湿法炼锌净液钴离子浓度在线检测及预测模型的研究与实现
|
摘要
锌粉锑盐净化除钴过程是湿法炼锌净化过程中的一个重要环节,其钴离子浓度不仅决定了锌粉的消耗量,而且直接影响电解的效率和产品的质量。因此,及时、准确地获得净化Ⅰ段后液钴离子浓度并预估净化Ⅱ段出口处钴离子浓度对合理调整锌粉加入量,保证生产正常进行和企业节能降耗,提高经济效益具有重要意义。
本文在分析工业现场锌粉锑盐除钴过程工艺和影响因素的基础上,结合电化学理论、误差补偿技术、专家系统知识和人工神经网络思想,实现了净化Ⅰ段后液钴离子浓度在线检测和净化Ⅱ段出口处钴离子浓度的预测,其主要研究内容有:
①研究了各种钴离子浓度检测方法的原理和特点,分析了极谱法在线检测痕量钴离子浓度的可行性,提出了基于极谱法的净化Ⅰ段后液钴离子浓度在线检测方法,研制了基于极谱法的在线分析仪。在分析了现场分析仪在现场应用中误差较大不能满足工业需求的原因后,利用数据筛选和最小二乘拟合数据处理方法对在线检测的算法进行了改进,补偿了由环境和设备引起的误差;根据极谱法原理补偿了由温度引起的误差,成功解决了存在的问题,满足了工业应用的需求。
②为了解决检测滞后难以及时获得Ⅱ段后液钴离子浓度信息以调整锌粉添加量的问题,在运用专家知识和数据预处理技术对大量工业历史数据进行预处理的基础上,利用处理后的数据作为样本建立了基于传统BP神经网络算法的净化Ⅱ段出口处钴离子浓度预测模型。针对传统BP神经网络算法在泛化能力方面的不足,建立了基于L-M算法的神经网络预测模型,经现场数据验证,该模型能依据Ⅱ段工况信息准确预测出Ⅱ段净化出口处的钴离子浓度。
③开发了钴离子浓度在线检测与预测系统,论文给出了系统的需求分析和结构、软件开发平台与功能实现。所开发的系统实现了Ⅰ段后液钴离子浓度在线检测和Ⅱ段后液钴离子浓度预测,并提供了工作记录表、报表统计等信息处理与管理功能,对提高工作效率和企业的现代化水平,发挥了重要作用。
论文最后总结了全文的主要研究工作,指出了可进一步改进的地方和完善的途径。
The purification-cobalt removal with antimony trioxide and zinc powder is an important part of zinc hydrometallurgy process. In this processing, cobalt concentration is a critical process parameter, which not only decides zinc powder consumption, but also affects greatly on electrolysis-efficiency and product quality. Therefore, getting the cobalt concentration value timely and exactly has great significance on assuring production succeed in completing at industrial produce, energy consumption reduction and enhancing the economic benefits.
A cobalt concentration online detection and predication modal is presented based on the analysis of the techniques and influencing factors of the purification-cobalt removal with antimony trioxide and zinc powder. It combines some technologies of electrochemistry theory, data processing technology, expert system and artificial intelligence theory. The main subjects of my researches are:
①Based on the analysis of the theories and features of all kinds of cobalt concentration detection methods and the feasibility of detecting trace amounts of cobalt with polarography method, a cobalt concentration of section I online detection with polarography method is developed. After analyzing the factors of the error, a new data processing algorithm based on data sieve, least-square fitting and temperature compensation technique is proposed to meet the demand of industry.
②A back propagation neural network predication model based on lots of historical data in industry is presented and simulated. A Levenberg-Marquardt neural network predication model is built to make up the disadvantages of traditional back propagation neural network, which can predict cobalt concentration timely and exactly based on the industrial information.
③A cobalt concentration online detection and predication system is developed based on the studies above. The overall design of system, the design platform, tools and the function of software are listed in the thesis. The system not only has the basic function of detecting and predicting online, but also has other useful functions such as providing work records forms and statistics of reports. It improves the efficiency of working and the level of modernization of enterprises.
Finally, the main research work is summarized and some improvable aspects and further researches are presented.
