基于阳极电流波动信号的铝电解槽槽况诊断方法研究
|
摘要
铝电解槽的稳定性研究是铝电解生产中的热点课题,从铝电解槽可在线采集量入手诊断其运行状况,进行相应的操作是提高其稳定性的重要措施。阳极电流是铝电解槽中一个能够在线采集的重要参数,其信号中包含大量的与电解槽运行状况相关的信息。可从阳极电流信号中提取槽况特征,建立相应的槽况诊断模型。
以160kA系列预焙铝电解槽为研究对象,采集了正常槽和典型故障槽的阳极电流信号,利用“频谱—小波包—神经网络”相结合的方法对信号进行分析处理,得到以下结论：
(1)利用频谱分析方法,研究不同槽况下阳极电流信号的频谱特点。采用Yule-Walker方法的功率谱估计提取出不同槽况阳极电流的频谱特征,较好区分了正常槽、冷槽及阴极破损槽,并得到正常槽、冷槽及阴极破损槽功率谱曲线主谱峰的频率范围,分别为：0.003～0.018Hz、0.023～0.027 Hz、0.027～0.031 Hz。将功率谱分析结果与HHT的边际谱分析结果进行对比,结果表明功率谱分析结果具有较好的可靠性。
(2)根据功率谱分析结果,提出了阳极电流信号的小波包能量特征概念。对阳极电流信号进行小波除噪,对除噪后的阳极电流信号进行小波包4层分解和5层分解,并提取其能量特征向量。根据小波包4层分解和5层分解的结果,优化分析,选择合适的小波包重构系数,提取了不同槽况下的能量特征向量,并做归一化处理,使不同槽况下能量特征向量的差距明显,较好的区分了不同槽况下的阳极电流信号的特征,为槽况的诊断提供依据。
(3)根据阳极电流信号小波包能量特征向量,建立了基于BP神经网络的故障诊断模型,并对其验证。仿真与验证结果表明：此网络辨别的模式简单,识别的准确率高,有利于在线监测和实时识别。
Stability is one of the spotlights in aluminum electrolysis process research. Diagnosing the working conditions of an aluminum reduction cell from its online parameters in order to take proper measures is an important strategy to enhance its stability. Anode current is one of the important parameters that can be collected online, whose signals contain a lot of information concerning the performance of an aluminum reduction cell. Diagnosis method is established by extracting characteristics from anode current signals.
As the object of 160kA pre-baked anode cells, anode current signals were collected under either normal cells or typical problem cells. These signals were analyzed by the method of "spectrum-wavelet packet-neural network" with results are as follows:
(1) Spectral characteristics of anode current signals in different cell states were investigated by spectral analysis method. Power spectrum estimation based on the Yule-Walker method was put forward to extract the spectral characteristics of anode current signals in different cell states, by which normal, cold and cathode fail cell would be distinguished by recognizing their frequency ranges of main peak as 0.003-0.018Hz,0.023-0.027Hz, and 0.027-0.031Hz respectively. The results of power spectrum estimation analysis have been compared with the results of HHT (Hilbert-Huang Transform) marginal spectrum analysis, which shows it is reliable for power spectrum estimation to analyze the anode current signals.
(2) According to the results of power spectrum analysis, the concept of wavelet packets energy characteristics for anode current signal was introduced. Wavelet packets decomposition was performed at level 4 and 5 for anode current signals de-noised with wavelet, and related decomposition energy characteristics vectors were extracted. The results of wavelet packets decomposition was optimized, the appropriate wavelet packets reconstruction was selected, and the energy characteristics vectors of anode current signals were extracted in different cell states. Energy characteristics vectors that were normalized have obvious difference, which make a better distinction between the different cell states, provide the evidence for the diagnosis of cell states.
(3) According to the wavelet packet energy characteristics vectors extracted from anode current signals, diagnosis model based on BP neural network was established and verified. Simulation results show that the model of network identification is simple in construction, high accuracy in recognition, and convenient to realize on-line monitoring and real-time identification.
