密闭鼓风炉炉况智能优化操作指导系统
|
摘要
铅锌密闭鼓风炉熔炼过程反应机理复杂,具有强非线性、强耦合性、不确定性、时变时滞性、多约束、多输入多输出等特点,本文深入研究了密闭鼓风炉炉况识别策略和炉况优化操作指导技术,建立了密闭鼓风炉炉况智能优化操作指导系统。
鼓风炉炉况是一个综合反映铅锌密闭鼓风炉冶炼工艺中熔炼过程进行好坏的指标。论文提出了一种炉况智能识别策略,它采用数学模型结合知识模型、模糊控制与专家系统相结合的方法进行鼓风炉炉况的识别,确定是正常炉况还是异常炉况。利用综合透气性和炉渣指标确定炉况的正常与异常,如果是正常炉况则保持当前操作,如果是异常炉况则综合热风压力和风量、偏料、炉气(方箱处测)、下料速度等指标判断发生何种设备故障或工艺故障。
在炉况识别的基础上,提出了炉况智能优化操作指导策略。采用灰色理论模型结合神经网络模型进行透气性的预测,采用聚类搜索结合专家系统进行正常工作模式下的优化操作,针对故障模式则采用基于产生式规则的专家系统,利用双向推理进行故障的识别和故障处理(故障显示、报警、故障原因、专家操作指导)。
系统投入运行后,有效地抑制了透气性的波动,产量质量得到了很大的提高,取得了显著的经济效益和社会效益。
The Lead-Zinc Imperial Smelting Process is a complex mechanism, with the characteristics of strong nonlinear, strong coupling, uncertain, time varying and time-delay, multi-constrained, multi-input and multi-output. The estimation strategy and operation guide technique of ISF state are mainly involved in this thesis, so as to establish the intelligent optimization operation guide system of ISF state.The furnace state is a synthetical value, which reflects the ongoing degree of smelting process in the Lead-Zinc Imperial Smelting Process. An intelligent estimation strategy of furnace state proposed in the paper can estimate the normal state or the abnormal state of ISF state through using compositive method of mathematics model & knowledge method, and fuzzy control & expert system. The compositive permeability and slag index are the main reference to estimate the furnace state. If the real-time state is good, then the operation values remain; otherwise, the synthetical values, such as pressure and flux of hot fan, gradient material distribution, furnace air (square trunk), velocity of material feeding, become the major factors to estimate which kind of equipments faults or craftwork faults are arisen.The intelligent optimization operation guide strategy of furnace state is proposed on the base of estimation method of furnace state. The compositive model of grey theory model and neural network model is built to predict the permeability. The clustering searching and expert system are used to realize the state optimization operation in normal mode. In fault mode, the expert system, basing on produce-mode rules, adopts bi-directional ratiocination to realize estimation and disposal of faults (faults display, alert, faults reason display, expert operation guides).The application of the intelligent optimization operation guide system shows that it can restrain the fluctuation of permeability effectively, improve the output and quality of products largely, and reap the remarkable economic and social benefit.