本文在分析工业现场锌粉锑盐除钴过程工艺和影响因素的基础上,结合电化学理论、误差补偿技术、专家系统知识和人工神经网络思想,实现了净化Ⅰ段后液钴离子浓度在线检测和净化Ⅱ段出口处钴离子浓度的预测,其主要研究内容有:
①研究了各种钴离子浓度检测方法的原理和特点,分析了极谱法在线检测痕量钴离子浓度的可行性,提出了基于极谱法的净化Ⅰ段后液钴离子浓度在线检测方法,研制了基于极谱法的在线分析仪。在分析了现场分析仪在现场应用中误差较大不能满足工业需求的原因后,利用数据筛选和最小二乘拟合数据处理方法对在线检测的算法进行了改进,补偿了由环境和设备引起的误差;根据极谱法原理补偿了由温度引起的误差,成功解决了存在的问题,满足了工业应用的需求。
②为了解决检测滞后难以及时获得Ⅱ段后液钴离子浓度信息以调整锌粉添加量的问题,在运用专家知识和数据预处理技术对大量工业历史数据进行预处理的基础上,利用处理后的数据作为样本建立了基于传统BP神经网络算法的净化Ⅱ段出口处钴离子浓度预测模型。针对传统BP神经网络算法在泛化能力方面的不足,建立了基于L-M算法的神经网络预测模型,经现场数据验证,该模型能依据Ⅱ段工况信息准确预测出Ⅱ段净化出口处的钴离子浓度。
③开发了钴离子浓度在线检测与预测系统,论文给出了系统的需求分析和结构、软件开发平台与功能实现。所开发的系统实现了Ⅰ段后液钴离子浓度在线检测和Ⅱ段后液钴离子浓度预测,并提供了工作记录表、报表统计等信息处理与管理功能,对提高工作效率和企业的现代化水平,发挥了重要作用。
论文最后总结了全文的主要研究工作,指出了可进一步改进的地方和完善的途径。
The purification-cobalt removal with antimony trioxide and zinc powder is an important part of zinc hydrometallurgy process. In this processing, cobalt concentration is a critical process parameter, which not only decides zinc powder consumption, but also affects greatly on electrolysis-efficiency and product quality. Therefore, getting the cobalt concentration value timely and exactly has great significance on assuring production succeed in completing at industrial produce, energy consumption reduction and enhancing the economic benefits.
A cobalt concentration online detection and predication modal is presented based on the analysis of the techniques and influencing factors of the purification-cobalt removal with antimony trioxide and zinc powder. It combines some technologies of electrochemistry theory, data processing technology, expert system and artificial intelligence theory. The main subjects of my researches are:
①Based on the analysis of the theories and features of all kinds of cobalt concentration detection methods and the feasibility of detecting trace amounts of cobalt with polarography method, a cobalt concentration of section I online detection with polarography method is developed. After analyzing the factors of the error, a new data processing algorithm based on data sieve, least-square fitting and temperature compensation technique is proposed to meet the demand of industry.
②A back propagation neural network predication model based on lots of historical data in industry is presented and simulated. A Levenberg-Marquardt neural network predication model is built to make up the disadvantages of traditional back propagation neural network, which can predict cobalt concentration timely and exactly based on the industrial information.
③A cobalt concentration online detection and predication system is developed based on the studies above. The overall design of system, the design platform, tools and the function of software are listed in the thesis. The system not only has the basic function of detecting and predicting online, but also has other useful functions such as providing work records forms and statistics of reports. It improves the efficiency of working and the level of modernization of enterprises.
Finally, the main research work is summarized and some improvable aspects and further researches are presented.
