烧结法氧化铝生料浆质量预测及应用研究
摘要
配料作为氧化铝生产的第一道工序,配制的生料浆指标的好坏不仅直接关系到熟料质量的高低,而且对整个系统的碱平衡和水平衡具有极大的影响。然而,目前生料浆成分检测严重滞后且波动范围大,配比难于及时调整,从而导致整个生产流程的实时控制陷入被动。为此,研究如何建立生料浆质量预测模型从而实现配比的实时调整,对实现氧化铝生产过程的稳产高产、提高企业竞争力都具有重要意义。
本文以中铝集团中州分公司氧化铝生产过程为背景,着重研究了生料浆质量预测模型的建立和应用。首先,在分析氧化铝配料过程的基础上,确定了影响生料浆质量的主要因素,提出了先建立硅渣成分含量预测模型和碱液成分含量预估模型,然后再建立生料浆质量预测模型的分步建模法;紧接着采用BP神经网络模型对硅渣成分含量进行预测,同时根据碱液槽和碱液泵的工作情况,运用流体力学知识简化,建立了碱液预估模型,其中神经网络的输出为预估模型的输入;最后,在预估模型的基础上,建立基于物料平衡的机理模型,并采用GM(1,1)对机理模型的偏差进行补偿,从而实现了生料浆质量指标的实时预测。
在此研究基础上,开发了基于模型的生料浆配料优化专家系统,论文描述了系统的结构、功能和系统软件设计,着重介绍了数据通信、数据库和报表打印几个技术的软件实现。系统软件采用Visual C++6.0编制,实现了配比监控、配比优化、数据导入、数据管理和帮助等功能。现场运行结果表明模型满足配比优化计算的精度要求,整个系统实现了配比的实时控制,提高了生料浆质量,稳定了生产。
The quality of raw slurry made up by blending, which is the first process of Alumina Production by Sintering Method, not only directly relates to the quality of the sintered grog, but also make great influence on the alkali and water balance in the whole system. However, at present it is difficult to timely measure and stabilize the quality of raw slurry, which complicates the adjustment of ratio and makes it is hard to achieve real-time control of the whole process of production. Therefore, how to implement real-time adjustment of the ratio by establishing the predictive model of quality for raw slurry, is significant for realizing steady and high production of alumina and enhancing the competitive power of enterprises.On the background of production process of alumina in Zhongzhou Brach China Aluminium, this paper mainly does research in the establishment and application of the predictive model of quality. Firstly, the key factors that influence the quality of raw slurry are acquired by analyzing the process of blending, then the multi-step modeling method is proposed. According to it, before the predictive model of quality for raw slurry is established, the predictive model of content for components of Ca-Al silicate slime and alkali liquor should be established. Secondly, the predictive model about Ca-Al silicate slime is established using BP neural networks(NN). Meanwhile according to the condition of alkali tanks and alkali pumps, the predictive model about alkali liquor, which is simplified by the principle of hydromechanics, is also established. One of its inputs is NN's output and its outputs is the inputs of the mechanism model that is based on the material balance principle. Finally, in order to compensate errors of the mechanism model, GM(1,1) is put forward. The compensated model realizes the real-time prediction of raw slurry indices.On the research mentioned above, a blending expert system based on the predictive model of quality is developed in this paper. The structure and function of system are presented, in which the technologies such as data communication, database, reports print are introduced in detail. The system software, developed by VC++, realizes functions of ratio
monitoring, ratio optimization, data leading-in, data management and help. An application result in Zhongzhou Brach China Aluminium shows the predictive model is effective and that the system implements the computation of blending ratio fast and efficiently, improves the quality of result and stabilizes industrial production.
