铝电解槽极距与铝液波动在线监测系统的研究与应用
|
摘要
随着铝电解槽向大型化发展，铝液波动和极距非均一化问题也越来越突出。因此掌握铝电解槽动态、非均一信息以及信息的可视化变得越来越重要，这将为铝电解槽的进一步节能降耗提供有力依据。
本文以中国铝业广西分公司320kA铝电解槽为研究对象，研究并开发了铝液高度连续测量仪器以及配套的软件(已申请专利)，实现了铝液高度的连续测量。通过对铝液波动、阳极电流、极距等参数的工业实验与测试，分析找出铝液波动、阳极电流、极距三者之间的关系。结果表明阳极导杆等距压降的变化与铝液高度的变化相对应，而阳极气泡对阳极电流波动的影响不大。
根据大量实验分析，选取阳极导杆等距压降、槽电压、阳极系电阻、阳极导杆温度、阳极底掌下垂直电流与阳极导杆电流之比等五个参数作为极距软测量模型的辅助变量。由于这些变量的非线性、强藕合特点，利用matlab6.5的模糊逻辑工具箱与神经网络工具箱建立了基于模糊C-均值聚类的多神经网络极距软测量模型。利用仿真手段拟合出阳极导杆温度、阳极电阻随天数变化的公式；考虑了阳极底掌电流分布问题，利用仿真计算，提出实验槽阳极电流流过阳极倒角与阳极侧部的电流占流过阳极导杆电流的18.7％。同时选择合适的配置安装系统硬件，用delphi7.0开发系统软件，成功实现了极距和铝液界面波动三维图像的在线监测，为铝电解槽的操作优化提供了信息支持。
将系统应用于铝电解槽的操作优化方面，监测分析了铝电解槽有代表性的几种工况，提出了优化思想。以更深一层挖掘铝电解槽节能的潜力为目标，本文提出了以极距和铝液波动在线监测系统为基础的极距均一化思想与实施方案，使铝电解槽在低极距情况下保持磁流体稳定性和较高的电流效率变成可能。
在中国铝业广西分公司004#槽上稳定运行所有系统，在工业实验中降低槽电压40mV，同时保持了槽子的稳定性与电流效率，取得了降低吨铝电耗127kWh的节能效果。
At present, the problem of aluminium wave and asymmetrical ACD(anode-cathode distance) become more and more serious along with the increase of cells' capacity. Therefore, it becomes more and more important, which can provide some basis for energy saving in aluminium industry.
The study in this dissertation was performed on CHLCO. An instrument and its software (which is applying as a patent) were researched and developed to measure molten metal waves. So molten metal waves can be measured continually. The relationship among vertical local oscillations of molten metal, anodic current and anode-cathode distance was found by means of measure and analysis. The corresponding relationship between equal-distant voltage-drop and aluminium wave was found, and anode-bubble has little influence on anode current.
As a great deal of experiment, five pantmeters including anodic equal-distant voltage drops, Anodic resistance, temperature of anodic leader, cell voltage and the percentage of vertical current and anodic current become assistant variables of the oft-measure model. Because the variables are non-linear and correlative, Soft-measure model based on FCM and Multiple Neural Network was set up by Fuzzy Logic Toolbox and Neural Network Toolbox of Matlab6.5. The relationship among Anodic resistance & temperature of anodic leader and numbers of days was expressed by formula. Current distributing of anodic bottom was researched. Because of simulation, the author indicate that the current which don't come through anodic bottom account for 18.7% of the anodic leader current. Then appropriate hardware was chosen to the system, and software of the system was developed with Delphi7.0. Anode-cathode distance(ACD) and three-dimensional of aluminium waves could be visualization online, and it could provide information to operational optimization to cells.
The system was running for optimizing operation, and some representative cases was analysed. A thought make ACD equality which based on the system and its implement ways were developed. So, the cells can steadily work in low ACD and keep magnetohydrodynamic (MHD) steady and high current efficiency.
The system was steady-going in Guangxi Filiale 320kA cell of CHLCO. During the experiment, cell voltage reduced to 4.17V, at the same time the cell was well-balanced. In the conclusion, electrical consumption went down 127kWh/t.
