摘要
铝电解槽是炼铝的主要设备，其能耗巨大，吨铝直流电耗在13000kWh以上，因此如何降低其能耗是一个很有意义的研究课题。本论文建立了槽壳温度在线检测系统，采集到了槽壳温度值，研究了槽壳温度与槽膛内形之间的关系，设计了槽况诊断专家系统，实现了对电解槽能量及物料平衡的监视，达到了节能降耗的目的。
(1)建立了槽壳温度在线检测系统。
槽壳温度在线检测系统由工控机、温度转换模块和温度转换器及热电偶组成。热电偶将温度信息传送至温度转换模块，温度转换模块将模拟量转换为数字量后传送至温度转换器，再由温度转换器传送至工控机，进行信息处理、数据库管理和报警提示等。系统所测温度值一方面以图表形式存储和显示；另一方面作为槽况诊断专家系统进行槽况诊断的依据。
(2)设计了槽膛内形仿真系统。
应用有限容积方法，根据槽壳温度以及相关工艺参数，基于软测量原理，建立了电解槽槽膛内形准三维仿真模型；运用Fortran和Visual Basic语言，开发了电解槽温度场的仿真程序，显示槽膛内形。
(3)开发了槽况诊断专家系统。
根据槽壳温度值、槽膛内形仿真值及电解槽工艺参数，用对象—属性—值三元组模式建立了数据库；根据各种病槽的发生原因、现象及处理方法，用产生式方法建立了规则库；在它们的基础上构建了专家系统，实现了电解槽槽能量及物料平衡的诊断。
本论文的创新点：将温度在线检测和数值仿真技术及人工智能技术三者结合，把槽壳温度检测值和槽膛内形仿真结果作为槽况诊断专家系统对槽况的诊断依据，实现了铝电解槽能量及物料平衡的监视。
半年多的试用表明系统运行稳定可靠，智能化程度高。槽壳温度在线检测系统能实时可靠的检测槽壳温度；槽膛内形仿真系统对槽膛的仿真结果与其测量值误差不超过1cm；诊断专家系统能有效地抑制电解槽病槽的发生，保证电解槽的稳定生产。该课题研究成果的应用使电解槽的经济技术指标明显改善，电流效率由92.8％升高到了93.5％，直流电耗由13715kWh／t-Al降到了13578 kWh／t-Al。
Aluminum cell is the main aluminum producing equipment with so much energy consumption as 13000kWh per ton, so it makes sense to study the method to reduce energy consumption. In this paper, the aluminum cell temperature on-line inspecting system was constructed, shell temperature was collected, relationship between shell temperature and freeze profile was studied, aluminum cell status diagnosing expert system was designed, realizing the inspection of aluminum cell energy and material balance, achieving the object of reducing energy consumption.
(1) The on-line detecting system for shell temperature was constructed.
Shell temperature detecting system consists of industrial computer, temperature transform module, temperature transform apparatus and thermocouple. Temperature signal is transferred to the transform module through thermocouple. After the temperature is transformed into digital signals, the signals are transferred to the temperature transform apparatus and again to the industrial computer through temperature transform apparatus, followed by signal processed, database managed and alarm notice. On one hand, the measured temperature values are stored and displayed as the form of table, and on the other hand, its values are provided as basis for the aluminum cell status diagnosing expert system.
(2) The temperature and freeze profile simulation system was designed.
Based on shell temperature and related technical parameter, finite volume method was used to construct freeze profile quasi three dimension simulation model based on soft measure principle. Fortran and Visual Basic software was used to develop aluminum cell temperature field simulation program displaying the current freeze profile。
(3) The aluminum cell status diagnosing expert system was developed.
According to inspected temperature of shell, the result of freeze profile simulation and aluminum cell technical parameter, the expert system database was built in the mode of object-property-value and stored in tree structure. Based on the cause of its failure, phenomenon and disposal of the diseased aluminum cell, rules for expert system was produced by the fuzzy producing method, and the energy balance diagnosing for aluminum cell was realized.
Some original points in this paper: Based on shell temperature inspected values and freeze profile simulation results, the on-line shell temperature detection, numeric simulation and AI technology were integrated to fulfill the diagnosis for aluminum cell status.
More than half year's application verified its stability and high intelligence. The error between freeze profile simulation result and the measured value was less than 2cm. The diagnosis expert system can effectively prevent malfunction of aluminum cell, ensure its stable production. The technical indexes of aluminum cell were improved with application of our research, with electric efficiency enhanced from 92.8% to 93.5%, and energy consumption decreased from 13715 kWh/t-Al to 13578 kWh/t-Al.
