摘要
机械故障诊断是以机器学为基础的一门综合性技术,它的本质是一个机器运行状态的模式识别问题,其关键就是故障信号的特征提取与分类。近年来,为满足对机器故障进行早期检测与精确诊断的需要,非平稳、非高斯信号处理方法在机械故障诊断领域受到了广泛关注。为此,本文基于国家自然科学基金项目“局域波法及其工程应用研究”(50475155),利用非平稳、非高斯信号处理理论中的局域波法、Wigner高阶时频表示和盲源分离理论,结合模式识别与机器学习领域的研究成果,对非平稳、非高斯的机械振动信号特征提取与故障诊断问题进行了广泛而深入的研究。主要的工作如下:
1.应用局域波法对机械振动非平稳信号进行了研究。通过与小波变换和几种时频分析方法的比较,表明局域波法对于非平稳信号的分析更具有效性。实验结果表明,局域波时频分析能够清晰地表征不同故障的时变特征。由于局域波时频谱是一种二维的信号表示形式,在计算机对故障自动分类时,涉及到维数压缩的问题。为了用尽可能少的维数表示时频谱而不损失分类精度,几何矩和边缘分布可以作为时频分布的特征。在此基础上,结合人工神经网络,提出了一种基于局域波几何矩和边缘的故障诊断方法。
2.研究了基于局域波法的多分量神经网络预测模型的有效性,用于对非平稳系统时间序列进行建模。通过太阳黑子数据的仿真试验,验证了该多分量结构比对应的单一神经网络结构性能优越。最后根据该方法组成了一个自回归时间序列模型库,用于转子故障的模型诊断中。这些模型可以用做一步向前预测器,对检测和诊断信号进行比较。从预测误差提取特征,能够确定机器的状态。不同故障状态的转子振动信号用来训练和检验模型。实验数据表明,在适当训练样本长度下,这种方法用于故障诊断,可以实现故障的正确分类。
3.研究了高阶时频分布在振动冲击信号特征提取中的应用问题。在机械状态监控中,冲击信号的检测对于提取机器的状态信息是很有用的。通过Wigner高阶矩谱可以有效地对这样的非平稳、非高斯振动冲击信号进行特征提取和检测。针对高阶时频分布分析多分量信号时存在交叉项的问题,提出了一种利用局域波分解来减少Wigner高阶矩谱交叉项的方法,以仿真信号为例,验证了此方法的有效性;通过对现场测试的柴油机爆燃阶段信号的Wigner高阶矩谱分析,验证了该方法在机械故障特征提取中具有很好的应用潜力。最终表明,通过该方法可以提取有价值的关于冲击信号的时间和谱特性的量化信息。
基于局域波法和盲源分离的故障诊断方法应用研究
4.提出了一种基于局域波时频图像的盲源分离故障诊断方法。独立成分分析
(ICA)是实现盲源分离最有效的方法之一。ICA可以认为是PCA特征提取技术的推
广。ICA能够提供图像的局部特征,给出较好的图像表示。针对局域波时频图像可以表
征不同故障振动信号的特点,应用盲源分离技术对不同故障信号的局域波时频图像进行
独立分量分离,提取代表当前工况特征的投影系数矩阵,作为故障特征,利用神经网络
实现不同故障的自动分类。最后以转子的早期摩擦,基座松动,不对中故障振动信号为
例,应用该方法进行了研究,实验结果证明了该方法的可行性。
5.为了有效提取故障特征信号,需要在不同位置进行多传感器的振动信号测量。
针对多源混合的非平稳、非高斯设备故障振动信号,应用非平稳信号的盲源分离算法,
可以有效地提取各自独立的非平稳振动源,从而更加准确地进行机械故障诊断。首先,
针对不同时频分布的非平稳盲源分离算法,通过仿真信号比较了它们的分离效果。然后
以转子的复合故障为例进行了实验验证。在此基础上,提出了一种基于盲源分离的多传
感器数据融合故障诊断方法。实验结果证明该方法能够提高故障诊断的精度。
关键词:机械振动;故障诊断;局域波法;高阶时频分布;盲源分离
Mechinery diagnosis is a branch of mechanology. Its essence is the pattern recognition of machine operating condition. Its key issues are the feature extraction and classification for fault signals, hi recent years, to meet the needs of early detection and accurate diagnosis of mechanical faults, non-stationary and non-Gaussian signal processing techniques attracted more and more attention in mechanical fault feature extraction, In this dissertation, based on " Research on Local Wave method and its Engineering Application" (National nature science fund project. No:50475155 ), combined with the pattern recognition and machine learning techniques, the problems of feature extraction and fault diagnosis are addressed using Local Wave method, Wigner higher-order time-frequency representation and blind source separation of non-stationary, non-Gaussian vibration signals. The main contents as follows:1. Local-wave method is studied and applied to mechanical vibration non-stationary signal. By comparing with wavelet and several time-frequency methods, the Local-wave method can be proved to more effective than others. The experiment result shows that Local-wave time-frequency analysis can clearly explain time varying characteristics of different fault modes. But this time-frequency method is a two-dimensional signal representation. It arises the dimensionality problem. To describe the signal with as few variables as possible, the geometric moments and margins were used as the features of the time-frequency distribution. Then, combined with the artificial neural network, the fault diagnosis method is proposed based on geometric moments and marginal densities of Local-wave.2. The effectiveness of a multi-component neural-network architecture