本文以中铝集团中州分公司氧化铝生产过程为背景,着重研究了生料浆质量预测模型的建立和应用。首先,在分析氧化铝配料过程的基础上,确定了影响生料浆质量的主要因素,提出了先建立硅渣成分含量预测模型和碱液成分含量预估模型,然后再建立生料浆质量预测模型的分步建模法;紧接着采用BP神经网络模型对硅渣成分含量进行预测,同时根据碱液槽和碱液泵的工作情况,运用流体力学知识简化,建立了碱液预估模型,其中神经网络的输出为预估模型的输入;最后,在预估模型的基础上,建立基于物料平衡的机理模型,并采用GM(1,1)对机理模型的偏差进行补偿,从而实现了生料浆质量指标的实时预测。
在此研究基础上,开发了基于模型的生料浆配料优化专家系统,论文描述了系统的结构、功能和系统软件设计,着重介绍了数据通信、数据库和报表打印几个技术的软件实现。系统软件采用Visual C++6.0编制,实现了配比监控、配比优化、数据导入、数据管理和帮助等功能。现场运行结果表明模型满足配比优化计算的精度要求,整个系统实现了配比的实时控制,提高了生料浆质量,稳定了生产。
The quality of raw slurry made up by blending, which is the first process of Alumina Production by Sintering Method, not only directly relates to the quality of the sintered grog, but also make great influence on the alkali and water balance in the whole system. However, at present it is difficult to timely measure and stabilize the quality of raw slurry, which complicates the adjustment of ratio and makes it is hard to achieve real-time control of the whole process of production. Therefore, how to implement real-time adjustment of the ratio by establishing the predictive model of quality for raw slurry, is significant for realizing steady and high production of alumina and enhancing the competitive power of enterprises.On the background of production process of alumina in Zhongzhou Brach China Aluminium, this paper mainly does research in the establishment and application of the predictive model of quality. Firstly, the key factors that influence the quality of raw slurry are acquired by analyzing the process of blending, then the multi-step modeling method is proposed. According to it, before the predictive model of quality for raw slurry is established, the predictive model of content for components of Ca-Al silicate slime and alkali liquor should be established. Secondly, the predictive model about Ca-Al silicate slime is established using BP neural networks(NN). Meanwhile according to the condition of alkali tanks and alkali pumps, the predictive model about alkali liquor, which is simplified by the principle of hydromechanics, is also established. One of its inputs is NN's output and its outputs is the inputs of the mechanism model that is based on the material balance principle. Finally, in order to compensate errors of the mechanism model, GM(1,1) is put forward. The compensated model realizes the real-time prediction of raw slurry indices.On the research mentioned above, a blending expert system based on the predictive model of quality is developed in this paper. The structure and function of system are presented, in which the technologies such as data communication, database, reports print are introduced in detail. The system software, developed by VC++, realizes functions of ratio
monitoring, ratio optimization, data leading-in, data management and help. An application result in Zhongzhou Brach China Aluminium shows the predictive model is effective and that the system implements the computation of blending ratio fast and efficiently, improves the quality of result and stabilizes industrial production.
引文
[1] 王雅琳.智能集成建模理论及其在有色冶炼过程优化控制中的应用研究[D]:[博士学位论文].长沙:中南大学,2001
[2] Wang Yalin, Gui Weihua, et al. A multi-objective optimization for the mixture ratio of Pb-Zn sintering process based on an intelligent integrated model[C]. AMSMA'2000, GuangZhou, 2000