引文
[1] 邱竹贤．预焙槽炼铝(修订版)[M]．北京：冶金工业出版社，1988．2～4
[2] 邱竹贤．铝电解[M]．北京：冶金工业出版社，1993．3～6
[3] 李春喜．调整工艺技术参数、强化电流是提高生产的有效途径[J]．轻金属，2000，(1)：39～40
[4] R.P.Pawlek. Primary Aluminum Smelters of the World: Nameplate Capacities and Shut Down Capacities[J]. Light Metal Age, 1999, 57(1): 8～10
[5] 王捷．电解铝生产工艺与设备[M]．北京：冶金工业出版社，2006．5～9
[6] 殷恩生．160kA中心下料预焙铝电解槽生产工艺及管理[M]．湖南：中南工大出版社，1996．45～51
[7] 徐军，陈学森．我国铝工业现状及今后发展建议[J]．轻金属，2001，(10)：3～6
[8] C. Vanvoren. The Pechiney 500kA Cell[J]. Light Metals, 2001, (5): 221～226
[9] Andrew Hall. Arabal'95-Dubai[J]. Light Metal Age, 1996, 54(2): 52～60
[10] 张丽燕，胡青春．计算铝电解槽内磁场分布的方法研究[J]．青岛大学学报，1996，9(4)：82～88
[11] 梅炽，汤洪清．铝电解槽电、热解析数学模型及数值仿真试验[J]．中南矿冶学院学报，1986，17(6)：29-37
[12] 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]：[硕士学位论文]．长沙：中南工业大学，1988
[13] 游旺．大型预焙铝电解槽槽膛内形在线动态仿真优化[D]：[博士学位论文]．长沙：中南工业大学，1997
[14] 李景江，邱竹贤．铝电解槽槽帮结壳形状的计算机模拟[J]．东北工学院学报，1980，60(3)：232～237
[15] 刘海石．大型预焙槽的槽膛[J]．轻金属，1994，(7)：34～36．
[16] 李德祥，郭天立．贵州铝厂160kA电解槽热场计算[J]．有色冶金，1993，(2)：20～23
[17] D.Mota, G.E.De.Andrade. Magnetic Compensation Project at ALBRAS Smelter[J]. Light Metals, 2001: 413～417
[18] M.V.Romerio, J.Antille. The Numerical Approach to Analysing Flow Stability in the Aluminum Reduction Cell[J]. Aluminum 76, 2000(12): 1031～1037
[19] 梁学民．铝电解槽物理场数学模型及计算机仿真研究[J]．轻金属，1998，增刊：145～149
[20] E. Skybakmoen. Chemical Resistance of Sidelining Materials Based on SiC and Carbon in Cyrolitic Melts[J]. Light Metals, 1999, (2): 215～222
[21] 姚世焕．近代铝电解技术发展与现有企业挖潜改造[J]．铝镁信息，1998，(7)：13～17
[22] 贾志善．增大氟化镁添加量对铝电解生产的影响[J]．轻金属，1983，(4)：28～31．
[23] 刘业翔．互联网上最新轻金属动态[J]．轻金属，2001，(1)：3～5
[24] 李晋宏，冷正旭，席灿明．模型控制和模型专家系统技术在大型预焙铝电解槽中的开发与应用[J]．轻金属，2001，(3)：44～47
[25] R. Bernd. Thermal Bake-out of Aluminum Reduction Cells. Minerals[J], Metals and Materials Society, 2002: 343～346
[26] 邱竹贤，孙淑萍等．中国铝电解工业迈入21世纪的新发展[J]．淄博学院学报(自然科学与工程版)，1999，1(1)：34～37
[27] G.D.Brown, G.J.Hardie, E.W.Shaw. TiB_2 Coated Aluminum Reduction Cells: Status and Future Direction of Coated in Comalco[J]. Proc6th Aust Al Smelting Workshop, 1998: 499～508
[28] T.Palmer. Changing Trends in Aluminum Production. Proc6th Aust Al Smelting Workshop[J], 1998: 1～2
[29] H.A.Oye, B.J.Weelch. Cathode Performance. The Influence of Design, Operations and Operating Conditions[J]. Light Metals, 1998: 18～23
[30] 周乃君．导流型铝电解槽技术进展与应用基础研究[J]．轻金属，2000，(9)：29～31
[31] 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
[32] 中国金属工业“十五”规划(经济论坛)[J]．有色设备，2001，(5)：20～27
[33] 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
[34] Adams J B. A probability model of medical reasoning and the MYCIN model[J]. Maghmatical Biosciences, 1976, (32) : 177～186
[35] Aiello N. A comparative study of control strategies for expert systems: AGE implementation of three variations of PUFF[C]. Proceedings of the National Conference on Artificial intelligence. 1983.