本文以中国铝业广西分公司320kA铝电解槽为研究对象，研究并开发了铝液高度连续测量仪器以及配套的软件(已申请专利)，实现了铝液高度的连续测量。通过对铝液波动、阳极电流、极距等参数的工业实验与测试，分析找出铝液波动、阳极电流、极距三者之间的关系。结果表明阳极导杆等距压降的变化与铝液高度的变化相对应，而阳极气泡对阳极电流波动的影响不大。
根据大量实验分析，选取阳极导杆等距压降、槽电压、阳极系电阻、阳极导杆温度、阳极底掌下垂直电流与阳极导杆电流之比等五个参数作为极距软测量模型的辅助变量。由于这些变量的非线性、强藕合特点，利用matlab6.5的模糊逻辑工具箱与神经网络工具箱建立了基于模糊C-均值聚类的多神经网络极距软测量模型。利用仿真手段拟合出阳极导杆温度、阳极电阻随天数变化的公式；考虑了阳极底掌电流分布问题，利用仿真计算，提出实验槽阳极电流流过阳极倒角与阳极侧部的电流占流过阳极导杆电流的18.7％。同时选择合适的配置安装系统硬件，用delphi7.0开发系统软件，成功实现了极距和铝液界面波动三维图像的在线监测，为铝电解槽的操作优化提供了信息支持。
将系统应用于铝电解槽的操作优化方面，监测分析了铝电解槽有代表性的几种工况，提出了优化思想。以更深一层挖掘铝电解槽节能的潜力为目标，本文提出了以极距和铝液波动在线监测系统为基础的极距均一化思想与实施方案，使铝电解槽在低极距情况下保持磁流体稳定性和较高的电流效率变成可能。
在中国铝业广西分公司004#槽上稳定运行所有系统，在工业实验中降低槽电压40mV，同时保持了槽子的稳定性与电流效率，取得了降低吨铝电耗127kWh的节能效果。
At present, the problem of aluminium wave and asymmetrical ACD(anode-cathode distance) become more and more serious along with the increase of cells' capacity. Therefore, it becomes more and more important, which can provide some basis for energy saving in aluminium industry.
The study in this dissertation was performed on CHLCO. An instrument and its software (which is applying as a patent) were researched and developed to measure molten metal waves. So molten metal waves can be measured continually. The relationship among vertical local oscillations of molten metal, anodic current and anode-cathode distance was found by means of measure and analysis. The corresponding relationship between equal-distant voltage-drop and aluminium wave was found, and anode-bubble has little influence on anode current.
As a great deal of experiment, five pantmeters including anodic equal-distant voltage drops, Anodic resistance, temperature of anodic leader, cell voltage and the percentage of vertical current and anodic current become assistant variables of the oft-measure model. Because the variables are non-linear and correlative, Soft-measure model based on FCM and Multiple Neural Network was set up by Fuzzy Logic Toolbox and Neural Network Toolbox of Matlab6.5. The relationship among Anodic resistance & temperature of anodic leader and numbers of days was expressed by formula. Current distributing of anodic bottom was researched. Because of simulation, the author indicate that the current which don't come through anodic bottom account for 18.7% of the anodic leader current. Then appropriate hardware was chosen to the system, and software of the system was developed with Delphi7.0. Anode-cathode distance(ACD) and three-dimensional of aluminium waves could be visualization online, and it could provide information to operational optimization to cells.
The system was running for optimizing operation, and some representative cases was analysed. A thought make ACD equality which based on the system and its implement ways were developed. So, the cells can steadily work in low ACD and keep magnetohydrodynamic (MHD) steady and high current efficiency.
The system was steady-going in Guangxi Filiale 320kA cell of CHLCO. During the experiment, cell voltage reduced to 4.17V, at the same time the cell was well-balanced. In the conclusion, electrical consumption went down 127kWh/t.
引文
[1] 陈远望．美国铝电解槽技术革新新进展[J]．世界有色金属，2004，9：50～52
[2] 罗天骄，魏昶，黄孟阳，等．世界铝工业的现状与发展[J]．云南冶金，2004，33(1)：42～46
[3] 狄贵华，任剑．浅析国内外铝电解能耗状况[J]．有色冶金节能，2004，21(4)：41～43
[4] 吕定雄，干益人，毛继红，等．我国开发大容量预焙铝电解槽技术进展及今后努力方向[J]．有色设备，2000，6：6～11
[5] 陆钦芳，金秀岩．关于我国铝电解工业技术发展路线的建议[J]．世界有色金属，2001，2：10～13
[6] 陈廷贵，代国才，姚世焕．300kA以上大型铝电解槽的设计理念[J]．轻金属，2003，12：19～26
[7] 刘海石．300kA大型预焙槽的优化[J]．世界有色金属，2004，9：50～52
[8] 陈颖．350kA大型预焙阳极铝电解槽优化设计论述[J]．有色金属设计，2004，31(1)：28～33
[9] 邱竹贤．世界铝工业与新技术发展趋势．有色冶炼，2000，29(2)：1～7
