450.
[14] 肖云魁. 汽车故障诊断学[M].北京:北京理工大学出版社,2006: 25.
XIAO Yun-kui. Research on vehicle fault diagnosis[M]. Beijing: Beijing Institute of Technology Press, 2006:25.

PDF(2376 KB)

Accesses

Citation

Detail

段落导航
相关文章

/