以160kA系列预焙铝电解槽为研究对象,采集了正常槽和典型故障槽的阳极电流信号,利用“频谱—小波包—神经网络”相结合的方法对信号进行分析处理,得到以下结论：
(1)利用频谱分析方法,研究不同槽况下阳极电流信号的频谱特点。采用Yule-Walker方法的功率谱估计提取出不同槽况阳极电流的频谱特征,较好区分了正常槽、冷槽及阴极破损槽,并得到正常槽、冷槽及阴极破损槽功率谱曲线主谱峰的频率范围,分别为：0.003～0.018Hz、0.023～0.027 Hz、0.027～0.031 Hz。将功率谱分析结果与HHT的边际谱分析结果进行对比,结果表明功率谱分析结果具有较好的可靠性。
(2)根据功率谱分析结果,提出了阳极电流信号的小波包能量特征概念。对阳极电流信号进行小波除噪,对除噪后的阳极电流信号进行小波包4层分解和5层分解,并提取其能量特征向量。根据小波包4层分解和5层分解的结果,优化分析,选择合适的小波包重构系数,提取了不同槽况下的能量特征向量,并做归一化处理,使不同槽况下能量特征向量的差距明显,较好的区分了不同槽况下的阳极电流信号的特征,为槽况的诊断提供依据。
(3)根据阳极电流信号小波包能量特征向量,建立了基于BP神经网络的故障诊断模型,并对其验证。仿真与验证结果表明：此网络辨别的模式简单,识别的准确率高,有利于在线监测和实时识别。
Stability is one of the spotlights in aluminum electrolysis process research. Diagnosing the working conditions of an aluminum reduction cell from its online parameters in order to take proper measures is an important strategy to enhance its stability. Anode current is one of the important parameters that can be collected online, whose signals contain a lot of information concerning the performance of an aluminum reduction cell. Diagnosis method is established by extracting characteristics from anode current signals.
As the object of 160kA pre-baked anode cells, anode current signals were collected under either normal cells or typical problem cells. These signals were analyzed by the method of "spectrum-wavelet packet-neural network" with results are as follows:
(1) Spectral characteristics of anode current signals in different cell states were investigated by spectral analysis method. Power spectrum estimation based on the Yule-Walker method was put forward to extract the spectral characteristics of anode current signals in different cell states, by which normal, cold and cathode fail cell would be distinguished by recognizing their frequency ranges of main peak as 0.003-0.018Hz,0.023-0.027Hz, and 0.027-0.031Hz respectively. The results of power spectrum estimation analysis have been compared with the results of HHT (Hilbert-Huang Transform) marginal spectrum analysis, which shows it is reliable for power spectrum estimation to analyze the anode current signals.
(2) According to the results of power spectrum analysis, the concept of wavelet packets energy characteristics for anode current signal was introduced. Wavelet packets decomposition was performed at level 4 and 5 for anode current signals de-noised with wavelet, and related decomposition energy characteristics vectors were extracted. The results of wavelet packets decomposition was optimized, the appropriate wavelet packets reconstruction was selected, and the energy characteristics vectors of anode current signals were extracted in different cell states. Energy characteristics vectors that were normalized have obvious difference, which make a better distinction between the different cell states, provide the evidence for the diagnosis of cell states.
(3) According to the wavelet packet energy characteristics vectors extracted from anode current signals, diagnosis model based on BP neural network was established and verified. Simulation results show that the model of network identification is simple in construction, high accuracy in recognition, and convenient to realize on-line monitoring and real-time identification.
引文
[1]邱竹贤.21世纪伊始铝工业的新进展[J].中国工程科学,2003,5(4)：41～46.
[2]王祝堂.铝电解技术的发展历史与进展[J].轻金属,2003,3.
[3]Petrucci R.H. General Chemistry, Principles and Modern applications(5th edition). New York:Macmillan,1989:79～85.
[4]Grijotheim K., Kvande H. Introduction to Aluminum Electrolysis(2nd Edition). Dusseldorf:Aluminum-Verlag,1993:102～105.
[5]Edwards J.D. The Aluminum Industry, Aluminum and its Production, Vol.1. New York:McGraw-Hill,1930:1030～1045.
[6]欧宝成,周渤,马骀.大型预焙槽阴极破损原因的探讨[J].轻金属,2008,4：34～36.
[7]刘业翔,李劫等.现代铝电解[M].北京：冶金工业出版社,2008：231～237.
[8]McFadden, Fiona J. Stevens, Bearne, Geoffrey P., Austin, Paul C. et al. Application of advanced process control to aluminium reduction cells-A review[J]. Light Metals,2001:1233～1242.
[9]Rolland,W.K., Steinsnes,A., Larsen,A.S. et al. Haldris-an expert system for process control and supervision of aluminum smelters[J]. Light Metals, 1991:437～443.
