CIPS结构下炼铁厂集成自动化系统及高炉铁水硅含量预报实现
|
摘要
近年来，CIMS(Computer Integrated Manufacturing Systems，计算机集成制造系统)在离散行业中的应用越来越广泛。与此同时，国际市场的竞争日益激烈，迫使企业加速推出质量好、成本低的产品，以期更有力地参与市场竞争。于是，流程工业开始在本企业执行CIPS(Computer Integrated Process System，计算机集成过程系统)，即CIMS在流程工业的应用。本文旨在研究CIPS在钢铁行业中的应用。
由于CIPS源于CIMS，因此本文首先简要介绍了CIMS的相关内容，包括CIMS的概念、构成、实施等。接着给出了CIPS的定义，讨论了流程工业集成自动化的发展过程，研究了CIPS的内容、体系结构、关键理论技术，并对CIPS的研究进行了展望。
然后，在对钢铁企业生产过程和流程特点分析的基础上，对钢铁企业CIPS进行了具体的研究。在此基础上，鉴于我国钢铁行业整体自动化和管理水平较低的现状，从功能和控制两个方面提出了一种CIPS的4层体系结构，并提出了如何具体实施CIPS。
最后，以CIPS为指导，建立了某钢铁厂炼铁分厂高炉集成自动化系统，并构造了高炉铁水硅含量预报模型。前者采用SCADA／HMI(Supervisory Control And Data Acquisition／Human Machine Interface，监控及数据采集／人机界面)实现系统功能，网络连接采用工业以太网，以实现数据的透明化。后者基于BP神经网络建模，在对传统的BP算法加以改进的基础上，结合时间序列的方法建立神经网络预报模型。
经实际投产运行后，该集成自动化系统效果显著，铁水硅含量预报模型预报效果理想，达到了预期目标。
In recent years, the CIMS (Computer Integrated Manufacturing Systems) has been applied in the discrete industry wider and wider. At the same time, in the face of the increased international competition in the global marketplace which has heightened the need to develop high quality and low cost products that can compete, the flow industry attempted to implement the CIPS (Computer Integrated Process System), i.e. CIMS application in the flow process industry. This paper aims at studying the application of CIPS in the steel trade.
The CIMS are first introduced briefly in this paper, as CIPS is derived from CIMS. The content and structure of CIMS are presented. Then the CIPS is defined, and the development of the integrated automation in the flow industry is discussed. After the detailed comparison between the CIPS and CIMS, the primary content and and the key theories and technologies of CIPS are investigated.
Next, based on the analyzing of the operations of the steel trade, the CIPS of the steel trade is studied. Considering the actuality of our country's steel trade, A CIPS 4 layers systematic structure is presented. The implement of CIPS in the steel trade is also studied.
In the end of this paper, according to the principle of CIPS, the iron-smeltery integrated automation system is build. Furthermore, a time-series prediction model is developed to predict silicon content in hot metal. The former uses SCADA/HMI(Supervisory Control And Data Acquisition/ Human Machine Interface) to implement system function, and the network is connected by industry Ethernet. The latter is based on improved BP neural network.
The result shows the integrated automation system works effectively and the prediction effect of the model is ideal.
由于CIPS源于CIMS，因此本文首先简要介绍了CIMS的相关内容，包括CIMS的概念、构成、实施等。接着给出了CIPS的定义，讨论了流程工业集成自动化的发展过程，研究了CIPS的内容、体系结构、关键理论技术，并对CIPS的研究进行了展望。
然后，在对钢铁企业生产过程和流程特点分析的基础上，对钢铁企业CIPS进行了具体的研究。在此基础上，鉴于我国钢铁行业整体自动化和管理水平较低的现状，从功能和控制两个方面提出了一种CIPS的4层体系结构，并提出了如何具体实施CIPS。
最后，以CIPS为指导，建立了某钢铁厂炼铁分厂高炉集成自动化系统，并构造了高炉铁水硅含量预报模型。前者采用SCADA／HMI(Supervisory Control And Data Acquisition／Human Machine Interface，监控及数据采集／人机界面)实现系统功能，网络连接采用工业以太网，以实现数据的透明化。后者基于BP神经网络建模，在对传统的BP算法加以改进的基础上，结合时间序列的方法建立神经网络预报模型。
经实际投产运行后，该集成自动化系统效果显著，铁水硅含量预报模型预报效果理想，达到了预期目标。
In recent years, the CIMS (Computer Integrated Manufacturing Systems) has been applied in the discrete industry wider and wider. At the same time, in the face of the increased international competition in the global marketplace which has heightened the need to develop high quality and low cost products that can compete, the flow industry attempted to implement the CIPS (Computer Integrated Process System), i.e. CIMS application in the flow process industry. This paper aims at studying the application of CIPS in the steel trade.