[3] Chen Xiaofang, Gui Weihua, Wang Yalin. A prediction model of sulfur based on intelligent integrated strategy[C]. Proceedings of IFAC 15th Congress, Barcelona, 2002
[4] 杨重愚.轻金属冶金学[M].北京:冶金工业出版社,1996
[5] 陈红武,王新武.稳定熟料指标降低氧化铝工艺能耗[J].有色金属节能,2001,0(1):13～15
[6] 谭天乐.基于粗糙集的过程建模、控制与故障诊断[D]:[博士学位论文].杭州:浙江大学,2003
[7] 杜玉晓.铅锌烧结过程智能集成优化控制技术及其应用研究[D]:[博士学位论文].长沙:中南大学,2004
[8] 应苏如.氧化铝生产过程的苛性比值与溶出率智能集成预测模型的研究与应用[D]:[硕士学位论文].长沙:中南大学,2004
[9] 王耀南,宋明.复杂工业系统的广义知识模型与智能模型[J].中南工业大学学报(自然科学版),2003,34(4):335～341
[10] 杨斌,田永青等.智能建模方法中的数据预处理[J].信息与控制,2002,31(4):380～384
[11] Schlang Martin. Neural Networks in Real World Applications[J]. IEEE EXPERT, 1996, 11(4): 8～9
[12] Mastumaoto H, Kurda C, Palosaari S, et al. Neural Network modeling of serum protein fractionation using gel filtration chromatography[J]. Journal of Chemical Engineering of Japan, 1999, 32(1): 1～7
[13] Babuska R, Berbruggon H B, Vancan H J L. Fuzzy Modeling of Enzymatic penicillin-G Anbersion[J]. Engineering Application of Artificial Intelligence, 1999, 12(1): 79～92
[14] Xiao Weirong, Weidemann B. Fuzzy Modeling and Its Application to Magentic Bearing Systems[J]. Fuzzy Sets and Systems, 1995, 73(20): 201～217
[15] 吕宏辉,钟络,夏红霞.灰色系统与神经网络融合技术探索[J].微机发展, 2000,10(3):3～5
[16] 邓聚龙.灰色控制系统(第二版)[M].武汉:华中理工大学出版社,1997
[17] 刘思峰,郭天榜,党耀国等.灰色系统理论及应用[M].北京:科技出版社,1999
[18] 陈竟鹏.用灰色理论模型预测供电量[J].渝州大学学报(自然科学版),2002,19(2):9～13
[19] 任国全.灰色预测模型的Delphi编程实现[J].微计算机信息,2002,18(3):51～52
[20] 杜玉晓,吴敏,岑丽辉等.铅锌烧结过程的集成建模方法及智能优化算法[J].小型微型计算机系统,2004,25(8):1458～1463
[21] 杜玉晓,吴敏.桂卫华.面向生产目标的铅锌烧结过程智能集成建模与优化控制技术[J].中国有色金属学报,2004,14(1):142～148
[22] Wan W H, Shi W X, Yuan Y F. Fuzzy modeling of nickel flash smelting process [J]. Metallurgical industry automation, 2000, 24(2): 9～12
[23] Xu G L, Li S P. Study on neural network forecast model of the secondary cooling of continuous casting[J]. Basic automation, 1999, 6(1): 5～7
[24] Kodkinen J, Yliniemi L, Leiviska K. Fuzzy modeling of a rotary dryer. In: Preprints of the IFAC Workshop[C]. Finland: Elsevier Science, 2000.
[25] Wu M, Nakano M, She J H. A model-based expert control strategy using neural networks for the coal blending process in an iron and steel plant[J]. Expert Systems with Applications, 1999, 16(6): 271～281
[26] 《联合法生产氧化铝》编写组.联合法生产氧化铝:基础知识[M].北京:冶金工业出版社,1975
[27] Zhang Sheng, Liu Hong-Xing, Gao Dun-Tang et al. Determining the input dimension of a neural network for nonlinear time series prediction[J]. Chinese Physics, 2003, 12(6): 594～598
[28] 冯志鹏,宋希庚,薛冬新.基于广义回归神经网络的时间序列预测研究[J].振动、测试与诊断,2003,23(2):105～109
[29] Hagan M T, Demouth H B, Beale M H. Neural Network Design[M]. Boston, MA: PWS Publishing, 1996.
[30] 戚德虎,康继昌.BP神经网络的设计[J].计算机工程与设计,1998,19(2):48～50
[31] Wang Yalin, Gui Weihua, Chen Xiaofang, et al. Neural network modeling for composition prediction of Pb-Zn sinter in imperial smelting process[C]. Proceedings of the Annual Chinese Automation Conference in the UK. Derby, England, 1999(Sept): 21～24
[32] G. Bloch, E Sirou, V. Eustache. Neural Intelligent Control for a Steel Plant[J]. IEEE Trans. on Neural Networks, 1997, 8(4): 910～918
[33] Yarlagadda Prasad K D V. Development of an integrated neural network system for prediction of process parameters in metal injection moulding[J]. Journal of Materials Processing Technology, 2002, 130:315～320
[34] 陈敏恒,丛德滋,方图南等.化工原理(第二版):上册[M].北京:化学工业出版社,1999
[35] 蒋维钧,戴猷元,顾惠君等.化工原理:上册[M].北京:清华大学出版社,1992