[10] 邱竹贤．中国铝工业应用新型电极材料的研究与展望[J]．中国工程科学，2001，3(5)：50～54
[11] 刘业翔．铝电解惰性阳极与可湿润性阴极的研究与开发进展[J]．轻金属，2001，5：26～29
[12] 邱竹贤．21世纪伊始铝电解工业的新进展[J]．中国工程科学，2003，5(4)：41～46
[13] 周卫铭，郭忠诚．铝电解惰性阳极材料的研究现状[J]．冶金丛刊，2004，1：1～3
[14] 李劼，李相鹏，赖延清，等．导流槽阴极导杆安装方式对阴极表面电流分布的影响[J]．中国有色金属学报，2003，13(4)：1017～1020
[15] ZHOU Nai-jun, MEI Chi, JIANG Chang-wei. A method of determining and designing the drained slope in drained aluminum reduction cells[J]. J.CENT.SOUTH UNIV.TECHNOL., 2003, 10(1): 74～77
[16] LIU Ye-xiang, LI Xiang-peng, LAI Yan-qing, et al. Heat balance simulation of drained aluminum reduction cell[J]. Trans. Nonferrous Met. Soc. China, 2003, 13 (5): 1199～1202
[17] LI Xiang-peng, LI Jie, LAI Yan-qing, et al. Freeze profile and heat balance calculation of the 160kA drained cell[J]. ACTA METALLURGICA SINICA (ENGLISH LETTERS), 2004, 17 (2): 215～220
[18] 邱竹贤，何鸣鸿，范立满．融盐铝电解中若干物理化学问题的研究[J]．东北大学学报，2001，22(2)：119～122
[19] 田忠良，赖延清，银瑰，等．低温铝电解研究进展[J]．有色金属(冶炼部分)，2004，5：26～28
[20] 苟护生，左良．我国铝工业发展的方针、目标及建设重点[J]．轻金属，2004，7：3～5
[21] 冯乃祥，田福泉，等．我国铝电解工业现状和与国外先进技术水平的差距[J]．轻金属，2000，7：29～33
[22] 伍进伟．我国电解铝工业国际竞争力的分析[J]．有色金属设计，2003，30(3)：6～10
[23] 张稳．浅谈我国电解铝工业的现状[J]．有色金属设计，2004，31(2)：66～68
[24] Geir M Haarberg, Karen S Osen, Jomar Thonstad. Measurement of electronic conduction in cryolite-alumina melts and estimation of its effect on current efficiency[J]. Light Metals, 1991: 283-287.
[25] Jacques Antille, Rene von Kaenel. Busbar optimization using cell stability criteria and its impact on cell performance[J]. Light Metals, 1999: 165-170.
[26] M F El-Demerdash, A A Adly, S E Abu-Shady et al. Towards a more stable aluminum cell via busbar configuration optimization[J]. Light Metals, 2000: 291-295.
[27] Pierre Homsi, Jean Michel Peyneau, Michel Recerdy. Overview of process control in reduction cells and potlines[J]. Light Metals, 2000: 223-229.
[28] Halvor Kvande, Bjorn P Moxnes, Jorn Skaar et al. Pseudo resistance curves for aluminium cell control-alumina dissolution and cell dynamics[j]. Light Metals, 1997: 403-409.
[29] Aureliu Panaitescu, Augustin Moraru. Research on the instabilities in the aluminum electrolysis cell[J]. Light Metals, 2003: 359-365.
[30] Fiona J Stevens, Weidong Zhang, Mark P Taylor et al. The interaction between current efficiency and energy balance in aluminium[J]. Light Metals, 1992: 541-547.
[31] A. Moraru, A. Panaitescu, and A. Crisu. Current Field In An Aluminum Electrolysis cell[J]. Light Metals, 2005: 469-476.
[32] M Guden, I Karakaya. Electrolysis of MgCl_2 with a top inserted anode and an Mg-Pb cathode[J]. Journal of Applied Electrochemistry, 1994, 24(8): 791-797.
[33] R C Dorward. Energy consumption of aluminium smelting cells containing solid wetted cathodes[J]. Journal of Applied Electrochemistry, 1983, 13(5): 569-575
[34] R G Haverkamp. Eliminating anode effects[J]. Light Metals, 1999: 285-288.
[35] R G Haverkamp. Towards zero anode effect. The 6th Australasian Aluminium Smelting Technology Conference and Workshop [J]. Queenstown. New Zealand, 1998: 267-274.
[36] R G Haverkamp. Removal of the anode effect [J]. Aluminum Transactions (USA), 2000, 2(1): 169-174.
[37] I Mantha. Reduction of anode effect frequency in Arvida prebaked smelter[J]. Light Metals, 2001: 25-33.
[38] S Rolseth, T Muftuoglu, A Solheim. Current efficiency at short anode--cathode distance in aluminum electrolysis[J]. Light Metals, 1986: 517-523.
[39] 谢叶明．155kA预焙槽针振的危害及处理对策[J]．中国有色冶金，2004(4)：78-79．
[40] 卢琳．240kA电解槽电压针振探讨[J]．有色金属，2004(5)：22-25
[41] 王绍鹏，马松堂．280kA电解槽针振初探[J]．轻金属，2002(12)：31-34．
[42] Jean Frederic, Tony Lelievre, Claude LE. Metal pad roll instabilities[J]. Light Metal, 2002: 483-487.
[43] Wu Jiangkang, Huang Ming, Huang Jun et al. Finite element analysis of incompressible viscous flow with moving free surface by selective volume of fluid method[J]. Light Metals, 2002: 511-514.
[44] 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.
[45] 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.
[46] A F LaCamera, D P Ziegler, R L Kozarek. Magnetohydrodynamics in the Hall-Heroult process, an overview [J]. Light Metals, 1992: 1179-1186.
[47] 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.
[48] Improved 2-dimensional model for magnetohydrodynamic stability analysis in reduction cells [J]. Light Metals, 1998: 419-428.
[49] M Segatz, Ch Droste, D Vogelsang. Magnetohydrodynamic effect of anode set pattern on cell performance[J]. Light Metals, 1997: 429-435.
[50] M. Chiampi, M. Repetto, etc. Magnetic modeling and magneto-hydro- dynamic simulations of an aluminuium production electrolytic cell[J]. The International Journal for Computation and Mathematics in Electrical and Electronic Engineering, 1999, 18(3): 528-538.
[51] 吴进康，李光正．铝电解槽磁液体稳定性研究[J]．力学进展．Vol32(2)：275-283．
[52] 梁汉．铝电解槽磁流体稳定性判据[J]．轻金属，2005，8：40-45．
[53] 曾水平，张秋萍．预焙铝电解槽电流效率与阳极电流分布的数学模型[J]．中国有色金属学报，2004，14(4)：681-685．
[54] J Descloux, M V Romerio. On the analysis by perturbation method of the anodic current fluctuation in an electrolytic cell for aluminium [J]. Light Metals, 1989: 237-243.
[55] J Antille, M Flueck, M V Romerio. Steady velocity field in aluminium reduction cells derived from measurements of the anodic current fluctuations[J]. Light Metals, 1994: 305-312.
[56] Aureliu Panaitescu, Augustin Moraru. Research on the instabilities in the aluminum electrolysis cell. Light Metals, 2003: 359-365.
[57] Aureliu Panaitescu, Augustin Moraru. Analysis of the phenomena at the start-up of an aluminum electrolysis cell bu the visualization of the current distributions [J]. Light Metals, 2002: 389-396.
[58] B D Ovsyannikov, M P Lipinskii, A M Tsyplakov. frequency method for estimating the anode-cathode distance in aluminum electrolysis cells[J]. Tsvetn. Met, 1984, 2: 36-39.
[59] 铁军，韩至成，邱竹贤，等．铝电解中阳极气泡形成的电化学研究[J]．有色金属，1996，48(1)：44-48．
[60] 路忠胜，白勇鹏，江波．降低铝用炭阳极净消耗对铝电解参数影响的理论和与实践[J]．炭素，2004，3：7-10．
[61] 杨世铭．传热学[M]．北京：高等教育出版社，1987
[62] 中华人民共和国有色金属行业标准．YS／T 120—92．铝电解槽能量平衡测试与计算方法．
[63] 梅炽．有色冶金炉窑仿真与优化[M]．北京．冶金工业出版社，2001
[64] 梅炽，王前普．铝电解槽热场研究[J]．轻金属，1992(1)：29～32
[65] 吴乐谋．铝电解槽热场研究——熔体与槽帮之间传热系数的计算及研究[D]．中南工业大学硕士论文，1988，6
[66] 李德祥，郭天立．铝电解槽内熔体与槽帮间换热系数计算[J]．有色金属，1993：45
[67] D. W. Dow, W.H. Goodnow. Influence of Operating Variables on Reduction Cell Bath Temperature[J]. Light Metals. 1972: 246～251
[68] 殷恩生．160kA中心下料预焙铝电解槽生产工艺及管理[M]．长沙：中南工业大学出版社．
[69] 农国武．铝电解槽尝试控制论域自调整的开发应用[J]．轻金属，2004，11：21-25．
[70] 高琳，高峰，管晓宏等．电力系统短期负荷预测的多神经网络Boosting集成模型．西安交通大学学报，2004，38(10)：1026-1030．
[71] 潘立登，朱宇宁．多神经网络在软测量仪表中的应用．北京化工大学学报，2001，28(1)：67-69．
[72] 韦巍，蒋静坪．基于多神经网络的机器人轨迹学习控制研究．浙江大学学报，．1997，31(4)：505-511．
[73] 臧朝平，韩芳，张思等．基于多神经网络多参数综合的旋转机械故障诊断系统研究．振动与冲击，1997，16(4)：65-68．
[74] 何耀华，韩守木，程尚模．基于多神经网络协同推理的故障诊断系统的研制．中国电机工程学报，1999，19(12)：57-60．
[75] 车录锋，周晓军，程耀东．基于多神经网络模糊积分集成的无损检测缺陷分类．机械科学与技术，2000，19(1)：111-112．
[76] Lalu Mangal, V G. Idichandy, C Ganapathy. ART-based multiple neural networks for monitoring offshore platforms. Applied Ocean Research, 1996, 18(2): 137-143.