鼓风炉炉况是一个综合反映铅锌密闭鼓风炉冶炼工艺中熔炼过程进行好坏的指标。论文提出了一种炉况智能识别策略,它采用数学模型结合知识模型、模糊控制与专家系统相结合的方法进行鼓风炉炉况的识别,确定是正常炉况还是异常炉况。利用综合透气性和炉渣指标确定炉况的正常与异常,如果是正常炉况则保持当前操作,如果是异常炉况则综合热风压力和风量、偏料、炉气(方箱处测)、下料速度等指标判断发生何种设备故障或工艺故障。
在炉况识别的基础上,提出了炉况智能优化操作指导策略。采用灰色理论模型结合神经网络模型进行透气性的预测,采用聚类搜索结合专家系统进行正常工作模式下的优化操作,针对故障模式则采用基于产生式规则的专家系统,利用双向推理进行故障的识别和故障处理(故障显示、报警、故障原因、专家操作指导)。
系统投入运行后,有效地抑制了透气性的波动,产量质量得到了很大的提高,取得了显著的经济效益和社会效益。
The Lead-Zinc Imperial Smelting Process is a complex mechanism, with the characteristics of strong nonlinear, strong coupling, uncertain, time varying and time-delay, multi-constrained, multi-input and multi-output. The estimation strategy and operation guide technique of ISF state are mainly involved in this thesis, so as to establish the intelligent optimization operation guide system of ISF state.The furnace state is a synthetical value, which reflects the ongoing degree of smelting process in the Lead-Zinc Imperial Smelting Process. An intelligent estimation strategy of furnace state proposed in the paper can estimate the normal state or the abnormal state of ISF state through using compositive method of mathematics model & knowledge method, and fuzzy control & expert system. The compositive permeability and slag index are the main reference to estimate the furnace state. If the real-time state is good, then the operation values remain; otherwise, the synthetical values, such as pressure and flux of hot fan, gradient material distribution, furnace air (square trunk), velocity of material feeding, become the major factors to estimate which kind of equipments faults or craftwork faults are arisen.The intelligent optimization operation guide strategy of furnace state is proposed on the base of estimation method of furnace state. The compositive model of grey theory model and neural network model is built to predict the permeability. The clustering searching and expert system are used to realize the state optimization operation in normal mode. In fault mode, the expert system, basing on produce-mode rules, adopts bi-directional ratiocination to realize estimation and disposal of faults (faults display, alert, faults reason display, expert operation guides).The application of the intelligent optimization operation guide system shows that it can restrain the fluctuation of permeability effectively, improve the output and quality of products largely, and reap the remarkable economic and social benefit.
引文
[1] 《铜铅锌冶炼参考设计资料》编写组.《铜铅锌冶炼参考设计资料》.北京:冶金工业出版社,1979
[2] 韶冶教培处.韶冶工人岗位培训教程.韶关冶炼厂,1994
[3] 刘洪霖,包宏著.化工冶金过程人工智能优化.北京:冶金工业出版社,1999
[4] 金川镍闪速炉计算机在线控制专题组.金川镍闪速炉计算机在线控制技术研究报告,1995
[5] 万维汉,万百五,杨金义.闪速炉的神经网络冰镍质:量模型与稳态优化控制研究.自动化学报,1999,25(6):800-803
[6] 史卓群.“ISP”工艺发展的主要特点.鼓风炉炼铅锌,1990,3
[7] 傅世敏等.高炉过程气体动力学.冶金工业出版社,1990.毕学工,高炉气体动力学.冶金工业出版社,1996
[8] 杨尚宝,杨天钧.基于神经网络的高炉异常炉况判断专家系统.北京科技大学学报,1994,16(6):517-521
[9] 李景银,郭宏飞,周伟.高炉异常炉况诊断专家系统的设计与实现.东北大学学报(自然科学版),1997,18(2):178-182
[10] 杨尚宝,杨天钧,董一诚.神经网络高炉炉况预测与判断专家系统.北京科技大学学报,1996,18(3):220-225
[11] Dubois D, Prade H. Fuzzy Sets, and Systems: Theory and Appliction Academic Press, New York,1980
[12] Takagi T, Sugeno M. Fuzzy Identification of Systems and its applications to modeling and control. IEEE Tran. SMC 1985, 15(1):116-132
[13] 唐加福,汪定伟.模糊优化理论与方法的研究综述.控制理论与应用,2000,17(2):159-164
[14] 刘曙光,王志宏,费佩燕,王斌.模糊控制的发展与展望.2000,17(1):9-12
[15] 郭永红,魏庆福.模糊控制在工程应用中的实现方法.1995,11 (5):3-6
[16] Wu Min, Nakano Mieho, She Jin-Hua. A Model-based Expert Control System for the Leaching Process in Zinc Hydrometallurgy. Expert Systems with Applications, 1999, 16(1):135-1431
[17] Wu Min, Nakano Micho, She Jin-Hua. An Expert Control Strategy Using Neural Networks for the Electrolytic Process in Zinc Hydrometallurgy. Proc. the 1999 IEEE CCA, 1999, 2: 1044-1049
[18] 方醉敏,赵捧,王春鸣,臧年顺.专家模糊控制算法及其实现.自动化仪表,1995,2:20-22
[19] 吴敏,桂卫华.锌浸出过程的专家控制和故障诊断方法.控制理论与应用,2001,18(6):861-866
[20] 傅世敏等.高炉过程气体动力学.冶金工业出版社,1990
[21] 毕学工.高炉气体动力学.冶金工业出版社,1996
[22] 冯地清,雷明.模糊控制技术新进展.自动化博览,2000,1
[23] 夏宏泉等.基于灰色神经网络的测井预测地层破裂压力.西南石油学院学报,1996,4
[24] 朱红求,密闭鼓风炉熔炼过程透气性预测智能集成建模研究:[硕士学位论文].长沙:中南大学,2000
[25] 邓聚龙.灰色控制系统。华中工学院出版社,1987
[26] 邓聚龙.灰色系统理论教程.华中理工大学出版社,1990
[27] 毕效辉,姚琼荟.灰色预测在过程控制中的应用.西南工学院学报,1997,3
[28] 夏宏泉等.基于灰色神经网络的测井预测地层破裂压力.西南石油学院学报,1996,4
[29] 冯建生,王秀芝.一个基于神经网络的配矿专家系统.冶金自动化,1999,23(4):7
[30] 谢安国,陆钟武.神经网络BP模型在烧结工序能耗分析中的应用.冶金能源,1998,17(5):8-10
[31] 胡守仁,余少波,戴葵.神经网络导论.国防科技大学出版社,1993
[32] 戚德虎,康继昌.BP神经网络的设计.计算机工程与设计,1998,19(2):48-50
[33] 吕宏辉,钟络,夏红霞.灰色系统与神经网络融合技术探索.微机发展,2000,3:3-5
[34] 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:[博士学位论文].长沙:中南大学,2001
[35] 中山大学数学力学系.概率论及数理统计.高等教育出版社,1980,4
[36] 李文奇.概率论与数理统计.北京师范大学出版社,1990;杨德保.函数概率论与数理统计.北京工业学院出版社,1985:84-97
[37] 仲蔚,俞金寿.基于模糊c均值聚类的多模型软测量建模.华东理工大学学报,2000,26(1):83-87
[38] 姚宏伟,梅晓榕,庄显义.模糊聚类分析在模糊神经网络结构优化中的应用.高技术通讯,2000,10:64-66
[39] 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:[博士学位论文].长沙:中南大学,2004
[40] 桂卫华,阳春华,吴敏等.智能集成优化控制技术及其在工业中的应用.有色冶金设计与研究,2003,24:6-14
[41] 王永初,任秀珍.工业过程控制系统设计范例.科学出版社,1986
[42] 颜纶亮等.微机在过程控制中的应用.清华大学出版社,1989
[43] 裴景玉.基于Windows95平台的实时控制技术.工业控制计算机,2000,3
[44] 王耀南,王辉,彭建春等.复杂工业过程的综合集成智能控制.信息与控制,1999,28(4):298-304
[45] David J.Kruglinsld.Visual C++技术内幕.北京:清华大学出版社,1999
[46] 汤天浩.智能监控系统:新的理论、方法与进展.上海海运学院学报,2001,22(3):17-21
[2] 韶冶教培处.韶冶工人岗位培训教程.韶关冶炼厂,1994
[3] 刘洪霖,包宏著.化工冶金过程人工智能优化.北京:冶金工业出版社,1999
[4] 金川镍闪速炉计算机在线控制专题组.金川镍闪速炉计算机在线控制技术研究报告,1995
[5] 万维汉,万百五,杨金义.闪速炉的神经网络冰镍质:量模型与稳态优化控制研究.自动化学报,1999,25(6):800-803
[6] 史卓群.“ISP”工艺发展的主要特点.鼓风炉炼铅锌,1990,3
[7] 傅世敏等.高炉过程气体动力学.冶金工业出版社,1990.毕学工,高炉气体动力学.冶金工业出版社,1996
[8] 杨尚宝,杨天钧.基于神经网络的高炉异常炉况判断专家系统.北京科技大学学报,1994,16(6):517-521
[9] 李景银,郭宏飞,周伟.高炉异常炉况诊断专家系统的设计与实现.东北大学学报(自然科学版),1997,18(2):178-182
[10] 杨尚宝,杨天钧,董一诚.神经网络高炉炉况预测与判断专家系统.北京科技大学学报,1996,18(3):220-225
[11] Dubois D, Prade H. Fuzzy Sets, and Systems: Theory and Appliction Academic Press, New York,1980
[12] Takagi T, Sugeno M. Fuzzy Identification of Systems and its applications to modeling and control. IEEE Tran. SMC 1985, 15(1):116-132
[13] 唐加福,汪定伟.模糊优化理论与方法的研究综述.控制理论与应用,2000,17(2):159-164
[14] 刘曙光,王志宏,费佩燕,王斌.模糊控制的发展与展望.2000,17(1):9-12
[15] 郭永红,魏庆福.模糊控制在工程应用中的实现方法.1995,11 (5):3-6
[16] Wu Min, Nakano Mieho, She Jin-Hua. A Model-based Expert Control System for the Leaching Process in Zinc Hydrometallurgy. Expert Systems with Applications, 1999, 16(1):135-1431
[17] Wu Min, Nakano Micho, She Jin-Hua. An Expert Control Strategy Using Neural Networks for the Electrolytic Process in Zinc Hydrometallurgy. Proc. the 1999 IEEE CCA, 1999, 2: 1044-1049
[18] 方醉敏,赵捧,王春鸣,臧年顺.专家模糊控制算法及其实现.自动化仪表,1995,2:20-22
[19] 吴敏,桂卫华.锌浸出过程的专家控制和故障诊断方法.控制理论与应用,2001,18(6):861-866
[20] 傅世敏等.高炉过程气体动力学.冶金工业出版社,1990
[21] 毕学工.高炉气体动力学.冶金工业出版社,1996
[22] 冯地清,雷明.模糊控制技术新进展.自动化博览,2000,1
[23] 夏宏泉等.基于灰色神经网络的测井预测地层破裂压力.西南石油学院学报,1996,4
[24] 朱红求,密闭鼓风炉熔炼过程透气性预测智能集成建模研究:[硕士学位论文].长沙:中南大学,2000
[25] 邓聚龙.灰色控制系统。华中工学院出版社,1987
[26] 邓聚龙.灰色系统理论教程.华中理工大学出版社,1990
[27] 毕效辉,姚琼荟.灰色预测在过程控制中的应用.西南工学院学报,1997,3
[28] 夏宏泉等.基于灰色神经网络的测井预测地层破裂压力.西南石油学院学报,1996,4
[29] 冯建生,王秀芝.一个基于神经网络的配矿专家系统.冶金自动化,1999,23(4):7
[30] 谢安国,陆钟武.神经网络BP模型在烧结工序能耗分析中的应用.冶金能源,1998,17(5):8-10
[31] 胡守仁,余少波,戴葵.神经网络导论.国防科技大学出版社,1993
[32] 戚德虎,康继昌.BP神经网络的设计.计算机工程与设计,1998,19(2):48-50
[33] 吕宏辉,钟络,夏红霞.灰色系统与神经网络融合技术探索.微机发展,2000,3:3-5
[34] 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:[博士学位论文].长沙:中南大学,2001
[35] 中山大学数学力学系.概率论及数理统计.高等教育出版社,1980,4
[36] 李文奇.概率论与数理统计.北京师范大学出版社,1990;杨德保.函数概率论与数理统计.北京工业学院出版社,1985:84-97
[37] 仲蔚,俞金寿.基于模糊c均值聚类的多模型软测量建模.华东理工大学学报,2000,26(1):83-87
[38] 姚宏伟,梅晓榕,庄显义.模糊聚类分析在模糊神经网络结构优化中的应用.高技术通讯,2000,10:64-66
[39] 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:[博士学位论文].长沙:中南大学,2004
[40] 桂卫华,阳春华,吴敏等.智能集成优化控制技术及其在工业中的应用.有色冶金设计与研究,2003,24:6-14
[41] 王永初,任秀珍.工业过程控制系统设计范例.科学出版社,1986
[42] 颜纶亮等.微机在过程控制中的应用.清华大学出版社,1989
[43] 裴景玉.基于Windows95平台的实时控制技术.工业控制计算机,2000,3
[44] 王耀南,王辉,彭建春等.复杂工业过程的综合集成智能控制.信息与控制,1999,28(4):298-304
[45] David J.Kruglinsld.Visual C++技术内幕.北京:清华大学出版社,1999
[46] 汤天浩.智能监控系统:新的理论、方法与进展.上海海运学院学报,2001,22(3):17-21