[10]赵显富.国外铝电解计算机控制技术的发展动态[J].冶金自动化,1991,15(2)：3～4.
[11]Berezin, A.I., Poliakov, P.V., Rodnov, O.O. et al. Neural network qualifier of noises of aluminium reduction cell[J]. Light Metals,2003:437～440.
[12]Berezin, A.I., Polyakov, P.V., Rodnov, O.O. et al. FMEA-based expert system for electrolysis diagnosis[J]. Light Metals,2005:429～434.
[13]Meghlaoui, A., Thibault, J., Bui, R.T. et al. Neural networks for the identification of the aluminium electrolysis process[J]. Computers and Chemical Engineering, 1998:1419-1428.
[14]Meghlaoui, A., Bui, R.T., Thibault, J. al. Intelligent control of the feeding of aluminum electrolytic cells using neural networks[J]. Metallurgical and Materials Transactions,1997:215-221.
[15]Meghlaoui, A., Bui, R.T., Thibault, J. et al. Predictive control of aluminum electrolytic cells using neural networks[J]. Metallurgical and Materials Transactions, 1998:1007～1019.
[16]Majid, Nazatul Aini Abd,Young, Brent R., Taylor, Mark P. et al. Detecting abnormalities in aluminium reduction cells based on process events using multi-way principal component analysis (MPCA)[J]. TMS Annual Meeting,2009:589-593.
[17]Stam, Marco A., Taylor, Mark P., Chen, John J. J. et al. Common behaviour and abnormalities in aluminium reduction cells[J]. Light Metals,2008:309-314.
[18]Stam, Marco A., Taylor, Mark P., Chen, John J.J. et al. Development of a multivariate process control strategy for aluminium reduction cells[J]. TMS Annual Meeting,2009:311-315.
[19]Stam, Marco A., Taylor, Mark P., Chen, John J. J. et al. Operational and control improvements in reduction lines at aluminium delfzijl[J]. Light Metals, 2007:243～247.
[20]Banta, Larry, Dai, Congxia, Biedler, Philip. Noise classification in the aluminum reduction process[J]. Light Metals,2003:431～435.
[21]Banta, Larry, Biedler, Philip, Dai, Congxia. Process signal classification for aluminum cells[J]. Proceedings of SPIE-The International Society for Optical Engineering,2001:91～100.
[22]Biedler, Philip, Banta, Larry, Dai, Congxia. Development of a state observer for an aluminum reduction cell[J]. Light Metals,2002:1091-1098.
[23]雷恩清,萧德云.基于COM的铝电解槽健康诊断系统软件平台[J].计算机工程,2004,30(9)：186～188.
[24]邢杰,萧德云.基于PCA的概率神经网络结构优化[J].清华大学学报(自然科学版),2008,48(1)：141～144.
[25]刘敏华,萧德云.基于SDG模型和模糊融合的故障诊断方法[J].控制工程,2006,13(1)：10～14.
[26]刘敏华.基于SDG模型的故障诊断及应用研究：[博士学位论文].北京：清华大学,2005.
[27]Hesong Li, Chi Mei, Naijun Zhou, et al. Diagnosis of working conditions of an aluminum reduction cell based on wavelet packets and fuzzy neural network[J]. Chemical Engineering and Processing,2006:1074-1080.
[28]姚恩秀,袁丁,袁工.模糊专家系统在铝电解控制中的应用[J].基础自动化,2000,7(1)：9～11.
[29]李界家,吴成.模糊神经网络技术在铝电解故障诊断中的应用[J].沈阳建筑大学学报(自然科学版),2005,21(4)：390～394.
[30]Li Jie, Xiao Jin, Liu Yexiang et al. Hierarchical intelligent control system for aluminum reduction cells[J]. Light Metals,1997:463～467.
[31]Li Jie, Wang Qianpu, Xiao Jin et al. Intelligent fuzzy control system for aluminum reduction cells[J]. Chinese Journal of Nonferrous Metals,1998:557～562.
[32]王有来,李晋宏,刘卫东等.模糊专家系统技术在大型预焙铝电解槽中的开发与应用[J].北方工业大学学报,200,13(1)：59～62.
[33]冯磊.经典功率谱估计与现代功率谱估计的对比[J].商业文化,2009,5：223～239.
[34]L. Colen. Time-frequency distributions:a review. IEEE. Proc.1989,77(7): 941～981.