The CIMS are first introduced briefly in this paper, as CIPS is derived from CIMS. The content and structure of CIMS are presented. Then the CIPS is defined, and the development of the integrated automation in the flow industry is discussed. After the detailed comparison between the CIPS and CIMS, the primary content and and the key theories and technologies of CIPS are investigated.
Next, based on the analyzing of the operations of the steel trade, the CIPS of the steel trade is studied. Considering the actuality of our country's steel trade, A CIPS 4 layers systematic structure is presented. The implement of CIPS in the steel trade is also studied.
In the end of this paper, according to the principle of CIPS, the iron-smeltery integrated automation system is build. Furthermore, a time-series prediction model is developed to predict silicon content in hot metal. The former uses SCADA/HMI(Supervisory Control And Data Acquisition/ Human Machine Interface) to implement system function, and the network is connected by industry Ethernet. The latter is based on improved BP neural network.
The result shows the integrated automation system works effectively and the prediction effect of the model is ideal.
引文
[1] 董名垂．CIMS系统集成．中国计算机报，1995，(15)：99-117
[2] 吴澄．CIMS在我国的进展：研究、应用和经验．计算机集成制造系统，1995，(4)：1-6
[3] 宋国宁，蒋新松．连续生产过程CIMS的研究．计算机集成制造系统CIMS，1995，1(1)：7-12
[4] Pratap K.Nair,FrankB. Canfield.Consider Computer Integrated Manufacturing for Continuous Process Plants. Chemical Engineering Progress, 1992,(11):71-81.
[5] R AOM,Wang Q. Computer integrated process systems in continuous manufacturing industrys. Computer Integred Manufacturing Systems,1993,6(4):260-272
[6] 秦斌，王欣编著．流程工业计算机集成过程系统CIPS．湖南：国防大学出版社，2002，4-6
[7] 柴天佑，李平，桂卫华．流程工业CIMS的研究现状与对策．见：CIMS-China’98论文集，1998，48-54
[8] 熊刚，许晓鸣，张钟俊．努力加强流程工业综合自动化的理论研究与实践．计算机集成制造系统CIMS，1995，1(3)：32～35
[9] 周明，秦民生．高炉铁水含硅预报数学模型．钢铁，1990，21(5)：8-12
[10] 孙克勤．高炉铁水含硅量预报自适应数学模型的研制与试验．钢铁，1990，24(6)：4-8
[11] Saxen H,Karilainen L. Model for Short Term Prediction of Silicon Content in the Furnace Process. in:Ironmaking Conference Proceedings,1992,185-191
[12] 刘永宽．美国普度大学CIM递阶控制系统参考模型简介．信息与控制，1994，23(6)：345-350
[13] T.J.Williams. The Purdue Enterprise Reference Architecture. Computer in Industry, 1994, (24):141-158
[14] Williams T J. Architecture for integrating manufacturing activities and enterprises. In:Theodore J Williams.in:Proceedings of 12th World Congress IFAC International Federation of Automatic Control. USA:West Lafayette,Indiana, 1993,273-283