based on Local wave for the time series prediction of non-stationary, nonlinear dynamic systems has been investigated. The simulated experiment for sunspots' benchmark suggests that the multi-component architecture outperforms the corresponding single-scale architectures. Then, an observer bank of autoregressive time series models based on multi-component neural-network architecture is used for model diagnosis of rotor fault vibration signals. These models can be used as one step ahead predictors allowing comparison of signals for the purposes of fault detection and diagnosis. From the prediction error, features can be extracted and used to determine the machine's condition. Vibration data from a rotor placed under different fault conditions were used for training and testing models. The experiment results indicate that this approach could be used to diagnose fault conditions.3. Higher order time-frequency distributions and their applications to feature extraction of machine vibration signals are studied. The detection of these impulses can be useful for fault diagnosis purposes in condition monitoring environment. The features can be effectively
extracted for these non-stationary, non-Gaussian vibration signals by the Wigner higher order moment spectrum. The Local wave decomposition method is used to suppress the cross-term for multiple signals in higher order time-frequency distributions and the simulated results are satisfactory. The vibration signals measured from diesel engine in the stage of deflagrate are analyzed with Wigner higher order moment spectrum. Experimental results indicate that this method has good potential in mechanical fault feature extraction and can be used to extract the valuable quantifiable information about the time-frequency features of impacts.4. The blind source separation fault diagnosis method based on Local wave time-frequency images is developed. ICA has been mainly used on the problem of blind signal separation. ICA is a feature extraction technique which can be considered a generalization of principal component analysis (PCA). ICA can be used to gain the local features which give better image representation. Because the Local wave time-frequency image can reflect the character of different fault conditions, blind source separation as a new signal pr
引文
[1] 黄文虎,夏松波,刘瑞吉等.设备故障诊断原理、技术及应用.北京:科学出版社.1996.
[2] 徐敏,黄邵毅.设备故障诊断手册—机械设备状态检测和故障诊断.西安:西安交通大学出版社.1998.
[3] 孙惠国,张瑞林.设备状态监测与故障诊断.郑州:河南科学技术出版社.1993.
[4] 邝朴生,蒋文科,刘刚等.设备诊断工程.北京:中国农业科学出版社.1997.
[5] 陈进.信号处理在机械设备故障诊断中的应用(连载).振动与冲击.1999.18(3):91-93
[6] Newland D E. Wavelet analysis of vibration, part 1: theory, Transactions of the ASME. Journal of Vibration and Acoustics. 1994. 116. 409-416.
[7] Cohen L著,白居宪译.时频分析:理论与应用.西安:西安交通大学出版社.1998.
[8] 王宏禹.非平稳随机信号分析与处理.北京:国防工业出版社.1999.
[9] 张贤达,保铮.非平稳信号分析与处理.北京:国防工业出版社.1998.
[10] 张贤达.时间序列分析—高阶统计量方法.北京:清华大学出版社.1996.
[11] 郑海波.非平稳非高斯信号特征提取与故障诊断技术研究:(博士学位论文).合肥:合肥工业大学.2002.
[12] 何正嘉,訾艳阳,孟庆丰等.机械设备非平稳信号的故障诊断原理及应用.北京:高等教育出版社.2000.
[13] 訾艳阳.基于非平稳信号特征提取原理的实用诊断技术研究:(博士学位论文).西安:西安交通大学.2000.
[14] 王奉涛.非平稳信号故障特征提取与智能诊断方法的研究及应用:(博士学位论文).大连:大连理工大学.2003.
[15] 陈仲生.直升机旋转部件故障特征提取的高阶统计量方法研究:(博士学位论文).长沙:国防科学技术大学.2004.
[16] Daubechies Ⅰ. The wavelet transform, time-frequency localization and signal analysis. IEEE Trans. on Information Theory. 1990. 36(5): 961-1005.