[35]Huang N E. The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis. Proc. R. Soc. Lond. A(1998)454:903-995.
[36]Mallat S. A wavelet tour of signal processing[M]. New York:Academic Press, 1998:46～50.
[37]Coifman R.R., Wickerhauser M.V. Entropy-based algorithms for best basis selection. IEEE Trans, on Inf. Theory,1992:713～718.
[38]Peters A.K., Wellesley. The World According to Wavelets. New York:Academic Press,1996:45～52.
[39]Proakis J. G, Manolakis D.G Digital Signal Processing:Principles, Algorithms, and Applications. Englewood Cliffs, NJ:Prentice Hall,1996:18～25.
[40]S.GMallat. A theory for multiresolution signal decomposition:The wavelet representation[J]. IEEE Pattern Anal and Machine Intell,1989,11(7):674～693.
[41]Strang G, Nguyen T. Wavelets and filter banks[M]. Cambridge: Wellesley-Cambridge Press,1996:75～86.
[42]胡昌华,张军波,夏军,张伟等.基于MATLAB的系统分析与设计—小波分析(第一版)[M].西安：电子科技大学出版,1999：48～62.
[43]M Cocchi, R Seeber, A Ulrici,et al. Wavelet packet transform for efficient pattern recognition of signals[J]. Chemometrics and Intelligent Laboratory Systems, 2001,57:97～119.
[44]B Walczak, D L Massart, Wavelet packet transform applied to a set of signal:a new approach to the best-basis selection[J]. Chemometrics and Intelligent Laboratory System,1997,38:39～50.
[45]N Saito, R R Coifman, F B Geshwind, F Warner. Discriminant feature extraction using empirical probability density estimation and a local basis library[J]. Pattern Recognition,2002,35:2841-2852.
[46]Y Wu, R Du, Feature extraction and assessment using wavelet packet for monitoring of machining processes[J]. Mechanical Systems and Signal Processing, 1996,10(1):29～53.
[47]秦树人,季忠,尹爱军.工程信号处理[M].北京：高等教育出版社,2008：456-463.
[48]飞思科技产品研发中心.神经网络理论与MATLAB7实现[M].北京：电子工业出版社,2005：20～23,99～108.
[49]孙建国.铝电解槽阳极电流分布特征及设计上的思考[J].轻金属,2007,3：29～31.
[50]李贺松.大型预焙铝电解非稳态、非均一模型及关键极节能技术研究：[博士学位论文].长沙：中南大学,2005.
[51]N. Urata. Magnetics and metal pad instability[J].Light Metals,1985:581-589.
[52]H.Sun, O.Zikanov, B.A.Finlayson. Effect of background melt flow and interface distortion on the stability of Hall-Heroult cells[J]. Magnetohydrodynamics, 2005,41(3):273-287.
[53]S.J.S.MORRIS, P.A.DAVIDSON. Hydromagnetic edge waves and instability in reduction cells[J]. Journal of Fluid Mechanics,2003,493(2):121-130.
[54]Xiangwen Wang, Alton T Tabereaux. Anodic phenomena-observations of anode overvoltage and gas bubbling during aluminum electrolysis [J]. Light Metals,2000:239-247.
[55]Jilai Xue, A Harald. Bubble behaviour-cell voltage oscillation during aluminium electrolysis and the effects of sound and ultrasound[J] Light Metals,1995:265-271.
[56]Jacques Antille, Rene Von Kaenel. Using a magnetohydrodynamic model to analyze pot stability in order to identify an abnormal operation condition[J]. Light Metals,2002:477-482.
[57]D Shin, A D Sneyd. Metal pad instabilities in aluminium reduction cells[J]. Light Metals,2000:279-283.
[58]P Maillard, M V Romerio. A stability criterion for an in finitely long Hall-Heroult cell[J]. Journal of computational and applied mathematics,1996,71:47-65.
[59]飞思科技产品研发中心.MATLAB7辅助信号处理技术与应用[M].北京：电子工业出版社,2005：319-321.
[60]丁蕾,曾水平,曾铮.350kA铝电解槽槽电阻信号的频谱分析[J].仪器仪表 与检测技术,2005,24(12)：68-70.
[61]曾水平,丁蕾,李晋宏.基于槽电阻频谱分析的铝电解槽槽况诊断[J].中南大学学报(自然科学版),2007,38(1)：768-771.
[62]Huang N E, Shen Z, Long S R. A new view of nonlinear water waves:the Hilbert spectrum. Annu Rev Fluid Mech,1999,31:417-457.
[63]Leisk G,Hsu N, Huang N E. Application of the Hilbert-Huang Transform to machine tool condition/health monitoring. AIP conference, Brunswick, 2002,615(1):1711-1718.
[64]谭善文.多分辨希尔伯特—黄(Hilbert-Huang)变换方法的研究：[博士学位论文].重庆：重庆大学,2001.
[65]廖庆斌,李舜酩,覃小攀.车辆振动信号的特征提取方法比较[J].吉林大学学报(工学版),2007,37(4)：910-914.
[66]李舜酩,李香莲.振动信号的现代分析技术与应用[M].北京：国防工业出版.2008：217-220.
[67]杨建国.小波分析及其工程应用[M].北京：机械工业出版社，2005：68～79.
[68]单峰.大型预焙铝电解槽槽电压波动特征提取：[硕士学位论文].长沙：中南大学,2009.
[69]刘娟花.小波分析在语音去噪中的应用：[硕士学位论文].西安：西安理工大学,2003.
[70]周辉.齿轮故障的特征提取与模式识别技术研究[博士学位论文].郑州：郑州大学,2005：23～24.
[71]张立明.人工神经网络的模型及其应用[M].上海：复旦大学出版社,1993.
[72]周开利,康耀红.神经网络模型及其MATLAB仿真程序设计[M].北京：清华大学出版社,2005：67～83.
[73]Howard Demuth, Mark Beale. Neural network toolbox for use with MATLAB[M]. The Mathworks,2001:16-19.
[2]王祝堂.铝电解技术的发展历史与进展[J].轻金属,2003,3.
[3]Petrucci R.H. General Chemistry, Principles and Modern applications(5th edition). New York:Macmillan,1989:79～85.
[4]Grijotheim K., Kvande H. Introduction to Aluminum Electrolysis(2nd Edition). Dusseldorf:Aluminum-Verlag,1993:102～105.
[5]Edwards J.D. The Aluminum Industry, Aluminum and its Production, Vol.1. New York:McGraw-Hill,1930:1030～1045.
[6]欧宝成,周渤,马骀.大型预焙槽阴极破损原因的探讨[J].轻金属,2008,4：34～36.
[7]刘业翔,李劫等.现代铝电解[M].北京：冶金工业出版社,2008：231～237.
[8]McFadden, Fiona J. Stevens, Bearne, Geoffrey P., Austin, Paul C. et al. Application of advanced process control to aluminium reduction cells-A review[J]. Light Metals,2001:1233～1242.
[9]Rolland,W.K., Steinsnes,A., Larsen,A.S. et al. Haldris-an expert system for process control and supervision of aluminum smelters[J]. Light Metals, 1991:437～443.
[10]赵显富.国外铝电解计算机控制技术的发展动态[J].冶金自动化,1991,15(2)：3～4.
[11]Berezin, A.I., Poliakov, P.V., Rodnov, O.O. et al. Neural network qualifier of noises of aluminium reduction cell[J]. Light Metals,2003:437～440.
[12]Berezin, A.I., Polyakov, P.V., Rodnov, O.O. et al. FMEA-based expert system for electrolysis diagnosis[J]. Light Metals,2005:429～434.
[13]Meghlaoui, A., Thibault, J., Bui, R.T. et al. Neural networks for the identification of the aluminium electrolysis process[J]. Computers and Chemical Engineering, 1998:1419-1428.
[14]Meghlaoui, A., Bui, R.T., Thibault, J. al. Intelligent control of the feeding of aluminum electrolytic cells using neural networks[J]. Metallurgical and Materials Transactions,1997:215-221.
[15]Meghlaoui, A., Bui, R.T., Thibault, J. et al. Predictive control of aluminum electrolytic cells using neural networks[J]. Metallurgical and Materials Transactions, 1998:1007～1019.
[16]Majid, Nazatul Aini Abd,Young, Brent R., Taylor, Mark P. et al. Detecting abnormalities in aluminium reduction cells based on process events using multi-way principal component analysis (MPCA)[J]. TMS Annual Meeting,2009:589-593.
[17]Stam, Marco A., Taylor, Mark P., Chen, John J. J. et al. Common behaviour and abnormalities in aluminium reduction cells[J]. Light Metals,2008:309-314.
[18]Stam, Marco A., Taylor, Mark P., Chen, John J.J. et al. Development of a multivariate process control strategy for aluminium reduction cells[J]. TMS Annual Meeting,2009:311-315.
[19]Stam, Marco A., Taylor, Mark P., Chen, John J. J. et al. Operational and control improvements in reduction lines at aluminium delfzijl[J]. Light Metals, 2007:243～247.
[20]Banta, Larry, Dai, Congxia, Biedler, Philip. Noise classification in the aluminum reduction process[J]. Light Metals,2003:431～435.
[21]Banta, Larry, Biedler, Philip, Dai, Congxia. Process signal classification for aluminum cells[J]. Proceedings of SPIE-The International Society for Optical Engineering,2001:91～100.
[22]Biedler, Philip, Banta, Larry, Dai, Congxia. Development of a state observer for an aluminum reduction cell[J]. Light Metals,2002:1091-1098.
[23]雷恩清,萧德云.基于COM的铝电解槽健康诊断系统软件平台[J].计算机工程,2004,30(9)：186～188.
[24]邢杰,萧德云.基于PCA的概率神经网络结构优化[J].清华大学学报(自然科学版),2008,48(1)：141～144.
[25]刘敏华,萧德云.基于SDG模型和模糊融合的故障诊断方法[J].控制工程,2006,13(1)：10～14.
[26]刘敏华.基于SDG模型的故障诊断及应用研究：[博士学位论文].北京：清华大学,2005.
[27]Hesong Li, Chi Mei, Naijun Zhou, et al. Diagnosis of working conditions of an aluminum reduction cell based on wavelet packets and fuzzy neural network[J]. Chemical Engineering and Processing,2006:1074-1080.
[28]姚恩秀,袁丁,袁工.模糊专家系统在铝电解控制中的应用[J].基础自动化,2000,7(1)：9～11.
[29]李界家,吴成.模糊神经网络技术在铝电解故障诊断中的应用[J].沈阳建筑大学学报(自然科学版),2005,21(4)：390～394.
[30]Li Jie, Xiao Jin, Liu Yexiang et al. Hierarchical intelligent control system for aluminum reduction cells[J]. Light Metals,1997:463～467.
[31]Li Jie, Wang Qianpu, Xiao Jin et al. Intelligent fuzzy control system for aluminum reduction cells[J]. Chinese Journal of Nonferrous Metals,1998:557～562.
[32]王有来,李晋宏,刘卫东等.模糊专家系统技术在大型预焙铝电解槽中的开发与应用[J].北方工业大学学报,200,13(1)：59～62.
[33]冯磊.经典功率谱估计与现代功率谱估计的对比[J].商业文化,2009,5：223～239.
[34]L. Colen. Time-frequency distributions:a review. IEEE. Proc.1989,77(7): 941～981.
[35]Huang N E. The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis. Proc. R. Soc. Lond. A(1998)454:903-995.
[36]Mallat S. A wavelet tour of signal processing[M]. New York:Academic Press, 1998:46～50.
[37]Coifman R.R., Wickerhauser M.V. Entropy-based algorithms for best basis selection. IEEE Trans, on Inf. Theory,1992:713～718.
[38]Peters A.K., Wellesley. The World According to Wavelets. New York:Academic Press,1996:45～52.
[39]Proakis J. G, Manolakis D.G Digital Signal Processing:Principles, Algorithms, and Applications. Englewood Cliffs, NJ:Prentice Hall,1996:18～25.
[40]S.GMallat. A theory for multiresolution signal decomposition:The wavelet representation[J]. IEEE Pattern Anal and Machine Intell,1989,11(7):674～693.
[41]Strang G, Nguyen T. Wavelets and filter banks[M]. Cambridge: Wellesley-Cambridge Press,1996:75～86.
[42]胡昌华,张军波,夏军,张伟等.基于MATLAB的系统分析与设计—小波分析(第一版)[M].西安：电子科技大学出版,1999：48～62.
[43]M Cocchi, R Seeber, A Ulrici,et al. Wavelet packet transform for efficient pattern recognition of signals[J]. Chemometrics and Intelligent Laboratory Systems, 2001,57:97～119.
[44]B Walczak, D L Massart, Wavelet packet transform applied to a set of signal:a new approach to the best-basis selection[J]. Chemometrics and Intelligent Laboratory System,1997,38:39～50.
[45]N Saito, R R Coifman, F B Geshwind, F Warner. Discriminant feature extraction using empirical probability density estimation and a local basis library[J]. Pattern Recognition,2002,35:2841-2852.
[46]Y Wu, R Du, Feature extraction and assessment using wavelet packet for monitoring of machining processes[J]. Mechanical Systems and Signal Processing, 1996,10(1):29～53.
[47]秦树人,季忠,尹爱军.工程信号处理[M].北京：高等教育出版社,2008：456-463.
[48]飞思科技产品研发中心.神经网络理论与MATLAB7实现[M].北京：电子工业出版社,2005：20～23,99～108.
[49]孙建国.铝电解槽阳极电流分布特征及设计上的思考[J].轻金属,2007,3：29～31.
[50]李贺松.大型预焙铝电解非稳态、非均一模型及关键极节能技术研究：[博士学位论文].长沙：中南大学,2005.
[51]N. Urata. Magnetics and metal pad instability[J].Light Metals,1985:581-589.
[52]H.Sun, O.Zikanov, B.A.Finlayson. Effect of background melt flow and interface distortion on the stability of Hall-Heroult cells[J]. Magnetohydrodynamics, 2005,41(3):273-287.
[53]S.J.S.MORRIS, P.A.DAVIDSON. Hydromagnetic edge waves and instability in reduction cells[J]. Journal of Fluid Mechanics,2003,493(2):121-130.
[54]Xiangwen Wang, Alton T Tabereaux. Anodic phenomena-observations of anode overvoltage and gas bubbling during aluminum electrolysis [J]. Light Metals,2000:239-247.
[55]Jilai Xue, A Harald. Bubble behaviour-cell voltage oscillation during aluminium electrolysis and the effects of sound and ultrasound[J] Light Metals,1995:265-271.
[56]Jacques Antille, Rene Von Kaenel. Using a magnetohydrodynamic model to analyze pot stability in order to identify an abnormal operation condition[J]. Light Metals,2002:477-482.
[57]D Shin, A D Sneyd. Metal pad instabilities in aluminium reduction cells[J]. Light Metals,2000:279-283.
[58]P Maillard, M V Romerio. A stability criterion for an in finitely long Hall-Heroult cell[J]. Journal of computational and applied mathematics,1996,71:47-65.
[59]飞思科技产品研发中心.MATLAB7辅助信号处理技术与应用[M].北京：电子工业出版社,2005：319-321.
[60]丁蕾,曾水平,曾铮.350kA铝电解槽槽电阻信号的频谱分析[J].仪器仪表 与检测技术,2005,24(12)：68-70.
[61]曾水平,丁蕾,李晋宏.基于槽电阻频谱分析的铝电解槽槽况诊断[J].中南大学学报(自然科学版),2007,38(1)：768-771.
[62]Huang N E, Shen Z, Long S R. A new view of nonlinear water waves:the Hilbert spectrum. Annu Rev Fluid Mech,1999,31:417-457.
[63]Leisk G,Hsu N, Huang N E. Application of the Hilbert-Huang Transform to machine tool condition/health monitoring. AIP conference, Brunswick, 2002,615(1):1711-1718.
[64]谭善文.多分辨希尔伯特—黄(Hilbert-Huang)变换方法的研究：[博士学位论文].重庆：重庆大学,2001.
[65]廖庆斌,李舜酩,覃小攀.车辆振动信号的特征提取方法比较[J].吉林大学学报(工学版),2007,37(4)：910-914.
[66]李舜酩,李香莲.振动信号的现代分析技术与应用[M].北京：国防工业出版.2008：217-220.
[67]杨建国.小波分析及其工程应用[M].北京：机械工业出版社，2005：68～79.
[68]单峰.大型预焙铝电解槽槽电压波动特征提取：[硕士学位论文].长沙：中南大学,2009.
[69]刘娟花.小波分析在语音去噪中的应用：[硕士学位论文].西安：西安理工大学,2003.
[70]周辉.齿轮故障的特征提取与模式识别技术研究[博士学位论文].郑州：郑州大学,2005：23～24.
[71]张立明.人工神经网络的模型及其应用[M].上海：复旦大学出版社,1993.
[72]周开利,康耀红.神经网络模型及其MATLAB仿真程序设计[M].北京：清华大学出版社,2005：67～83.
[73]Howard Demuth, Mark Beale. Neural network toolbox for use with MATLAB[M]. The Mathworks,2001:16-19.