[36] Aiello L, Cecchi C, Sartini D. Representation and use of metaknowledge[J]. Proceedings of IEEE, 1986, (74): 1304～1321.
[37] Aikins J S. Prototypical Knowledge for Expert Systems[J]. Artificial Intelligence, 1983, 20: 163～210
[38] 朱海滨．面向对象技术—原理与设计[M]．长沙：国防科技大学出处版社，1992．106～109
[39] Gilbert N. Explaination and Dialogue[M]. The Knowledge Engineering Review, 1989, (4): 235～247
[40] Bachant J, McDermott J. R1 revisited: four years in the renches[M]. AI Magazine, 1984, Fall issue: 21～32
[41] Barnett J A. Computational methods for a mathematical teory of evidence[M]. International Joint Conference on Artificial Intelligence. 1981. 868～875
[42] Barr A, Feigenbaum E A, ed. The Handbook of Artificial Intelligence[M]. Los Altors CA: Morgan Kaufmann, 1981.45～51
[43] Boose J H. Expertise Transfer for Expert System Design[M]. New York: Elsevier, 1986. 88～90
[44] Boose J H, Gaines B. Knowledge Acquisition Tools for Expert Systems[M]. New York: Academic Press, 1988. 134～138
[45] Brachman R J, Levesque H J. Readings in Knowledge Representation[M]. Los Altos CA: Morgan Kaufmann, 1985. 99～100
[46] Brachman R J, Schmoize J G. An overview of the KL-ONE knowledge representation system[J]. Cognitive Science, 1985, (9): 171～216
[47] Chatalic P, Dubois D, Prade H. An approach to approximate reasoning based on the Dempster rule of combination[J]. Expert Systems, 1987, (1): 67～85
[48] Buxton R. Modelling uncertainty in expert systems[J]. International Journal of Man-Machine Studies, 1989, (31): 415～476
[49] 费宗铭．机器学习[J]．计算机科学，1991，(1)：56～58
[50] 沈清，胡德文，时春．神经网络应用技术[M]．长沙：国防科技大学出处版社，1987．34～35
[51] 张瑞英，陈松乔．锌沸腾焙烧专家系统的研究和开发[J]．中国有色金属学报，1997，7(2)：125～128
[52] 周志栋．连续式加热炉诊断性专家系统的研制[C]．第一届全球华人智能控制与智能自动化大会论文集．北京：ICIA，1993．153～156
[53] 刘日新，刘七新，张晓刚．ADAM模块在热电偶温度采集中的应用[J]．工业炉，2000，22(1)：23～24
[54] 阎鸿，常慧玲．ADAM数据采集网络在过程测控中的应用[J]．冶金自动 化，2000，(4)：43
[55] 席灿明．智能技术在铝电解过程控制中的开发应用[C]．中国有色金属学报第三届学术会议论文集．长沙：中南大学，1997．127～133
[56] 周铁托，张建．大中型预焙铝电解槽自适应控制过程的研究[J]．轻金属，1994，(2)：22～25．
[57] 黄永忠，王化章．铝电解生产[M]．长沙：中南工业大学出版社，1994．78～79
[58] 崔衡，谢刚，陈书荣等．智能控制在铝电解槽中的应用[J]．昆明理工大学学报，2001，26(6)：101～105
[59] TikAsz L. Expert Systems Applied to Control Aluminum Smelters[J]. Light Metals, 1990: 434～449
[60] A.Sannia. An Automation System in Potrooms by Microcomputer Chain[J]. Light Metals, 1980: 351～360
[61] 徐树田．铝电解工艺的自动控制及其问题[J]．轻金属，1985，(2)：43～48
[62] 黄永忠，刘业翔．铝电解槽工艺参数采集系统及工艺参数测试与研究[J]．轻金属，1992，(9)：11～14
[63] 冯乃祥，孙阳等．贵州铝厂160kA大型预焙阳极铝电解槽磁场测量与计算[J]．轻金属，2000，(11)：43～47
[64] Moen T. Adaptive Control of Reduction Cells with Point Feeder[J]. Light Metals, 1985: 459～469
[65] 梅炽，王前普等．有色冶金炉窑仿真与优化[J]．中国有色金属学报，1996，6(2)：56～58
[66] 刘业翔，梅炽，曾水平．80kA上插棒式自焙铝电解槽电解质溶体中电磁力场的计算与分析[J]．中国有色金属学报，1996，6(1)：34～38
[67] 刘业翔，黄永忠等．点式下料铝电解槽氧化铝浓度新型估计模型[J]．中国有色金属学报，1993，3(2)：41～43
[68] Rolland W K. Haldris-an Expert System for Process Control and Supervision of Aluminum Smelters[J]. Light Metals, 1991: 437～443