[17] Wang W J, McFadden P D. Application of wavelets to gearbox vibration signals for fault detecion. Journal of Sound and Vibration. 1996. 192(5): 927-939.
[18] Wang W J, McFadden P D. Application of orthogonal wavelets to early gear damage detection. Mechanical System and Signal Processing. 1995. 9(5): 497-507.
[19] Staszewski W J. Wavelet based compression and feature selection for vibration analysis. Journal of Sound and Vibration. 1998. 211(5): 735-760.
[20] 高强,何正嘉.谐波小波及其时频剖面图在旋转机械诊断中的应用.西安交通大学学报.2000.34(9):62-66.
[21] 王江萍,王鸿飞,王素英.基于小波多分辨率分析的往复机械故障诊断特征提取与识别.西安石油学院学报.1998.13(1):30-32.
[22] 刘献栋,李其汉.小波变换在转子系统动静件早期碰摩故障诊断中的应用.航空学报.1999.20(3):220-223.
[23] 赵纪元,何正嘉.小波包-自回归谱分析及在振动诊断中的应用.振动工程学报.1995.8(3):198-203.
[24] 耿中行,屈梁生.小波包原理及其在机械故障诊断中的应用.机械工程学报.1997.33(3):76-79.
[25] 张佩瑶,马孝江.小波包信号提取算法及其在故障诊断中的应用.大连理工大学学报.1997.37(1):67-72.
[26] 伍学奎,陈进,周轶尘.基于小波包变换的内燃机气阀漏气诊断方法.振动工程学报.2000.13(2):210-215.
[27] 杨国安,钟秉林,黄仁等.机械故障信号小波包分解的时域特征提取方法研究.振动与冲击.2001.20(2):25-28.
[28] 夏勇,商斌梁,张振仁.图像处理在内燃机故障诊断中的应用研究.机械科学与技术.2001.20(3):432-437.
[29] Rao P, Talor F. Real-time monitoring of vibration using the Wigner distribution. Sound and Vibration Acoustics. 1990. 5. 22-25.
[30] Meng Q F, Qu L S. Rotating machinery fault diagnosis using Wigner distribution. Mechanical Systems and Signal Processing. 1991. 5(3): 155-166.
[31] 袁小宏,屈梁生.机器振动诊断中信号处理方法的研究.西安交通大学学报.2001.35(7):714-717.
[32] Samimy B, Rizzoni G. Mechanical Signature Analysis Using Time Frequency Signal Processing: Application to Internal Combustion Engine Knock Detection. Proceeding of the IEEE. 1996. 84(9): 1330-1343.
[33] Staszewski W J, Worden K, Tomlinson D R. Time-frequency in gearbox fault detection using the Wigner-Ville distribution and pattern recognition. Mechanical System and Signal Processing. 1997. 11(5): 673-692.
[34] Les E A, Gray D B, Siva B N. Applications of time-frequency analysis from manufacturing and machine monitoring sensors. Proceedings of The IEEE. 1996. 84(9): 1319-1329.
[35] Huang N E, Shen Z, long S R. The empirical mode decomposition and the hilbert spectrum for nonlinear and non-stationary time series analysis. Proc. Roy. Soc. Vol. 454. 1998.
[36] 马孝江,余泊,张志新等.一种新的时频分析方法-局域波法.振动工程学报.2000.13(5):219-224.
[37] 余泊.自适应时频分析方法及其在故障诊断中的应用研究:(博士学位论文).大连:大连理工大学.1998.
[38] 盖强.局域波时频分析的理论研究与应用:(博士学位论文).大连:大连理工大学.2001.
[39] 钟佑明,秦树人,汤宝平.一种振动信号新变换法的研究.振动工程学报.2002.15(2):231-238.
[40] Zhang HY and Ma XJ. Wigner-Ville Distribution Based on Intrinsic Mode Functions. International Conference on Radar in Beijing. 2001. 10: 165-168.
[41] Zhang H Y, Ma X J. A New Method for the Analysis of Non-stationary Signal: The Time-Varying AR Model Based on Intrinsic Mode Functions. ISTM in ShangHai. 2001. 6: 201-205.
[42] Ga Q, Ma X J. The partial wave method for the analysis of non-stationary signals and its use in machine fault diagnosis. Proceedings of the International Symposium on Test and Measurement. 2001. 6: 1465-1468.
[43] 王珍.基于局域波分析的柴油机故障诊断方法的研究及应用:(博士学位论文).大连:大连理工大学.2002.
[44] 韩春明,郭华东,王长林.利用经验模态分解方法抑制SAR斑点噪声.遥感学报.2002.6:266-271.
