基于广义回归网络的铝电解阳极效应预报
|
摘要
应用广义回归神经网络对当前预焙槽铝电解阳极效应预报问题进行了研究。在简述广义回归神经网络的基本结构基础上,利用广义回归神经网络对铝电解槽阳极效应进行系统辨识建模。重点探讨了建模过程中模型样本结构的选择,实验分析了样本容量对模型预报准确率的影响。取自某铝厂400 k A大型预焙槽的单槽运行现场数据样本对模型进行训练和检验,结果表明该方法阳极效应预报准确率平均在90%以上,预报提前量可以达到半个小时。现场多台电解槽的建模测试结果进一步论证了该模型和样本结构的合理性和有效性,由此证实该方法在保证较高预报准确率同时,具有较好的普适性。
This paper aims to research the anode effect(AE) prediction of pre-baked aluminum electrolysis cell with the generalized regression neural network(GRNN). The structures and advantages of GRNN are introduced, then the anode effect system of aluminum electrolysis cell is modeled by the method of system identification based on GRNN. The structure of samples is analyzed emphatically in the process of modeling, and the influence of sample size to model prediction accuracy is analyzed by experiments. The AE model based on GRNN is trained and tested by sufficient samples which are extracted from the production data of the 400 k A aluminum electrolysis cell. It's proved that the accuracy rate of the AE prediction is more than 90% on average while predicting AE half an hour before the AE moment through this method. Results from experiments with sample data from different cells show that this analytic method is logical and effective, and has extensive applicability while keeping high prediction accuracy.
This paper aims to research the anode effect(AE) prediction of pre-baked aluminum electrolysis cell with the generalized regression neural network(GRNN). The structures and advantages of GRNN are introduced, then the anode effect system of aluminum electrolysis cell is modeled by the method of system identification based on GRNN. The structure of samples is analyzed emphatically in the process of modeling, and the influence of sample size to model prediction accuracy is analyzed by experiments. The AE model based on GRNN is trained and tested by sufficient samples which are extracted from the production data of the 400 k A aluminum electrolysis cell. It's proved that the accuracy rate of the AE prediction is more than 90% on average while predicting AE half an hour before the AE moment through this method. Results from experiments with sample data from different cells show that this analytic method is logical and effective, and has extensive applicability while keeping high prediction accuracy.
引文
[1]邱竹贤.预焙槽炼铝[M].北京:冶金工业出版社,2008:350,373-377.Qiu Z X.Pre-baked cell aluminium melting[M].Beijing:Metallurgical Industry Press,2008:350,373-377.
[2]杨玲,王智堂.铝电解阳极效应的分析及控制[J].有色金属(冶炼部分),2007,59(3):28-30.Yang L,Wang Z T.Analyzing and controlling of anode effect in aluminum electrolysis[J].Nonferrous Metals(Extractive Metallurgy),2007,59(3):28-30.
[3]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(自然科学版),2001,26(6):101-105.Cui H,Xie G,Chen S R,et al.Application of intelligent control in the aluminum electrolytic cell[J].Journal of Kunming University of Science&Technology(Natural Science Edition),2001,26(6):101-105.
[4]刘业翔,李劼.现代铝电解[M].北京:冶金工业出版社,2013:56-62.Liu Y X,Li J.Modern aluminum electrolysis[M].Beijing:Metallurgical Industry Press,2013:56-62.
[5]李界家.铝电解生产过程中阳极效应预报新方法研究[J].轻金属,2002,30(5):59-62.LI J J.Research of new method of anode effect prediction in the production process of aluminium electrolytic[J].Light Metals,2002,30(5):59-62.
[6]Meghlaoui A,Thibault J,Bui R T,et al.Neural networks for the identification of the aluminum electrolysis process[J].Computers&Chemical Engineering,1998,22(10):1419-1428.
[7]李界家,孙璐璐,王奔,等.基于Elman神经网络阳极效应故障预报方法[J].沈阳建筑大学学报(自然科学版),2010,26(5):1012-1016.Li J J,Sun L L,Wang B,et al.Research of fault prediction of anode effect based on elman neural network[J].Journal of Shenyang Jianzhu University(Natural Science),2010,26(5):1012-1016.
[8]Rye K A,Koninsson M,Solberg I.Current distribution among individual anode carbons in a Hall-Heroult prebake cell at low alumina concentrations[J].Light Metals 1998,241-246.