[77] Zhang Jie. Mproved on-line process fault diagnosis through information fusion in multiple neural networks. Computers and Chemical Engineering, 2006, 30(3): 558-571.
[78] 李树涛，王耀南，毛建旭．基于多神经网络分类器组合的火焰图像分割．数据采集与处理，2000，15(4)：443-446．
[79] 章新华，林良骥，王骥程．基于多神经网络融合的声纳目标分类．控制与决策，1997，12(4)：381-384．
[80] 陈芳信，马翔，姜新等．基于多神经网络的手写数字综合识别系统．信息技术，2003，27(9)：1-4．
[81] 郑文波，林雅梅．无限制性手写数字的多神经网络识别．福州大学学报，1998，26(2)：18-21．
[82] 曲雨水，黄德先，金以慧．基于多神经网络的发酵过程菌丝浓度估计．计算机工程与应用，2004，10：208-210．
[83] Anna K Jerebko; James D Malley, Marek Franaszek. Multiple neural network classification scheme for detection of colonic polyps in CT colonography data sets. Academic Radiology, 2003, 10(2): 154-160.
[84] Pottmann M, Unbehauen H, Seborg D E. Application of a general multi-model approach for identification of highly nonlinear processed: A case study[J]. Int J Control, 1993, 57(1): 97-120.
[85] 李柠，李少远，席裕庚．利用模糊满意聚类建立pH中和过程模型[J]．控制理论与应用．2002，17(2)：143-147．
[86] Zadeh L A. Fuzzy sets[J]. Inf Cont, 1965, (8): 338-353.
[87] 高新波，谢维信．模糊聚类理论发展及应用的研究进展[J]．科学通报，1999，44(21)：41．
[88] 周叙国，邢汉承．基于遗传模糊C-均值聚类算法的地图分割．计算机与数字工程，2005，33(6)：77-78．
[89] 李茂宽，关键．基于模糊C均值的支持向量机数据分类识别．系统仿真学报，2005，17(7)：1785-1787．
[90] Chen, Weijie; Giger, Maryellen L.; Bick, Ulrich. A Fuzzy C-Means (FCM)-Based Approach for Computerized Segmentation of Breast Lesions in Dynamic Contrast-Enhanced MR Images, 2006, 13(1): 63-72.
[91] Chuang, Keh-Shih; Tzeng, Hong-Long; Chen, Sharon; Fuzzy c-means clustering with spatial information for image segmentation, 2006, 30(1): 9-15.
[92] Noordam J C, van den Broek W H AM, Buydens L M C. Multivariate image segmentation with cluster size insensitive Fuzzy C-means. Chemometrics and Intelligent Laboratory Systems, 2002, 23(5): 65-78.
[93] Lo, Chi-Chun, Wang, Shuenn-Jyi. Video segmentation using a histogram-besed fuzzy c-means clustering algorithm. Computer Standards and Interfaces, 2001, 23(5): 429-438.
[94] Zhang, Dao-Qiang; Chen, Song-Can. A novel kernelized fuzzy C-means algorithm with application in medical image segmentation. Artificial Intelligence in Medicine, 2004, 115(1): 261-279.
[95] Pianykh, S Oleg. Analytically tractable case of fuzzy c-means clustering. Pattern Recognition, 2006, 39(1): 35-46.
[96] 赵雪红，张来斌，樊建春．基于模糊C-均值聚类算法的柴油机磨损状态评判．润滑与密封，2005，2：23-25．
[97] Teppola, Pekka; Mujunen, Satu-Pia; Minkkinen, Pentti. Adaptive Fuzzy C-Means clustering in process monitoring. Chemometrics and Intelligent Laboratory Systems, 1999, 45(1): 23-28.
[2] 罗天骄，魏昶，黄孟阳，等．世界铝工业的现状与发展[J]．云南冶金，2004，33(1)：42～46
[3] 狄贵华，任剑．浅析国内外铝电解能耗状况[J]．有色冶金节能，2004，21(4)：41～43
[4] 吕定雄，干益人，毛继红，等．我国开发大容量预焙铝电解槽技术进展及今后努力方向[J]．有色设备，2000，6：6～11
[5] 陆钦芳，金秀岩．关于我国铝电解工业技术发展路线的建议[J]．世界有色金属，2001，2：10～13
[6] 陈廷贵，代国才，姚世焕．300kA以上大型铝电解槽的设计理念[J]．轻金属，2003，12：19～26
[7] 刘海石．300kA大型预焙槽的优化[J]．世界有色金属，2004，9：50～52
[8] 陈颖．350kA大型预焙阳极铝电解槽优化设计论述[J]．有色金属设计，2004，31(1)：28～33
[9] 邱竹贤．世界铝工业与新技术发展趋势．有色冶炼，2000，29(2)：1～7
[10] 邱竹贤．中国铝工业应用新型电极材料的研究与展望[J]．中国工程科学，2001，3(5)：50～54
[11] 刘业翔．铝电解惰性阳极与可湿润性阴极的研究与开发进展[J]．轻金属，2001，5：26～29
[12] 邱竹贤．21世纪伊始铝电解工业的新进展[J]．中国工程科学，2003，5(4)：41～46
[13] 周卫铭，郭忠诚．铝电解惰性阳极材料的研究现状[J]．冶金丛刊，2004，1：1～3
[14] 李劼，李相鹏，赖延清，等．导流槽阴极导杆安装方式对阴极表面电流分布的影响[J]．中国有色金属学报，2003，13(4)：1017～1020
[15] ZHOU Nai-jun, MEI Chi, JIANG Chang-wei. A method of determining and designing the drained slope in drained aluminum reduction cells[J]. J.CENT.SOUTH UNIV.TECHNOL., 2003, 10(1): 74～77
[16] LIU Ye-xiang, LI Xiang-peng, LAI Yan-qing, et al. Heat balance simulation of drained aluminum reduction cell[J]. Trans. Nonferrous Met. Soc. China, 2003, 13 (5): 1199～1202
[17] LI Xiang-peng, LI Jie, LAI Yan-qing, et al. Freeze profile and heat balance calculation of the 160kA drained cell[J]. ACTA METALLURGICA SINICA (ENGLISH LETTERS), 2004, 17 (2): 215～220
[18] 邱竹贤，何鸣鸿，范立满．融盐铝电解中若干物理化学问题的研究[J]．东北大学学报，2001，22(2)：119～122
[19] 田忠良，赖延清，银瑰，等．低温铝电解研究进展[J]．有色金属(冶炼部分)，2004，5：26～28
[20] 苟护生，左良．我国铝工业发展的方针、目标及建设重点[J]．轻金属，2004，7：3～5
[21] 冯乃祥，田福泉，等．我国铝电解工业现状和与国外先进技术水平的差距[J]．轻金属，2000，7：29～33
[22] 伍进伟．我国电解铝工业国际竞争力的分析[J]．有色金属设计，2003，30(3)：6～10
[23] 张稳．浅谈我国电解铝工业的现状[J]．有色金属设计，2004，31(2)：66～68
[24] Geir M Haarberg, Karen S Osen, Jomar Thonstad. Measurement of electronic conduction in cryolite-alumina melts and estimation of its effect on current efficiency[J]. Light Metals, 1991: 283-287.
[25] Jacques Antille, Rene von Kaenel. Busbar optimization using cell stability criteria and its impact on cell performance[J]. Light Metals, 1999: 165-170.
[26] M F El-Demerdash, A A Adly, S E Abu-Shady et al. Towards a more stable aluminum cell via busbar configuration optimization[J]. Light Metals, 2000: 291-295.
[27] Pierre Homsi, Jean Michel Peyneau, Michel Recerdy. Overview of process control in reduction cells and potlines[J]. Light Metals, 2000: 223-229.
[28] Halvor Kvande, Bjorn P Moxnes, Jorn Skaar et al. Pseudo resistance curves for aluminium cell control-alumina dissolution and cell dynamics[j]. Light Metals, 1997: 403-409.
[29] Aureliu Panaitescu, Augustin Moraru. Research on the instabilities in the aluminum electrolysis cell[J]. Light Metals, 2003: 359-365.
[30] Fiona J Stevens, Weidong Zhang, Mark P Taylor et al. The interaction between current efficiency and energy balance in aluminium[J]. Light Metals, 1992: 541-547.
[31] A. Moraru, A. Panaitescu, and A. Crisu. Current Field In An Aluminum Electrolysis cell[J]. Light Metals, 2005: 469-476.
[32] M Guden, I Karakaya. Electrolysis of MgCl_2 with a top inserted anode and an Mg-Pb cathode[J]. Journal of Applied Electrochemistry, 1994, 24(8): 791-797.
[33] R C Dorward. Energy consumption of aluminium smelting cells containing solid wetted cathodes[J]. Journal of Applied Electrochemistry, 1983, 13(5): 569-575
[34] R G Haverkamp. Eliminating anode effects[J]. Light Metals, 1999: 285-288.
[35] R G Haverkamp. Towards zero anode effect. The 6th Australasian Aluminium Smelting Technology Conference and Workshop [J]. Queenstown. New Zealand, 1998: 267-274.
[36] R G Haverkamp. Removal of the anode effect [J]. Aluminum Transactions (USA), 2000, 2(1): 169-174.
[37] I Mantha. Reduction of anode effect frequency in Arvida prebaked smelter[J]. Light Metals, 2001: 25-33.
[38] S Rolseth, T Muftuoglu, A Solheim. Current efficiency at short anode--cathode distance in aluminum electrolysis[J]. Light Metals, 1986: 517-523.
[39] 谢叶明．155kA预焙槽针振的危害及处理对策[J]．中国有色冶金，2004(4)：78-79．
[40] 卢琳．240kA电解槽电压针振探讨[J]．有色金属，2004(5)：22-25
[41] 王绍鹏，马松堂．280kA电解槽针振初探[J]．轻金属，2002(12)：31-34．
[42] Jean Frederic, Tony Lelievre, Claude LE. Metal pad roll instabilities[J]. Light Metal, 2002: 483-487.
[43] Wu Jiangkang, Huang Ming, Huang Jun et al. Finite element analysis of incompressible viscous flow with moving free surface by selective volume of fluid method[J]. Light Metals, 2002: 511-514.
[44] 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.
[45] 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.
[46] A F LaCamera, D P Ziegler, R L Kozarek. Magnetohydrodynamics in the Hall-Heroult process, an overview [J]. Light Metals, 1992: 1179-1186.
[47] 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.
[48] Improved 2-dimensional model for magnetohydrodynamic stability analysis in reduction cells [J]. Light Metals, 1998: 419-428.
[49] M Segatz, Ch Droste, D Vogelsang. Magnetohydrodynamic effect of anode set pattern on cell performance[J]. Light Metals, 1997: 429-435.
[50] M. Chiampi, M. Repetto, etc. Magnetic modeling and magneto-hydro- dynamic simulations of an aluminuium production electrolytic cell[J]. The International Journal for Computation and Mathematics in Electrical and Electronic Engineering, 1999, 18(3): 528-538.
[51] 吴进康，李光正．铝电解槽磁液体稳定性研究[J]．力学进展．Vol32(2)：275-283．
[52] 梁汉．铝电解槽磁流体稳定性判据[J]．轻金属，2005，8：40-45．
[53] 曾水平，张秋萍．预焙铝电解槽电流效率与阳极电流分布的数学模型[J]．中国有色金属学报，2004，14(4)：681-685．
[54] J Descloux, M V Romerio. On the analysis by perturbation method of the anodic current fluctuation in an electrolytic cell for aluminium [J]. Light Metals, 1989: 237-243.
[55] J Antille, M Flueck, M V Romerio. Steady velocity field in aluminium reduction cells derived from measurements of the anodic current fluctuations[J]. Light Metals, 1994: 305-312.
[56] Aureliu Panaitescu, Augustin Moraru. Research on the instabilities in the aluminum electrolysis cell. Light Metals, 2003: 359-365.
[57] Aureliu Panaitescu, Augustin Moraru. Analysis of the phenomena at the start-up of an aluminum electrolysis cell bu the visualization of the current distributions [J]. Light Metals, 2002: 389-396.
[58] B D Ovsyannikov, M P Lipinskii, A M Tsyplakov. frequency method for estimating the anode-cathode distance in aluminum electrolysis cells[J]. Tsvetn. Met, 1984, 2: 36-39.
[59] 铁军，韩至成，邱竹贤，等．铝电解中阳极气泡形成的电化学研究[J]．有色金属，1996，48(1)：44-48．
[60] 路忠胜，白勇鹏，江波．降低铝用炭阳极净消耗对铝电解参数影响的理论和与实践[J]．炭素，2004，3：7-10．
[61] 杨世铭．传热学[M]．北京：高等教育出版社，1987
[62] 中华人民共和国有色金属行业标准．YS／T 120—92．铝电解槽能量平衡测试与计算方法．
[63] 梅炽．有色冶金炉窑仿真与优化[M]．北京．冶金工业出版社，2001
[64] 梅炽，王前普．铝电解槽热场研究[J]．轻金属，1992(1)：29～32
[65] 吴乐谋．铝电解槽热场研究——熔体与槽帮之间传热系数的计算及研究[D]．中南工业大学硕士论文，1988，6
[66] 李德祥，郭天立．铝电解槽内熔体与槽帮间换热系数计算[J]．有色金属，1993：45
[67] D. W. Dow, W.H. Goodnow. Influence of Operating Variables on Reduction Cell Bath Temperature[J]. Light Metals. 1972: 246～251
[68] 殷恩生．160kA中心下料预焙铝电解槽生产工艺及管理[M]．长沙：中南工业大学出版社．
[69] 农国武．铝电解槽尝试控制论域自调整的开发应用[J]．轻金属，2004，11：21-25．
[70] 高琳，高峰，管晓宏等．电力系统短期负荷预测的多神经网络Boosting集成模型．西安交通大学学报，2004，38(10)：1026-1030．
[71] 潘立登，朱宇宁．多神经网络在软测量仪表中的应用．北京化工大学学报，2001，28(1)：67-69．
[72] 韦巍，蒋静坪．基于多神经网络的机器人轨迹学习控制研究．浙江大学学报，．1997，31(4)：505-511．
[73] 臧朝平，韩芳，张思等．基于多神经网络多参数综合的旋转机械故障诊断系统研究．振动与冲击，1997，16(4)：65-68．
[74] 何耀华，韩守木，程尚模．基于多神经网络协同推理的故障诊断系统的研制．中国电机工程学报，1999，19(12)：57-60．
[75] 车录锋，周晓军，程耀东．基于多神经网络模糊积分集成的无损检测缺陷分类．机械科学与技术，2000，19(1)：111-112．
[76] Lalu Mangal, V G. Idichandy, C Ganapathy. ART-based multiple neural networks for monitoring offshore platforms. Applied Ocean Research, 1996, 18(2): 137-143.