[45] 宋平舰,张杰.二维经验模态分解在海洋遥感图象信息分离中的应用.高技术通讯.2001.9(62):67.
[46] 杨世锡,胡劲松,吴昭同.旋转机械振动信号基于EMD的希尔伯特变换和小波变换时频分析比较.中国电机工程学报.2003.23(6):102-107.
[47] 马孝江,王凤利,蔡悦.局域波时频分布在转子系统早期故障诊断中的应用研究.中国电机工程学报.2004.24(3):161-164.
[48] 邹岩昆.局域波分析的理论方法研究及应用:(博士学位论文).大连:大连理工大学.2005.
[49] Fonollosa J R, Nikias C L. Analysis of finite-energy signals using higher order moments- and spectra-based time-frequency distributions. Signal Processing. 1994. 36(3): 315-328.
[50] Fonollosa J R, Nikias C L. Transient signal detection using the wigner bispectrum. Conference record of the twenty-fifth asilomar conference on signal, systems and computers. 1991. 2: 1087-1092.
[51] 陈进,姜鸣.高阶循环统计量理论在机械故障诊断中的应用.振动工程学报.2001.14(2):125-134.
[52] Fackrell J W A, White P R, Hammond J K. The interpretation of the bispectra of vibration signal-Ⅰ: theory. Mechanical system and signal processing. 1995. 9(3): 257-266.
[53] Fackrell J W A, White P R, Hammond J K. The interpretation of the bispectra of vibration signal-Ⅱ: Experiment results ans applications. Mechanical system and signal processing. 1995. 9(3): 257-266.
[54] Nandi, A K, Dickie, J R, and Smith, J A. Classification of conditions of rotating machines using higher order statistics. IEE Colloquium on Higher Order Statistics in Signal Processing: Are They of Any Use? 1995. 3. 1-6.
[55] 马孝江,袁景侠.双谱分析及其在机械故障诊断中的应用.大连理工大学学报.1989.29(2):177-182.
[56] 王子平,叶佩青,杨叔子.基于实验的碰摩转子系统故障的双谱特征.振动工程学报.2002.15(1):62-67.
[57] 杨江天,陈家骥,曾子平.基于高阶谱的旋转机械故障征兆提取.振动工程学报.2002.14(1):13-17.
[58] 李志农,丁启全,吴昭同.旋转机械典型故障的高阶谱特征分析.汽轮机技术.2002.44(1):42-44.
[59] 刘天雄,郑明刚,朱继梅.基于双谱分析的设备状态检测研究.振动与冲击.2000.19(3):37-42.
[60] 苏文斌,史维祥,温熙森.故障诊断中的非线性耦合特征提取(一).机械研究与应用.1998.11(2):12-13.
[61] 苏文斌,史维祥,温熙森.故障诊断中非线性耦合特征提取的研究.西安交通大学学报.1999.33(1):88-92.
[62] Fonoliosa J R, Nikias C T. Wigner higher-order moment spectra: defination, properties, computation and application to transient signal analysis. IEEE Trans. On Signal Processing. 1993. 41(1): 245-266.
[63] Hinich M J. Detecting a transient signal by bispectral analysis. IEEE Trans. on Acoustics, Speech, and Signal Processing. 1990. 38(7): 1277-1283.
[64] Fonollosa J R, Nikias C L. Wigner higher-order moment spectra: defination, properties, computation and application to transient signal analysis. IEEE Trans. On Signal Processing. 1993. 41(1): 245-266.
[65] Boashash B and Ristich B. Application of cumulant TVHOS to the analysis of composite FM signals in multicative and additive noise. Proceedings of SPIE. 1994. 2027: 245-255.
[66] Swani A. Third-order Wigner distributions: defination properties. Proceedings of the ICASSP. 1991. 3081-3084.
[67] Lee S K, White P R. Higher-order time-frequency analysis and its application to fault detection in rotating machinery. Mechanical Systems and Signal Processing. 1997. 11(4): 637-650.
[68] 来五星,轩建平,史铁林等.Wigner-Ville时频分布研究及其在齿轮故障诊断中的应用.振动工程学报.2003.16(2):247-250.
[69] Lee S K. Higher order time-frequency analysis as a tool for health monitoring. Key Engineering Materials. 2003. Vols. 245-246. 175-182.
[70] Colas M, Gelle G, Delaunay G. Sliced higher order time-frequency representations for detection and classification. Acoustical diagnosis application to faulty fan belts. IEEE Signal Processing Workshop on Higher-Order Statistics. 1999. 183-186.
[71] Swani A, Movellan J R, Nikias C L. Higher-Order Spectral Analysis Toolbox User's Guide. 1997.
[72] Lee S K. A new method for smoothing non-oscillation cross-terms in sliced Wigner fourth order moment spectra. Mechanical Systems and Signal Processing. 2001. 15(5): 1023-1029.
[73] 张桂才,史铁林,杨叔子.基于高阶统计量的机械故障特征提取方法研究.华中理工大学学报.1999.27(3):6-8.
[74] Bell A J, Sejnowski T J. An information-maximization approach to blind separation and blind deconvolution. Neural Computation. 1995. 7(6): 1129-1159.
[75] Cichocki A, Amari S, and Thawonmas R. Blind signal extraction using self-adaptive nonlinear Hebbian learning rule. Proc. the 1996 International symposium on Linear theory and Application. 1996. 337-380.
[76] Cardoso J F, Souloumiac A. Blind beamforming for non Gaussian signals. IEE Proceedings-F. 1993. 140(6): 362-370.
[77] Comon P. Independent component analysis, a new concept? Signal Processing. 1994. 36(2): 287-314.
[78] Yang H H, Amari S. Adaptive on-line learning algorithms for blind separation-Maximum entropy and minimum mutual information. Neural Computation. 1997. 9(7): 1457-1482.
[79] Amari S, Cichocki A, and Yang H H. A new learning Algorithm for Blind Signal Separation. Advances in Neural Information Processing Systems, Cambridge: MIT Press. 1996. 8. 557-563.
[80] Amari S. Natural Gradient Works Efficiently in Learning. Neural Computation. 1998. 10(2): 251-276.
[81] Amari S. Differential geometrical methods of statistics. Springer Lecture Note in Statistics. Heidekberg: Springer-verlag. 1985. 28.
[82] Karhunen J, Oja E, Wang L, Vigaro R, Joutsensalo J. A class of neural networks for independent component analysis. IEEE trans on neural networks. 1997. 8(3): 486-504.
[83] Hyvarinen A, Oja E. A fast Fixed-point algorithm for Independent component analysis. Neural Computation. 1997. 9(7): 1483-1492.
[84] Hyvarinen A. Fast and Robust fixed-point algorithms for Independent Component Analysis. IEEE Trans. on Neural Networks. 1999. 10(3): 626-634.
[85] Amari S, Cichocki A. Adaptive blind signal processing-neural network approaches. Proceedings IEEE. 1998. 86(10): 2026-2048.
[86] Cardoso J F. Blind signal separation: statistical principles. Proceedings of IEEE. 1998. 86(10): 2009-2025.
[87] Hyvarinen A. Survey on Independent Component Analysis. Neural Computing surveys. Neural Computing surveys. 1999. 2. 94-128.
[88] Hyvarinen A. Gaussian Moments for Noisy Independent Component Analysis. IEEE Signal Processing Letters. 1999. 9(9): 145-147.
[89] Hochreiter S, Schmidhuber J. Feature extraction through LOCOCODE. Neural Computation. 1999. 11(3): 679-714.
[90] Cheung Y M, Xu L. Independent component ordering in ICA time series analysis. Neurocomputing. 2001. 41: 145-152.
[91] 陈华富,尧德中,Sue Becker.一种新的数据处理方法:邻域IC相关法及其初步应用.中国科学A辑.2004.32(5):685-692.
[92] 杨绿溪,李克,周长春等.一种用于超高斯和亚高斯混合信号盲分离的新算法.东南大学学报.1999.29(1):1-7.
[93] 刘琚,梅良模,何振亚.一种盲信号分离的信息理论方法.山东大学学报.1998.34(4):398-403.
[94] 刘琚,鲁子奕,何振亚等.基于信息论准则的盲源分离方法.应用科学学报.1999.17(2):156-162.
[95] 冯大政,史维祥.一种自适应信号盲分离和盲辩识的有效算法.西安交通大学学报.1998.35(2):76-79.
[96] 虞晓,胡光锐.基于统计估计的盲信号分离算法.上海交通大学学报.1999.33(5):566-569.
[97] 华蓉,苏中义.基于遗传算法过程的盲分离.上海交通大学学报.2001.35(2):319-322.
[98] 林家骏,乐慧丰,俞金寿.过程信号的盲分离.华东理工大学学报.1999.25(5):510-513.
[99] 虞晓,胡光锐.基于神经网络的非线性盲信号分离.上海交通大学学报.1999.33(9):1093-1096.
[100] 林秋华,殷福亮.盲源分离自适应算法的统一形式.大连理工大学学报.2002.42(4):498-503.
[101] 张旭秀,邱天爽.基于源信号之间统计独立性的ICA方法的等价性研究.探测与控制学报.2003.25(4):21-25.
[102] 杨竹青,李勇,胡德文.独立成分方法综述.自动化学报.2002.28(5):762-772.
[103] 陈华富,尧德中.独立成分分析及其应用的研究进展.生物医学工程学杂志.2003.20(2):366-370.
[104] 李木森,毛剑琴.盲信号分离的现状和展望.信息与电子工程.2003.1(1):69-79.
[105] Lee Te-Won, Independent component analysis: theory and application. La Jolla, Calofornia: Kluwer Academic Publishers, 1998.
[106] Sahlin Henrik, Broman Holger. Separation of real-world signals. Signal Processing. 1998. 64(1): 103-113.
[107] 杨竹青.ICA在脑图象序列中的应用:(硕士学位论文).长沙:国防科学技术大学.2002.
[108] 倪晋平.水声信号盲分离技术研究:(博士学位论文).西安:西北工业大学.2002.
[109] Ypma A, Pajunen P. Rotating machine vibration analysis with second-order independent component analysis. Proc. of the workshop on ICA and Signal separation. Aussois, France. 1999. 37-42.
[110] Ypma A, Tax D M J, Duin R P W. Robust machine fault detection with independent component analysis and support vector data description. Proc. IEEE Signal Processing Society Workshop on Neural Networks for Signal Processing. Madison, Wisconsin, USA. 1999. 67-76.
[111] Ypma A, Leshem A. Blind separation of machine vibration with bilinear forms, Proc. 2nd Int. Workshop on ICA and Signal Separation, Helsinki, Finland. 2000. 405-410.
[112] Ypma A, Leshem A, Duin R P W. Blind separation of rotating machine sources: bilinear forms and convolutive mixtures. Neurocomputing. 2002. 49(1-4): 349-368.
[113] Gelle G, Colas M, Delaunay G. Blind sources separation applied to rotating machines monitoring by acoustical and vibrations analysis. Mechanical Systems and Signal Processing. 2000. 14(3): 427-442.
[114] Gelle G, Colas M, Serviere C. Blind sources separation: A tool for rotating machine monitoring by vibration analysis. Journal of Sound and Vibration. 2001. 248(5): 865-885.
[115] Li W, Gu F. Ball A D, Leung A Y T, Phipps C E. A study of a noise from diesel engines using independent component analysis. Mechanical Systems and Signal Processing. 2001. 15(6): 1165-1184.
[116] Roan M J, Erling J G, Sibul L H. A new, non-linear, adaptive, blind source separation approach to gear tooth failure detection and analysis. Mechanical Systems and Signal Processing. 2002. 16(5): 719-740.
[117] Poyhonen S, Jover P, Hyotyniemi H. Independent component analysis of vibration for fault diagnosis of an induction motor. IASTED International Conference on Circuits, Signals, and Systems. Cancun, Mexico. 2003. 203-208.
[118] 张海军,温广瑞,屈梁生.一种提高诊断信息质量的方法.西安交通大学学报.2000.36(3):295-299.
[119] 李力,屈梁生.应用独立分量分析提取机器的状态特征.西安交通大学学报.2003.37(1):45-48.
[120] 张海军.机械故障诊断和预测中的信息获取:(博士学位论文).西安:西安交通大学,2002.
[121] 钟振茂,陈进,钟平.盲源分离用于机械故障诊断的研究初探.机械科学与技术.2002.21(2):282-284.
[122] 吴军彪,陈进,伍星.基于盲源分离技术的故障特征信号分离方法.机械强度.2002.24(4):485-488.
[123] 吴军彪,陈进,钟平等.机械噪声故障特征提取的盲分离法与小波提纯法.上海交通大学学报.2003.37(5):766-769.
[124] 杨世锡,焦卫东,吴昭同.独立分量分析基网络应用于旋转机械故障特征抽取与分类.机械工程学报.2004.40(3):45-49.
[125] 焦卫东,杨世锡,吴昭同.基于源数估计的旋转机械源盲分离.中国机械工程.2003.14(14):1184-1187.
[126] 焦卫东,杨世锡,吴昭同.机械故障模式识别的ICA基神经网络方法.计算技术与自动化.2003.22(2):63-67.
[127] 焦卫东,杨世锡,吴昭同.复合ICA_SVM 机机械状态模式分类.中国机械工程.2004.15(1):62-65.
[128] 陈仲生,杨拥民,沈国际.独立分量分析在直生机齿轮箱故障早期诊断中的应用.机械科学与技术.2004.23(4):481-500.
[129] 王永,时文刚,吕淘.ICA及其在往复泵多源信号分离中的应用.石油矿场机械.2003.32(5):62-64.
[130] 宋晓萍.多机组系统机器振动信号与声学信号的盲分离:(硕士学位论文).西安:西北工业大学.2004
[131] Melvile W K. Wave modulation and breakdown. Journal of Fluid Mech. 1983. 128. 489-506.
[132] Boashash B. Estimating and interpreting the instantaneous frequency of signal. Part Ⅰ: Fundamentals. Proceeding of the IEEE. 1992. 80. 520-538.
[133] 王凤利.基于局域波法的转子系统非线性动态特性及应用研究:(博士学位论文).大连:大连理工大学.2003.
[134] Tacer B, Loughlin P J. Non-stationary signal classification using the joint moments of time- frequency distributions. Pattern Recognition. 1998. 33. 1635-1641.
[135] Leonowicz Z and Lobos T. hpplication of Time-Frequency Distribution and Neural Networks for Fault Classification in Power Electronics. International symposium on Computational Intelligence for Measurements and applications. 2003. 67-71.
[136] 邹岩昆,马孝江,朱泓等.局域波法在柴油机气缸磨损故障诊断中的研究.中国机械工程.2004.15(20):1811-1814.
[137] Matlab ToolBox. Neural Networks. Radial Basis Networks, online http://www.mathworks.com. 1998.
[138] 陈庆瀚,杜金鸿. High-Level Synthesis of Probabilistic Neural Network Classifier. Online http://www. csie. cyut. edu. tw/TAAI2OO2/TAAI2002PDF/. 2002.
[139] Powell M J D, Radial Basis Functions for Multivariate Interpolation: A Review Clarendon Press, Oxford, 1987.
[140] 陈重阳,蔡萍,施文康等.基于径向基函数概率神经网络的心率失常自动识别.上海交通大学学报.2000.34:1476-1477.
[141] 王柏生,倪一清,高赞明.用概率神经网络进行结构损伤位置识别.振动工程学报 2001.14:60-64.
[142] Amir B G. ScaleNet-Multiscale Neural-Network Architecture for Time Series Prediction. IEEE Trans. on Neural Networks. 1998. 9(6): 1471-1482.
[143] 陈熙源,程启明,万德钧等.神经网络时间序列预测及其工程应用.应用科学学报.1999. 17(2):228-231.
[144] 钮永胜,赵新民,孙金玮.采用基于神经网络的时间序列预测器的传感器故障诊断新方法.仪器仪表学报.1998.19(4):383-388.
[145] 贺国光,马寿峰,李宇.基于小波分解与重构的时间序列预测法.自动化学报.2002.28(6):1012-1014.
[146] 徐科,徐金梧,班晓娟.基于小波分解的某些非平稳时间序列预测方法.电子学报.2001.29(4):566-568.
[147] 曹星平,易东云,吴翊.基于神经网络的时间序列预测方法进展.电脑与信息技术.1999.(6):1-3.
[148] Nielsen R H. Theory of the back propagation neural network. Proc. Int. Joint Conf. Neural Networks. 1989. 1: 593-606.
[149] Bishop C M. Neural Networks for Pattern Recognition. Oxford, U. K.: Oxford Univ. Press. 1995.
[150] More J J. The Levenberg-Marquardt algorithm: Implementation and theory. Numerical Analysis. New York: Springer-Verlag, 1977.
[151] 杨叔子,吴雅,王治藩等.时间序列分析的工程应用.华中理工大学出版社,1991.
[152] Hagan M T, Menhaj M B. Training feedforward networks works with the marquardt algorithm. IEEE Trans. on Neural Networks. 1994. 5(6): 989-993.
[153] 赵弘,周瑞祥,林廷圻.基于Levenberg算法的神经网络监督控制.西安交通大学学报.2002.36(5):523-527.
[154] Bakshi B R, Stephanopoulos G. Reasoning in time: Modeling, analysis, and pattern recognition of temporal process trends. Advance in Chem. Eng. 1995. 22: 485-547.
[155] Baillie D C, Mathew J. A comparison of autoregressive modeling techniques for fault diagnosis of rolling element bearings. Mechanical System and Signal Processing. 2004. 10(1): 1-17.
[156] McCormick A C, Nandi A K. Neural networks autoregressive modeling of vibration for condition monitoring of rotating shafts. International Conference on neural networks. 1997. 2214-2218.
[157] Mechefske C K, Mathew J. Fault detection and diagnosis in low speed roiling element bearings, part Ⅰ: The use of parametric Spectra. Mechanical Systems and Signal Processing. 1992. 6(4): 297-307.
[158] Sorsa T, Suontausta J, and Koivo M N. Fault diagnosis of dynamic systems using neural networks. 12th IFAC Conference. 1993.
[159] Hammond J K, White P R. The analysis of nonstationary signals using time-frequency methods. Journal of Sound and Vibration. 1996. 190(3): 419-447.
[160] Choi H I, Williams W J. Improved time-frequency representation of multicomponent signal using exponential kernels. IEEE Transactions on Acoustics, Speech, and Signal Processing. 1989. 37(6): 862-871.
[161] Gerr N L. Introducing a third-order wigner distribution. Proceedings of the IEEE. Vol. 76. 1988. 290-292.
[162] 沈民奋,孙丽莎.现代随机信号与系统分析.北京:科学出版社.1998.
[163] Jones D L, Baraniuk R O. An adaptive optimal-kernel time-frequency representation. IEEE Transactions on Acoustics, Speech, and Signal Processing. 1995. 43(10): 2361-2371.
[164] 张海勇.基于局域波法的非平稳随机信号分析中若干问题的研究:(博士学位论文).大连:大连理工大学.2001.
[165] 胡以怀,杨叔子,刘永长等.柴油机磨损故障振动诊断机理的研究.内燃机学报.1998.16(1):50-61.
[166] 王珍,马孝江.局域波时频法在柴油机缸套活塞磨损诊断中的应用研究.内燃机学报.2002.20(2):157-160.
[167] 余成波,何怀波,石晓辉.内燃机振动控制及应用。北京:国防工业出版社.1997.
[168] 谭达明.内燃机振动控制.成都:西南交通大学出版社.1993.
[169] E. Oja. Neural networks, principle components and subspaces. International Journal of Neural Systems. 1989. 1: 61-68.
[170] Hyvarinen A, Oja E. Independent component analysis: algorithms and applications. Neural Computation. 2000. 13: 411-430.
[171] Hyvarinen A, Oja E. A fast Fixed-point algorithm for Independent component analysis. Neural Computation. 1997. 9(7): 1483-1492.
[172] 王善永,陆颂元,马元奎.汽轮发电机组转子动静碰摩故障检测的小波分析方法研究.中国电机工程学报.1999.19(3):1-6.
[173] 彭志科,何永勇,卢青.用小波时频分析方法研究发电机碰摩故障特征.中国电机工程学报.2003.23(5):75-79.
[174] Bartlett M, Movellan J R, Sejnowski T J. Face recognition by independent component analysis. IEEE trans on neural networks. 2002. 13(6): 1450-1464.
[175] Bartlett M, Lades H, and Sejnowski T J. Independent component representations for face recognition. Procedings of the SPIE, Vol. 3299: Conference on Human Vision and Electronic Imaging Ⅲ. 1998. 528-539.
[176] 吴小培,冯焕清,周荷琴.基于独立分量分析的图像分离技术及应用.中国图像图形学报.2001.6(2):133-137.
[177] 丁佩律,梅剑锋,张立明等.基于独立分量分析的人脸自动识别方法研究.红外与毫米波学报.2001.20(5):361-364.
[178] 乌凌超,莫玉龙.基于独立分量分析的字符识别方法.上海大学学报.2003.2003(9):3-193.
[179] 王宏漫,欧宗瑛.采用PCA_ICA特征和SVM分类的人脸识别.计算机辅助设计与图形学学报.2003.15(4):416-420.
[180] Belouchrani A, Merain K, Cardoso J F. A blind source separation techneque using second order statistics. IEEE Trans. On Signal Processing. 1997. 45(2): 434-443.
[181] Belouchrani A, Amin M. Blind source separation based on time-frequency signal representations. IEEE trans. On signal processing. 1998. 46(11): 2888-2897.
[182] Baraniuk R, Jones D. A signal dependent time-frequency representation: Optimum kernel design. IEEE Trans. On Signal Processing. 1993. 41 1589-1603.
[183] Fevotte C, Doncarli C. Two contributions to blind source separation using time-frequency distributions. Signal Processing Letters, IEEE. 2004. 11(3): 386-389.
[184] Holobar A, Fevotte C, Doncarli C, and Zazula D. Single autoterms selection for blind source separation in time-frequency plane. Proc. EUSIPCO. 2002. 1: 565-568.
[185] 刘琚,杜正锋,梅良模.基于Wigner-Ville分布的非平稳信号盲分离.山东大学学报.2003.38(1):73-76.
[186] Choi H I, Williams W J. Improved time frequency representations of multicomponent signals using exponential kernels. IEEE Trans. On Signal Processing. 1989. 37(6): 862-871.
[187] 胡昌华,周涛,夏启兵等.基于Matlab的系统分析与设计—时频分析.西安:西安电子科技大学出版社.2002.