[15] 赵欣宇，柴天佑，赵新力．CIMS与CIPS体系结构对比分析的研究．信息与控制，1998，27(4)：304-309
[16] 曲润涛，彭健，杜秀华．CIPS和CIMS中关键技术的比较．化工自动化及仪表，1998，25(5)：23～26
[17] 王锦标．连续过程工业CIPS四维体系结构的研究．化工自动化及仪表．1996，23(5)：30-34
[18] 张海盛．连续制造业CIMS结构初探．见：第二届中国CIMS学术会议论文集CIMS-China．国家高技术计划自动化领域CIMS主题专家组，1992，8-10
[19] 宋国宁，蒋新松，刘宝义．大型过程工业自动化全过程体系结构．信息与控，1994，23(2)：65-70
[20] M Rao,Q Wang,L Yuan,M Zao. CIPS Architecture and implementation, in:Proceeding of Conference on Computer Integrated Manufacturing in Process Industries. 1994,25～26
[21] 赵欣宇，柴天佑，侯斌．CIPS体系结构的研究．东北大学学报(自然科学版)，1999，20(2)：137-139
[22] 侍洪波，黄道．CIPS系统的若干关键技术难题．微型电脑应用，1999，(4)：20-22
[23] 邓守强．高炉炼铁技术．北京：冶金工业出版社，1990，55-57
[24] 那树人．炼铁工艺计算[M]．北京：冶金工业出版社，1999，130-132
[25] 刘祥官，刘芳．高炉过程优化与智能控制模型．高校应用数学学报A辑，2001，16(4)：462-470．
[26] 孙铁栋，杨章远，许志宏等．神经网络方法在预报高炉铁水硅含量上的应用研究．钢铁[J]，1996，31(3)：18-20
[27] 王玉涛，周建常，王师．神经网络模型与时差方法结合预报铁水硅含量．钢铁[J]，2000，35(3)：32-35
[28] 胡守仁，余少波，戴葵．神经网络导论．长沙：国防科技大学出版社，1992，15-20
[29] 施鸿宝．神经网络及其应用．西安：西安交通大学出版社，1993，38-40
[30] Tesauro G. Practical Issues in Temporal Difference Learning. Machine Learning, 1992, (3):257-277
[31] Masumi Ishikawa. Prediction of Time Series by a Structural Learning of Neural Network. Fuzzy Sets and System,1996,8(2):167-176
[32] Guangqi Zuo,Jitang Ma,Bjorkman Bo. A Neural Network Modal for Predicting the Silicon Content of the HotMetal at No.2 Blast furnace of SSABLULEA. in:Ironmaking Conference Proceedings,1996,211-221
[33] 刘金琨，邓守强，苏士权．高炉铁水硅含量的神经网络时间序列预报[J]．钢铁研究学报，1996，8(3)：63-66．
[34] 周后盘，刘衍聪，应道宁．CIMS网络子系统的规划设计[J]．机械工业自动化，1997，2(16)：1-4．
[35] 刘宝义．CIM与工业自动化总体技术．它的目标、方法和实践．信息与控制，1993，22(1)：29-35
[36] 梁昱，徐用懋．连续工业CIMS实时信息集成应用．测控技术，1999，2(18)：42-43．
[37] 曲润涛，彭健，杜秀华，冯为民．基于CIPS体系结构的信息模型的实现．化工自动化及仪表，1998，25(2)：27～30
[38] 韩庆兰，高阳，陈松桥．论钢铁企业实施CIMS工程的管理模式及信息集成方案[J]．钢铁，1998，10(33)：73-76．
[39] Luis Puigjaner,Albert Huercio,Antonio Espuna. Batch Production Control in a Computer Integrated Manufacturing Environment J.Proc.Cont. 1994,4(4):281-290
[40] CEN TC310WG1. An Evaluation of CIM Modeling Constructs Evaluation Report of Constructs for Views According to ENV 40003. Computer in Industry,. 1994,(24):159-236
[41] 毕学工．人工智能和专家系统在钢铁工业中的应用[J]．武汉钢铁学院学报，1995，18(2)：146-154．
[42] 马竹梧．专家系统开发工具及其应用[J]．基础自动化，1998，5(5)：6-9．
[43] 杨天钧，杨尚宝．人工智能高炉冶炼专家系统．北京科技大学学报[J]，1994，16：42-48
[44] 秦民生，杨天钧．炼铁过程的解析与模拟．北京：冶金工业出版社，1991，12-15