[36] 成都科技大学化工原理教研室编著.化工原理:上册[M].成都:成都科技大学出版社,1986
[37] 王科俊,王克成.神经网络建模、预报与控制[M].哈尔滨:哈尔滨工程大学出版社,1996
[38] 胡军,梅炽等.吹炼炉入炉铜锍品位的灰色预测[J].中国有色金属学报,2000,10(5):736～739
[39] 王军平,樊文侠,敬忠良等.一种构建灰色预测模型的新算法[J].西北大学学报(自然科学版),2002,32(6):613～616
[40] 罗开元,高峰,胡俐蕊.灰色广义预测控制算法及仿真研究[J].控制理论与应用,2002,19(2):207～210
[41] 邹健,杨莹春,诸静.基于灰色模型的预测模糊控制策略及其应用研究[J].中国电机工程学报,2002,22(9):12～14
[42] 吕安民,柯美忠,李成名等.灰色系统模型在时间序列模式中的应用研究[J].微机发展,2002,12(5):41～43
[43] 唐朝晖,桂卫华,吴敏等.基于神经网络和灰色理论的密闭鼓风炉透气性预测模型[J].中国有色金属学报,2003,13(5):1306～1310
[44] 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[J].自动化学报,2000,26(2):226～232
[45] (美)David J. Kruglinski. Visual C++技术内幕[M].北京:清华大学出版社,1999
[46] 希望图书创作室译.Visual C++6.0技术内幕[M].北京:北京希望电子出版社,1995
[47] 侯俊杰.深入浅出MFC第2版[M].武汉:华中科技大学出版社,2002
[48] 何海江.OPC客户端关健技术的实现[J].微计算机信息,2003,19(7):76～78
[49] 马增良,兰斌.OPC数据访问服务器实现机制研究[J].计算机工程与应用,2003,39(21):65～67,71
[50] 郑城荣,段来盛.Microsoft Access2.0开发指南[M].北京:人民邮电出版社,1999
[51] 张晓黎,吴崇峰,高雨茁.ODBC应用于注塑模浇注系统的二次开发[J].塑料科技,2003,0(1):12～14
[52] 李天宏,薛安,杨海宏.采用ODBC技术实现数据库与矢量图形的相互查询[J].测绘信息与工程,1997,3:11～14
[53] 巩伟,宋勇江,陈庆文.组态软件RSView32与VB、EXCEL结合的复杂生产报表解决方案[J].自动化技术与应用,2003,22(9):53～55
[2] Wang Yalin, Gui Weihua, et al. A multi-objective optimization for the mixture ratio of Pb-Zn sintering process based on an intelligent integrated model[C]. AMSMA'2000, GuangZhou, 2000
[3] Chen Xiaofang, Gui Weihua, Wang Yalin. A prediction model of sulfur based on intelligent integrated strategy[C]. Proceedings of IFAC 15th Congress, Barcelona, 2002
[4] 杨重愚.轻金属冶金学[M].北京:冶金工业出版社,1996
[5] 陈红武,王新武.稳定熟料指标降低氧化铝工艺能耗[J].有色金属节能,2001,0(1):13～15
[6] 谭天乐.基于粗糙集的过程建模、控制与故障诊断[D]:[博士学位论文].杭州:浙江大学,2003
[7] 杜玉晓.铅锌烧结过程智能集成优化控制技术及其应用研究[D]:[博士学位论文].长沙:中南大学,2004
[8] 应苏如.氧化铝生产过程的苛性比值与溶出率智能集成预测模型的研究与应用[D]:[硕士学位论文].长沙:中南大学,2004
[9] 王耀南,宋明.复杂工业系统的广义知识模型与智能模型[J].中南工业大学学报(自然科学版),2003,34(4):335～341
[10] 杨斌,田永青等.智能建模方法中的数据预处理[J].信息与控制,2002,31(4):380～384
[11] Schlang Martin. Neural Networks in Real World Applications[J]. IEEE EXPERT, 1996, 11(4): 8～9
[12] Mastumaoto H, Kurda C, Palosaari S, et al. Neural Network modeling of serum protein fractionation using gel filtration chromatography[J]. Journal of Chemical Engineering of Japan, 1999, 32(1): 1～7
[13] Babuska R, Berbruggon H B, Vancan H J L. Fuzzy Modeling of Enzymatic penicillin-G Anbersion[J]. Engineering Application of Artificial Intelligence, 1999, 12(1): 79～92
[14] Xiao Weirong, Weidemann B. Fuzzy Modeling and Its Application to Magentic Bearing Systems[J]. Fuzzy Sets and Systems, 1995, 73(20): 201～217
[15] 吕宏辉,钟络,夏红霞.灰色系统与神经网络融合技术探索[J].微机发展, 2000,10(3):3～5
[16] 邓聚龙.灰色控制系统(第二版)[M].武汉:华中理工大学出版社,1997
[17] 刘思峰,郭天榜,党耀国等.灰色系统理论及应用[M].北京:科技出版社,1999
[18] 陈竟鹏.用灰色理论模型预测供电量[J].渝州大学学报(自然科学版),2002,19(2):9～13
[19] 任国全.灰色预测模型的Delphi编程实现[J].微计算机信息,2002,18(3):51～52
[20] 杜玉晓,吴敏,岑丽辉等.铅锌烧结过程的集成建模方法及智能优化算法[J].小型微型计算机系统,2004,25(8):1458～1463
[21] 杜玉晓,吴敏.桂卫华.面向生产目标的铅锌烧结过程智能集成建模与优化控制技术[J].中国有色金属学报,2004,14(1):142～148
[22] Wan W H, Shi W X, Yuan Y F. Fuzzy modeling of nickel flash smelting process [J]. Metallurgical industry automation, 2000, 24(2): 9～12
[23] Xu G L, Li S P. Study on neural network forecast model of the secondary cooling of continuous casting[J]. Basic automation, 1999, 6(1): 5～7
[24] Kodkinen J, Yliniemi L, Leiviska K. Fuzzy modeling of a rotary dryer. In: Preprints of the IFAC Workshop[C]. Finland: Elsevier Science, 2000.
[25] Wu M, Nakano M, She J H. A model-based expert control strategy using neural networks for the coal blending process in an iron and steel plant[J]. Expert Systems with Applications, 1999, 16(6): 271～281
[26] 《联合法生产氧化铝》编写组.联合法生产氧化铝:基础知识[M].北京:冶金工业出版社,1975
[27] Zhang Sheng, Liu Hong-Xing, Gao Dun-Tang et al. Determining the input dimension of a neural network for nonlinear time series prediction[J]. Chinese Physics, 2003, 12(6): 594～598
[28] 冯志鹏,宋希庚,薛冬新.基于广义回归神经网络的时间序列预测研究[J].振动、测试与诊断,2003,23(2):105～109
[29] Hagan M T, Demouth H B, Beale M H. Neural Network Design[M]. Boston, MA: PWS Publishing, 1996.
[30] 戚德虎,康继昌.BP神经网络的设计[J].计算机工程与设计,1998,19(2):48～50
[31] Wang Yalin, Gui Weihua, Chen Xiaofang, et al. Neural network modeling for composition prediction of Pb-Zn sinter in imperial smelting process[C]. Proceedings of the Annual Chinese Automation Conference in the UK. Derby, England, 1999(Sept): 21～24
[32] G. Bloch, E Sirou, V. Eustache. Neural Intelligent Control for a Steel Plant[J]. IEEE Trans. on Neural Networks, 1997, 8(4): 910～918
[33] Yarlagadda Prasad K D V. Development of an integrated neural network system for prediction of process parameters in metal injection moulding[J]. Journal of Materials Processing Technology, 2002, 130:315～320
[34] 陈敏恒,丛德滋,方图南等.化工原理(第二版):上册[M].北京:化学工业出版社,1999
[35] 蒋维钧,戴猷元,顾惠君等.化工原理:上册[M].北京:清华大学出版社,1992
[36] 成都科技大学化工原理教研室编著.化工原理:上册[M].成都:成都科技大学出版社,1986
[37] 王科俊,王克成.神经网络建模、预报与控制[M].哈尔滨:哈尔滨工程大学出版社,1996
[38] 胡军,梅炽等.吹炼炉入炉铜锍品位的灰色预测[J].中国有色金属学报,2000,10(5):736～739
[39] 王军平,樊文侠,敬忠良等.一种构建灰色预测模型的新算法[J].西北大学学报(自然科学版),2002,32(6):613～616
[40] 罗开元,高峰,胡俐蕊.灰色广义预测控制算法及仿真研究[J].控制理论与应用,2002,19(2):207～210
[41] 邹健,杨莹春,诸静.基于灰色模型的预测模糊控制策略及其应用研究[J].中国电机工程学报,2002,22(9):12～14
[42] 吕安民,柯美忠,李成名等.灰色系统模型在时间序列模式中的应用研究[J].微机发展,2002,12(5):41～43
[43] 唐朝晖,桂卫华,吴敏等.基于神经网络和灰色理论的密闭鼓风炉透气性预测模型[J].中国有色金属学报,2003,13(5):1306～1310
[44] 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[J].自动化学报,2000,26(2):226～232
[45] (美)David J. Kruglinski. Visual C++技术内幕[M].北京:清华大学出版社,1999
[46] 希望图书创作室译.Visual C++6.0技术内幕[M].北京:北京希望电子出版社,1995
[47] 侯俊杰.深入浅出MFC第2版[M].武汉:华中科技大学出版社,2002
[48] 何海江.OPC客户端关健技术的实现[J].微计算机信息,2003,19(7):76～78
[49] 马增良,兰斌.OPC数据访问服务器实现机制研究[J].计算机工程与应用,2003,39(21):65～67,71
[50] 郑城荣,段来盛.Microsoft Access2.0开发指南[M].北京:人民邮电出版社,1999
[51] 张晓黎,吴崇峰,高雨茁.ODBC应用于注塑模浇注系统的二次开发[J].塑料科技,2003,0(1):12～14
[52] 李天宏,薛安,杨海宏.采用ODBC技术实现数据库与矢量图形的相互查询[J].测绘信息与工程,1997,3:11～14
[53] 巩伟,宋勇江,陈庆文.组态软件RSView32与VB、EXCEL结合的复杂生产报表解决方案[J].自动化技术与应用,2003,22(9):53～55