[77] Zhang Jie. Mproved on-line process fault diagnosis through information fusion in multiple neural networks. Computers and Chemical Engineering, 2006, 30(3): 558-571.
[78] 李树涛，王耀南，毛建旭．基于多神经网络分类器组合的火焰图像分割．数据采集与处理，2000，15(4)：443-446．
[79] 章新华，林良骥，王骥程．基于多神经网络融合的声纳目标分类．控制与决策，1997，12(4)：381-384．
[80] 陈芳信，马翔，姜新等．基于多神经网络的手写数字综合识别系统．信息技术，2003，27(9)：1-4．
[81] 郑文波，林雅梅．无限制性手写数字的多神经网络识别．福州大学学报，1998，26(2)：18-21．
[82] 曲雨水，黄德先，金以慧．基于多神经网络的发酵过程菌丝浓度估计．计算机工程与应用，2004，10：208-210．
[83] Anna K Jerebko; James D Malley, Marek Franaszek. Multiple neural network classification scheme for detection of colonic polyps in CT colonography data sets. Academic Radiology, 2003, 10(2): 154-160.
[84] Pottmann M, Unbehauen H, Seborg D E. Application of a general multi-model approach for identification of highly nonlinear processed: A case study[J]. Int J Control, 1993, 57(1): 97-120.
[85] 李柠，李少远，席裕庚．利用模糊满意聚类建立pH中和过程模型[J]．控制理论与应用．2002，17(2)：143-147．
[86] Zadeh L A. Fuzzy sets[J]. Inf Cont, 1965, (8): 338-353.
[87] 高新波，谢维信．模糊聚类理论发展及应用的研究进展[J]．科学通报，1999，44(21)：41．
[88] 周叙国，邢汉承．基于遗传模糊C-均值聚类算法的地图分割．计算机与数字工程，2005，33(6)：77-78．
[89] 李茂宽，关键．基于模糊C均值的支持向量机数据分类识别．系统仿真学报，2005，17(7)：1785-1787．
[90] Chen, Weijie; Giger, Maryellen L.; Bick, Ulrich. A Fuzzy C-Means (FCM)-Based Approach for Computerized Segmentation of Breast Lesions in Dynamic Contrast-Enhanced MR Images, 2006, 13(1): 63-72.
[91] Chuang, Keh-Shih; Tzeng, Hong-Long; Chen, Sharon; Fuzzy c-means clustering with spatial information for image segmentation, 2006, 30(1): 9-15.
[92] Noordam J C, van den Broek W H AM, Buydens L M C. Multivariate image segmentation with cluster size insensitive Fuzzy C-means. Chemometrics and Intelligent Laboratory Systems, 2002, 23(5): 65-78.
[93] Lo, Chi-Chun, Wang, Shuenn-Jyi. Video segmentation using a histogram-besed fuzzy c-means clustering algorithm. Computer Standards and Interfaces, 2001, 23(5): 429-438.
[94] Zhang, Dao-Qiang; Chen, Song-Can. A novel kernelized fuzzy C-means algorithm with application in medical image segmentation. Artificial Intelligence in Medicine, 2004, 115(1): 261-279.
[95] Pianykh, S Oleg. Analytically tractable case of fuzzy c-means clustering. Pattern Recognition, 2006, 39(1): 35-46.
[96] 赵雪红，张来斌，樊建春．基于模糊C-均值聚类算法的柴油机磨损状态评判．润滑与密封，2005，2：23-25．
[97] Teppola, Pekka; Mujunen, Satu-Pia; Minkkinen, Pentti. Adaptive Fuzzy C-Means clustering in process monitoring. Chemometrics and Intelligent Laboratory Systems, 1999, 45(1): 23-28.