[45] 章天华，鲁世英．炼铁[M]．北京：冶金工业出版社，1986，455-464
[46] S M Pandit,J A Clum,S M Wu. Modelling,Prediction and Control of Blast Furnace Operation from Observed Data by Multivariate Time Series. in:Ironmaking Proceedings,Toronto, 1975,403-416
[47] Hunt K J. Extending the Functional Equivalence of Radial Basis Function Networksand Fuzzy Inference System. IEEE Transaction on NeuralNetworks[J], 1996,(3): 776-781
[48] Masumi Ishikawa. Learning of Modular Structured Network. Artificial Intelligence, 1995,75: 51-62
[49] Sutton R S. Learning to Predict by the Methods of Temporal Difference. Machine Learning, 1983,(3):9-44
[2] 吴澄．CIMS在我国的进展：研究、应用和经验．计算机集成制造系统，1995，(4)：1-6
[3] 宋国宁，蒋新松．连续生产过程CIMS的研究．计算机集成制造系统CIMS，1995，1(1)：7-12
[4] Pratap K.Nair,FrankB. Canfield.Consider Computer Integrated Manufacturing for Continuous Process Plants. Chemical Engineering Progress, 1992,(11):71-81.
[5] R AOM,Wang Q. Computer integrated process systems in continuous manufacturing industrys. Computer Integred Manufacturing Systems,1993,6(4):260-272
[6] 秦斌，王欣编著．流程工业计算机集成过程系统CIPS．湖南：国防大学出版社，2002，4-6
[7] 柴天佑，李平，桂卫华．流程工业CIMS的研究现状与对策．见：CIMS-China’98论文集，1998，48-54
[8] 熊刚，许晓鸣，张钟俊．努力加强流程工业综合自动化的理论研究与实践．计算机集成制造系统CIMS，1995，1(3)：32～35
[9] 周明，秦民生．高炉铁水含硅预报数学模型．钢铁，1990，21(5)：8-12
[10] 孙克勤．高炉铁水含硅量预报自适应数学模型的研制与试验．钢铁，1990，24(6)：4-8
[11] Saxen H,Karilainen L. Model for Short Term Prediction of Silicon Content in the Furnace Process. in:Ironmaking Conference Proceedings,1992,185-191
[12] 刘永宽．美国普度大学CIM递阶控制系统参考模型简介．信息与控制，1994，23(6)：345-350
[13] T.J.Williams. The Purdue Enterprise Reference Architecture. Computer in Industry, 1994, (24):141-158
[14] Williams T J. Architecture for integrating manufacturing activities and enterprises. In:Theodore J Williams.in:Proceedings of 12th World Congress IFAC International Federation of Automatic Control. USA:West Lafayette,Indiana, 1993,273-283
[15] 赵欣宇，柴天佑，赵新力．CIMS与CIPS体系结构对比分析的研究．信息与控制，1998，27(4)：304-309
[16] 曲润涛，彭健，杜秀华．CIPS和CIMS中关键技术的比较．化工自动化及仪表，1998，25(5)：23～26
[17] 王锦标．连续过程工业CIPS四维体系结构的研究．化工自动化及仪表．1996，23(5)：30-34
[18] 张海盛．连续制造业CIMS结构初探．见：第二届中国CIMS学术会议论文集CIMS-China．国家高技术计划自动化领域CIMS主题专家组，1992，8-10
[19] 宋国宁，蒋新松，刘宝义．大型过程工业自动化全过程体系结构．信息与控，1994，23(2)：65-70
[20] M Rao,Q Wang,L Yuan,M Zao. CIPS Architecture and implementation, in:Proceeding of Conference on Computer Integrated Manufacturing in Process Industries. 1994,25～26
[21] 赵欣宇，柴天佑，侯斌．CIPS体系结构的研究．东北大学学报(自然科学版)，1999，20(2)：137-139