[9]Kobbeltvedt O,Moxnes B P.On the bath flow,alumina distribution and anode gas in aluminum cells[J].Essential Readings in Light Metals:Aluminum Reduction Technology,2013,(2):257-264.
[10]杨军,邱天爽,邱竹贤,等.炭阳极气泡振动信号的谱分析与阳极效应预报[J].轻金属,2004,32(8):33-37.Yang J,Qiu T S,Qiu Z X,et al.Spectral analysis of the vibratory signals through carbon anodes and prediction of anodes effect[J].Light Metals,2004,32(8):33-37.
[11]李界家,李旸,于丰.模糊系统在阳极效应故障诊断中的应用[J].轻金属,2007,35(11):20-23.Li J J,Li Y,Yu F.Application of fuzzy system in diagnosing of anode effects[J].Light Metals,2007,35(11):20-23.
[12]曾水平.铝电解过程阳极效应预测[J].冶金自动化,2008,32(5):7-10.Zeng S P.Prediction of anode effect in aluminum electrolyzing process[J].Metallurgical Industry Automation,2008,32(5):7-10.
[13]Xing J,Xiao D Y.Ordered neural network and its application to prediction of anode effect[J].Control Engineering of China,2007,14(1):27-33.
[14]张愉,齐美星.铝电解槽电阻信号的小波包分析[J].苏州市职业大学学报,2012,23(1):19-23.Zhang Y,Qi M X.Study on resistance signal of aluminum reduction cells applying wavelet packet[J].Journal of Suzhou Vocational University,2012,23(1):19-23.
[15]郭宇奇.基于小波分析的铝电解槽故障特征提取[D].北京:北方工业大学(图书馆),2013:49.Guo Y Q.Extraction of aluminum electrolytic cell fault feature based on wavelet analysis[D].Beijing:North China University of Technology(Library),2013:49.
[16]丁立伟,聂婷,李停.基于BP网络和专家系统的铝电解槽分层故障诊断[J].计算机测量与控制,2014,22(11):3476-3479.Ding L W,Nie T,Li T.Aluminum cell hierarchical fault diagnosis method based on BP network and expert system[J].Computer Measurement&Control,2014,22(11):3476-3479.
[17]Sadler,Barry A.Frequency response analysis of anode current signals as a diagnostic aid for detecting approaching anode effects in aluminum smelting cells[J].Light Metals 2013,887-892.
[18]Majid N A A,Taylor M P,Chen J J J,et al.Diagnosing faults in aluminium processing by using multivariate statistical approaches[J].Journal of Materials Science,2012,47(3):1268-1279.
[19]Majid N A A,Taylor M P,Chen J J J,et al.Aluminium process fault detection by multiway principal component analysis[J].Control Engineering Practice,2011,19(4):367-379.
[20]Tessier J,Zwirz T G,Tarcy G P,et al.Multivariate statistical process monitoring of reduction cells[J].Light Metals 2009,305-310.
[21]Shuiping Z,Jinhong L.Diagnosis system of the anode faults for alumina reduction cell[C].Intelligent Systems Design and Applications,2006.ISDA'06.Sixth International Conference on.Los Angeles:IEEE,2006,1:844-849.
[22]史峰,王小川,郁磊,等.Matlab神经网络30个案例分析[M].北京:北京航空航天大学出版社,2010:73-80.Shi F,Wang X C,Yu L,et al.Matlab neural network 30 case analysis[M].Beijing:Beijing University of Aeronautics and Astronautics Press,2010:73-80.
[2]杨玲,王智堂.铝电解阳极效应的分析及控制[J].有色金属(冶炼部分),2007,59(3):28-30.Yang L,Wang Z T.Analyzing and controlling of anode effect in aluminum electrolysis[J].Nonferrous Metals(Extractive Metallurgy),2007,59(3):28-30.
[3]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(自然科学版),2001,26(6):101-105.Cui H,Xie G,Chen S R,et al.Application of intelligent control in the aluminum electrolytic cell[J].Journal of Kunming University of Science&Technology(Natural Science Edition),2001,26(6):101-105.
[4]刘业翔,李劼.现代铝电解[M].北京:冶金工业出版社,2013:56-62.Liu Y X,Li J.Modern aluminum electrolysis[M].Beijing:Metallurgical Industry Press,2013:56-62.
[5]李界家.铝电解生产过程中阳极效应预报新方法研究[J].轻金属,2002,30(5):59-62.LI J J.Research of new method of anode effect prediction in the production process of aluminium electrolytic[J].Light Metals,2002,30(5):59-62.
[6]Meghlaoui A,Thibault J,Bui R T,et al.Neural networks for the identification of the aluminum electrolysis process[J].Computers&Chemical Engineering,1998,22(10):1419-1428.
[7]李界家,孙璐璐,王奔,等.基于Elman神经网络阳极效应故障预报方法[J].沈阳建筑大学学报(自然科学版),2010,26(5):1012-1016.Li J J,Sun L L,Wang B,et al.Research of fault prediction of anode effect based on elman neural network[J].Journal of Shenyang Jianzhu University(Natural Science),2010,26(5):1012-1016.
[8]Rye K A,Koninsson M,Solberg I.Current distribution among individual anode carbons in a Hall-Heroult prebake cell at low alumina concentrations[J].Light Metals 1998,241-246.
[9]Kobbeltvedt O,Moxnes B P.On the bath flow,alumina distribution and anode gas in aluminum cells[J].Essential Readings in Light Metals:Aluminum Reduction Technology,2013,(2):257-264.
[10]杨军,邱天爽,邱竹贤,等.炭阳极气泡振动信号的谱分析与阳极效应预报[J].轻金属,2004,32(8):33-37.Yang J,Qiu T S,Qiu Z X,et al.Spectral analysis of the vibratory signals through carbon anodes and prediction of anodes effect[J].Light Metals,2004,32(8):33-37.
[11]李界家,李旸,于丰.模糊系统在阳极效应故障诊断中的应用[J].轻金属,2007,35(11):20-23.Li J J,Li Y,Yu F.Application of fuzzy system in diagnosing of anode effects[J].Light Metals,2007,35(11):20-23.
[12]曾水平.铝电解过程阳极效应预测[J].冶金自动化,2008,32(5):7-10.Zeng S P.Prediction of anode effect in aluminum electrolyzing process[J].Metallurgical Industry Automation,2008,32(5):7-10.
[13]Xing J,Xiao D Y.Ordered neural network and its application to prediction of anode effect[J].Control Engineering of China,2007,14(1):27-33.
[14]张愉,齐美星.铝电解槽电阻信号的小波包分析[J].苏州市职业大学学报,2012,23(1):19-23.Zhang Y,Qi M X.Study on resistance signal of aluminum reduction cells applying wavelet packet[J].Journal of Suzhou Vocational University,2012,23(1):19-23.
[15]郭宇奇.基于小波分析的铝电解槽故障特征提取[D].北京:北方工业大学(图书馆),2013:49.Guo Y Q.Extraction of aluminum electrolytic cell fault feature based on wavelet analysis[D].Beijing:North China University of Technology(Library),2013:49.
[16]丁立伟,聂婷,李停.基于BP网络和专家系统的铝电解槽分层故障诊断[J].计算机测量与控制,2014,22(11):3476-3479.Ding L W,Nie T,Li T.Aluminum cell hierarchical fault diagnosis method based on BP network and expert system[J].Computer Measurement&Control,2014,22(11):3476-3479.
[17]Sadler,Barry A.Frequency response analysis of anode current signals as a diagnostic aid for detecting approaching anode effects in aluminum smelting cells[J].Light Metals 2013,887-892.
[18]Majid N A A,Taylor M P,Chen J J J,et al.Diagnosing faults in aluminium processing by using multivariate statistical approaches[J].Journal of Materials Science,2012,47(3):1268-1279.
[19]Majid N A A,Taylor M P,Chen J J J,et al.Aluminium process fault detection by multiway principal component analysis[J].Control Engineering Practice,2011,19(4):367-379.
[20]Tessier J,Zwirz T G,Tarcy G P,et al.Multivariate statistical process monitoring of reduction cells[J].Light Metals 2009,305-310.
[21]Shuiping Z,Jinhong L.Diagnosis system of the anode faults for alumina reduction cell[C].Intelligent Systems Design and Applications,2006.ISDA'06.Sixth International Conference on.Los Angeles:IEEE,2006,1:844-849.
[22]史峰,王小川,郁磊,等.Matlab神经网络30个案例分析[M].北京:北京航空航天大学出版社,2010:73-80.Shi F,Wang X C,Yu L,et al.Matlab neural network 30 case analysis[M].Beijing:Beijing University of Aeronautics and Astronautics Press,2010:73-80.