引文
[1]有色金属提取冶金手册(锌镉铅铋卷).北京:冶金工业出版社,2000
[2]邱定藩.有色金属科技进步与展望.纪念《有色金属》创刊50周年,北京:冶金工业出版社.1999
[3]唐朝波.锑盐净化除钻工艺的研究:[硕士学位论文].长沙:中南大学,1999
[4]中国有色金属学会.锌冶金.长沙:中南大学出版社,2005
[5]徐采栋,林蓉,汪大成.冶金物理化学.上海:上海科学技术出版社,1997
[6]戴军.用锑砷混合添加剂从硫酸锌溶液中除钴的研究:[硕士学位论文].沈阳:东北大学,2002
[7]梅光贵.湿法炼锌学.长沙:中南大学出版社,2001
[8]株洲冶炼厂.锌的湿法冶炼.长沙:湖南人民出版社,1973
[9]黄卫平.微量元素钴的分析方法概况.广东微量元素科学,1999,6(6):1-6
[10]李竹英.自动分析仪测钴方法研究:[硕士学位论文].长沙:中南大学,2003
[11]张明浩,任凤莲,徐金华,李志红.示波极谱法同时测定锌电解液中镉和钴.中南工业大学学报,2000,31(6):235-239
[12]徐益军,唐有根,杨幼平,王艳.示波极谱法快速测定锌电解液中痕量钴.冶金分析,2003,23(1):1-4
[13]刘渺.钢铁企业主工序分厂煤气量预测方法研究:[硕士学位论文].长沙:中南大学,2006
[14]钟颖,汪秉文.基于遗传算法的BP神经网络时间序列预测模型.系统工程与电子技术,2002,24(4):9-11
[15]Vapnik V N.The Nature of Statistical Learning Theory.New york:Springer Verlag,2000
[16]Dai Jun,Wang Dequan,Jiang Lan,et al.Removal of cobalt from zinc sulphate solution using rode antimony trioxide as additive.Trans.of Nonferrous Met.Soc.China,2002,12(6):1172-1175
[17]曾桂生.硫酸锌溶液中锌粉置换除钴的机理研究:[博士学位论文].昆明:昆明理工大学,2006
[18]Mishra P K,Das R P.Kinetics of zinc and cobalt sulphide precipitation and its application in hydrometallurgical separation.Hydrometallurgy,1992,(28):373-379
[19]Yoshifijmi Yamamoto,Koiehi Fumino,Takumi Ueda,.etal.A potient-dynamie study of the lead electrode in sulphurie acid solution.Elecmchim-ica Acta,1992, 37,(2):199-203
[20]王文录.湿法炼锌工艺除钴的实践.湖南有色金属,2003,19(3):24-26
[21]Oluf Bockman.Cobalt cementation in zinc electrowinning[dissertation].Trondheim:Norges teknisk-naturvitenskapelige universitet,1999
[22]Rashkov St,Dobrev Ts,Noncheva Z,etal.Lead-cobalt anodes for electrowinning of zinc from sulphate electrolytes.Hydrometallurgy,1999,2:223-230
[23]Lawson F,Lee T.In:Kinetics of removal of cobalt from zinc sulfate electrolytes by cementation.Hydrometallurgy'81,Manchester.1981
[24]曹为民.砷盐和锑盐净化除钴的探讨.湖南有色金属,2001,17:17-22
[25]Kruus P,Robertson D A,McMillen L A.Effects of ultrasound on the cementation of cobalt on zinc.Ultrasonics,1991,29(5):30-37
[26]Zeng Guisheng,Xie Gang,Yang Dajin,Wang Dajian.Oxidation Resistivity of Boride Coating of Graphite Anode Sample.Materials Chemistry and Physics,2006,95:183-187
[27]Zeng Guisheng,Xie Gang,Yang Dajin.The Effect of Cadmium Ion on Cobalt Removal from Zinc Sulfate Soluntion.Minerals Engineering,2006,19(2):197-200
[28]张林.株冶三段逆锑盐连续净化试生产实践之我见.株冶科技,1997,25(4):17-20
[29]Karavasteva M.The effect of certain surfactants on the acid and neutral leaching of zinc calcine obtained by roasting of sphalerite concetrates.Hydrometallurgy,2001,62:151-156
[30]杨静.环境中痕量铜_钴_镍的检测新方法研究:[硕士学位论文].衡阳:南华大学,2007
[31]韩之俊,靳京民编著.测量质量工程学.北京:中国计量出版社,2000
[32]陈建铎.测量数据的筛选与程序设计.现代电子技术.1996,1:30-32
[33]金盛.环形线圈检测器交通数据预处理方法研究:[硕士学位论文].长春:吉林大学,2007
[34]俞龙江,杨英,孙圣和.基于最小二乘拟合法的焊点形状检测.仪器仪表学报,2007,28(7):1255-1258
[35]Thirwrnalaiah K,Deo M C.River stage forecasting using artificial neural networks.Hydrol Eng,1998,3(1):26-32
[36]Skipworth P J,Saul A J.Maehell J.Predicting water quality in distribution systems using artificial neural networks.Water Maritime and Energy,1999,136(1):1-8
[37]Gumrah F,OZ B.The application of artificial neural networks for the prediction of water quality of polluted aquifer.Water Air Soil Pollut,2000,119(1-4):275-291
[38]Maier H R,Dandy G C.The use of artificial neural networks for the prediction of water quality parameters.Water resource,1996,32(4):1013-1022
[39]Maier H R,Dandy G C.Use of artificial neural networks for modelling cyanobacteria Anabaena spp.In the River Murray,South Australia.Ecological Modelling,1998,105(2-3):257-272
[40]唐万梅.几个预测方法及模型的研究:[博士学位论文].呼和浩特:内蒙古大学,2006
[41]李勇刚,桂卫华.基于PCA的多神经网络软测量模型及其在工业中的应用.小型微型计算机系统,2004,25(10):1781-1784.
[42]Li Jin,Zhang Zi-ping,Chen Hong-fang.The PCA A method in neural networks.Journal of China University of Science and Technology,2001.21(1):40-44.
[43]范磊,张运陶,程正军.基于Matlab的改进BP神经网络及其应用.西华师范大学学报,2005,26(1):71-73
[44]蔡自兴.智能控制.北京:电子工业出版社,2004
[45]阳春华,沈德耀,吴敏.焦炉配煤专家系统的定性定量综合设计方法.自动化学报,2000,26(2):226-232
[46]阳春华,段小刚,王雅琳等.烧结法生产氧化铝生料浆的配料专家系统设计.中南大学学报,2005,36(4):648-652
[47]Cichocki A,Unbehauen R.Neural networks for optimization and signal processing.New York:John Wiley & Sons Ltd,1993:53
[48]喻寿益,王吉林,彭晓波.基于神经网络的铜闪速熔炼过程工艺参数预测模型.中南大学学报(自然科学版),2007,38(3):523-527
[49]伍春香,刘琳.三层BP网络隐层节点数确定方法的研究.武汉测绘科技大学学报,1999,24(2):177-178
[50]高大文,王鹏,蔡臻超.人工神经网络中隐含层节点与训练次数的优化.哈尔滨工业大学学报,2003,35(2):207-209
[51]Li B,Chow M Y,Tipsuwan Y,etal.Neural-network-based motorrolling bearing fault diagnosis.IEEE Transactions on Industrial Electronics,2000,47(5):1060-1069
[52]Tang Wan-Mei.The study of the optimal structure of BP neural network.System Engineering Theory and Practice,2005,25(10):95-100
[53]Ng S C,Cheung C C,Leung S H.Fast convergence for back-propagation network with magnified gradient function.IEEE International Joint Conference on Neural Networks,2003,9(3):1903-1908
[54]司捷,周贵安,李函等.基于梯度监督学习的理论与应用(Ⅰ)-基本算法.清华大学学报(自然科学版),1997,37(7):71-73
[55]司捷,周贵安,李函等.基于梯度监督学习的理论与应用(Ⅱ)-训练机制.清华大学学报(自然科学版),1997,37(9):104-107
[56]高雪鹏,从爽.BP网络改进算法的性能对比研究.控制与决策,2001,16(3):167-171
[57]刘鹰,赵琳.神经网络BP算法的改进和仿真.计算机仿真,1999,16(3):12-14
[58]Wilamowski B M,Iplikci S,Kaynak O,Efe M O.In:training neural netwokrs;Neuarl Netwokrs,2001.Proceedings.IJCNN'01.Intenrational Joint Conference on,Volume:3,2001,Page(s):1778-1782vol.3
[59]Ampazis N,Perantonis S J.Levenberg-Marquardt algorithm with adaptive momentum for the efficient training of feedforward networks;Nueral Netwokrs,2000.IJCNN2000,Proceedings of the IEEE-INNS-ENNS International Joint Conference on,Volume:1,2000,Pgae(s):126-131vol.1
[60]阳春华,谢明,桂卫华,彭晓波.铜闪速熔炼过程冰铜品位预测模型的研究与应用.信息与控制,2008,37(2):28-33
[61]任妮.改进L-M型BP模型的降雨径流预报:[硕士学位论文].大连:大连理工大学,2006
[62]王建梅,覃文忠.基于L-M算法的BP神经网络分类器.武汉大学学报,2005,30(10):928-931
[2]邱定藩.有色金属科技进步与展望.纪念《有色金属》创刊50周年,北京:冶金工业出版社.1999
[3]唐朝波.锑盐净化除钻工艺的研究:[硕士学位论文].长沙:中南大学,1999
[4]中国有色金属学会.锌冶金.长沙:中南大学出版社,2005
[5]徐采栋,林蓉,汪大成.冶金物理化学.上海:上海科学技术出版社,1997
[6]戴军.用锑砷混合添加剂从硫酸锌溶液中除钴的研究:[硕士学位论文].沈阳:东北大学,2002
[7]梅光贵.湿法炼锌学.长沙:中南大学出版社,2001
[8]株洲冶炼厂.锌的湿法冶炼.长沙:湖南人民出版社,1973
[9]黄卫平.微量元素钴的分析方法概况.广东微量元素科学,1999,6(6):1-6
[10]李竹英.自动分析仪测钴方法研究:[硕士学位论文].长沙:中南大学,2003
[11]张明浩,任凤莲,徐金华,李志红.示波极谱法同时测定锌电解液中镉和钴.中南工业大学学报,2000,31(6):235-239
[12]徐益军,唐有根,杨幼平,王艳.示波极谱法快速测定锌电解液中痕量钴.冶金分析,2003,23(1):1-4
[13]刘渺.钢铁企业主工序分厂煤气量预测方法研究:[硕士学位论文].长沙:中南大学,2006
[14]钟颖,汪秉文.基于遗传算法的BP神经网络时间序列预测模型.系统工程与电子技术,2002,24(4):9-11
[15]Vapnik V N.The Nature of Statistical Learning Theory.New york:Springer Verlag,2000
[16]Dai Jun,Wang Dequan,Jiang Lan,et al.Removal of cobalt from zinc sulphate solution using rode antimony trioxide as additive.Trans.of Nonferrous Met.Soc.China,2002,12(6):1172-1175
[17]曾桂生.硫酸锌溶液中锌粉置换除钴的机理研究:[博士学位论文].昆明:昆明理工大学,2006
[18]Mishra P K,Das R P.Kinetics of zinc and cobalt sulphide precipitation and its application in hydrometallurgical separation.Hydrometallurgy,1992,(28):373-379
[19]Yoshifijmi Yamamoto,Koiehi Fumino,Takumi Ueda,.etal.A potient-dynamie study of the lead electrode in sulphurie acid solution.Elecmchim-ica Acta,1992, 37,(2):199-203
[20]王文录.湿法炼锌工艺除钴的实践.湖南有色金属,2003,19(3):24-26
[21]Oluf Bockman.Cobalt cementation in zinc electrowinning[dissertation].Trondheim:Norges teknisk-naturvitenskapelige universitet,1999
[22]Rashkov St,Dobrev Ts,Noncheva Z,etal.Lead-cobalt anodes for electrowinning of zinc from sulphate electrolytes.Hydrometallurgy,1999,2:223-230
[23]Lawson F,Lee T.In:Kinetics of removal of cobalt from zinc sulfate electrolytes by cementation.Hydrometallurgy'81,Manchester.1981
[24]曹为民.砷盐和锑盐净化除钴的探讨.湖南有色金属,2001,17:17-22
[25]Kruus P,Robertson D A,McMillen L A.Effects of ultrasound on the cementation of cobalt on zinc.Ultrasonics,1991,29(5):30-37
[26]Zeng Guisheng,Xie Gang,Yang Dajin,Wang Dajian.Oxidation Resistivity of Boride Coating of Graphite Anode Sample.Materials Chemistry and Physics,2006,95:183-187
[27]Zeng Guisheng,Xie Gang,Yang Dajin.The Effect of Cadmium Ion on Cobalt Removal from Zinc Sulfate Soluntion.Minerals Engineering,2006,19(2):197-200
[28]张林.株冶三段逆锑盐连续净化试生产实践之我见.株冶科技,1997,25(4):17-20
[29]Karavasteva M.The effect of certain surfactants on the acid and neutral leaching of zinc calcine obtained by roasting of sphalerite concetrates.Hydrometallurgy,2001,62:151-156
[30]杨静.环境中痕量铜_钴_镍的检测新方法研究:[硕士学位论文].衡阳:南华大学,2007
[31]韩之俊,靳京民编著.测量质量工程学.北京:中国计量出版社,2000
[32]陈建铎.测量数据的筛选与程序设计.现代电子技术.1996,1:30-32
[33]金盛.环形线圈检测器交通数据预处理方法研究:[硕士学位论文].长春:吉林大学,2007
[34]俞龙江,杨英,孙圣和.基于最小二乘拟合法的焊点形状检测.仪器仪表学报,2007,28(7):1255-1258
[35]Thirwrnalaiah K,Deo M C.River stage forecasting using artificial neural networks.Hydrol Eng,1998,3(1):26-32
[36]Skipworth P J,Saul A J.Maehell J.Predicting water quality in distribution systems using artificial neural networks.Water Maritime and Energy,1999,136(1):1-8
[37]Gumrah F,OZ B.The application of artificial neural networks for the prediction of water quality of polluted aquifer.Water Air Soil Pollut,2000,119(1-4):275-291
[38]Maier H R,Dandy G C.The use of artificial neural networks for the prediction of water quality parameters.Water resource,1996,32(4):1013-1022
[39]Maier H R,Dandy G C.Use of artificial neural networks for modelling cyanobacteria Anabaena spp.In the River Murray,South Australia.Ecological Modelling,1998,105(2-3):257-272
[40]唐万梅.几个预测方法及模型的研究:[博士学位论文].呼和浩特:内蒙古大学,2006
[41]李勇刚,桂卫华.基于PCA的多神经网络软测量模型及其在工业中的应用.小型微型计算机系统,2004,25(10):1781-1784.
[42]Li Jin,Zhang Zi-ping,Chen Hong-fang.The PCA A method in neural networks.Journal of China University of Science and Technology,2001.21(1):40-44.
[43]范磊,张运陶,程正军.基于Matlab的改进BP神经网络及其应用.西华师范大学学报,2005,26(1):71-73
[44]蔡自兴.智能控制.北京:电子工业出版社,2004
[45]阳春华,沈德耀,吴敏.焦炉配煤专家系统的定性定量综合设计方法.自动化学报,2000,26(2):226-232
[46]阳春华,段小刚,王雅琳等.烧结法生产氧化铝生料浆的配料专家系统设计.中南大学学报,2005,36(4):648-652
[47]Cichocki A,Unbehauen R.Neural networks for optimization and signal processing.New York:John Wiley & Sons Ltd,1993:53
[48]喻寿益,王吉林,彭晓波.基于神经网络的铜闪速熔炼过程工艺参数预测模型.中南大学学报(自然科学版),2007,38(3):523-527
[49]伍春香,刘琳.三层BP网络隐层节点数确定方法的研究.武汉测绘科技大学学报,1999,24(2):177-178
[50]高大文,王鹏,蔡臻超.人工神经网络中隐含层节点与训练次数的优化.哈尔滨工业大学学报,2003,35(2):207-209
[51]Li B,Chow M Y,Tipsuwan Y,etal.Neural-network-based motorrolling bearing fault diagnosis.IEEE Transactions on Industrial Electronics,2000,47(5):1060-1069
[52]Tang Wan-Mei.The study of the optimal structure of BP neural network.System Engineering Theory and Practice,2005,25(10):95-100
[53]Ng S C,Cheung C C,Leung S H.Fast convergence for back-propagation network with magnified gradient function.IEEE International Joint Conference on Neural Networks,2003,9(3):1903-1908
[54]司捷,周贵安,李函等.基于梯度监督学习的理论与应用(Ⅰ)-基本算法.清华大学学报(自然科学版),1997,37(7):71-73
[55]司捷,周贵安,李函等.基于梯度监督学习的理论与应用(Ⅱ)-训练机制.清华大学学报(自然科学版),1997,37(9):104-107
[56]高雪鹏,从爽.BP网络改进算法的性能对比研究.控制与决策,2001,16(3):167-171
[57]刘鹰,赵琳.神经网络BP算法的改进和仿真.计算机仿真,1999,16(3):12-14
[58]Wilamowski B M,Iplikci S,Kaynak O,Efe M O.In:training neural netwokrs;Neuarl Netwokrs,2001.Proceedings.IJCNN'01.Intenrational Joint Conference on,Volume:3,2001,Page(s):1778-1782vol.3
[59]Ampazis N,Perantonis S J.Levenberg-Marquardt algorithm with adaptive momentum for the efficient training of feedforward networks;Nueral Netwokrs,2000.IJCNN2000,Proceedings of the IEEE-INNS-ENNS International Joint Conference on,Volume:1,2000,Pgae(s):126-131vol.1
[60]阳春华,谢明,桂卫华,彭晓波.铜闪速熔炼过程冰铜品位预测模型的研究与应用.信息与控制,2008,37(2):28-33
[61]任妮.改进L-M型BP模型的降雨径流预报:[硕士学位论文].大连:大连理工大学,2006
[62]王建梅,覃文忠.基于L-M算法的BP神经网络分类器.武汉大学学报,2005,30(10):928-931