[22] 侍洪波，黄道．CIPS系统的若干关键技术难题．微型电脑应用，1999，(4)：20-22
[23] 邓守强．高炉炼铁技术．北京：冶金工业出版社，1990，55-57
[24] 那树人．炼铁工艺计算[M]．北京：冶金工业出版社，1999，130-132
[25] 刘祥官，刘芳．高炉过程优化与智能控制模型．高校应用数学学报A辑，2001，16(4)：462-470．
[26] 孙铁栋，杨章远，许志宏等．神经网络方法在预报高炉铁水硅含量上的应用研究．钢铁[J]，1996，31(3)：18-20
[27] 王玉涛，周建常，王师．神经网络模型与时差方法结合预报铁水硅含量．钢铁[J]，2000，35(3)：32-35
[28] 胡守仁，余少波，戴葵．神经网络导论．长沙：国防科技大学出版社，1992，15-20
[29] 施鸿宝．神经网络及其应用．西安：西安交通大学出版社，1993，38-40
[30] Tesauro G. Practical Issues in Temporal Difference Learning. Machine Learning, 1992, (3):257-277
[31] Masumi Ishikawa. Prediction of Time Series by a Structural Learning of Neural Network. Fuzzy Sets and System,1996,8(2):167-176
[32] Guangqi Zuo,Jitang Ma,Bjorkman Bo. A Neural Network Modal for Predicting the Silicon Content of the HotMetal at No.2 Blast furnace of SSABLULEA. in:Ironmaking Conference Proceedings,1996,211-221
[33] 刘金琨，邓守强，苏士权．高炉铁水硅含量的神经网络时间序列预报[J]．钢铁研究学报，1996，8(3)：63-66．
[34] 周后盘，刘衍聪，应道宁．CIMS网络子系统的规划设计[J]．机械工业自动化，1997，2(16)：1-4．
[35] 刘宝义．CIM与工业自动化总体技术．它的目标、方法和实践．信息与控制，1993，22(1)：29-35
[36] 梁昱，徐用懋．连续工业CIMS实时信息集成应用．测控技术，1999，2(18)：42-43．
[37] 曲润涛，彭健，杜秀华，冯为民．基于CIPS体系结构的信息模型的实现．化工自动化及仪表，1998，25(2)：27～30
[38] 韩庆兰，高阳，陈松桥．论钢铁企业实施CIMS工程的管理模式及信息集成方案[J]．钢铁，1998，10(33)：73-76．
[39] Luis Puigjaner,Albert Huercio,Antonio Espuna. Batch Production Control in a Computer Integrated Manufacturing Environment J.Proc.Cont. 1994,4(4):281-290
[40] CEN TC310WG1. An Evaluation of CIM Modeling Constructs Evaluation Report of Constructs for Views According to ENV 40003. Computer in Industry,. 1994,(24):159-236
[41] 毕学工．人工智能和专家系统在钢铁工业中的应用[J]．武汉钢铁学院学报，1995，18(2)：146-154．
[42] 马竹梧．专家系统开发工具及其应用[J]．基础自动化，1998，5(5)：6-9．
[43] 杨天钧，杨尚宝．人工智能高炉冶炼专家系统．北京科技大学学报[J]，1994，16：42-48
[44] 秦民生，杨天钧．炼铁过程的解析与模拟．北京：冶金工业出版社，1991，12-15
[45] 章天华，鲁世英．炼铁[M]．北京：冶金工业出版社，1986，455-464
[46] S M Pandit,J A Clum,S M Wu. Modelling,Prediction and Control of Blast Furnace Operation from Observed Data by Multivariate Time Series. in:Ironmaking Proceedings,Toronto, 1975,403-416
[47] Hunt K J. Extending the Functional Equivalence of Radial Basis Function Networksand Fuzzy Inference System. IEEE Transaction on NeuralNetworks[J], 1996,(3): 776-781
[48] Masumi Ishikawa. Learning of Modular Structured Network. Artificial Intelligence, 1995,75: 51-62
[49] Sutton R S. Learning to Predict by the Methods of Temporal Difference. Machine Learning, 1983,(3):9-44
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |