基于生产周期特征的氧化铝生产调度优化系统设计
|
摘要
本文是以中国铝业贵州某分公司的现状为背景,针对氧化铝生产过程的特点以及存在的问题,通过对各生产工艺的分析和优化,全面整合氧化铝生产的底层控制系统之间以及底层控制系统和上层管理系统间的信息,提出了基于生产周期特征的氧化铝调度优化系统的解决方案,设计并实现氧化铝各工艺的优化方案和以数据库为核心的系统的基本结构框架。
围绕该课题,本文主要完成了如下几方面的研究工作:
①分析氧化铝生产调度优化系统的需求和特点,总结了目前氧化铝生产过程中存在问题和不足,提出了基于生产周期特征的氧化铝调度优化系统的解决方案;
②完成碱比、钙比、水分的生料浆质量指标预测控制结构图。主要包括配料工艺的分析及配料过程中存在的问题、生料浆质量的机理模型的搭建、基于模糊理论的生料浆智能修正模型得搭建;
③实现苛性比值和溶出率的精确在线检测。主要包括高压溶出机理的分析、二者机理模型和基于主元分析法的神经网络模型的搭建、复合神经网络算法的研究;模型在现场调试的效果;
④完成氧化铝调度优化系统基本框架的设计。主要包括系统网络设计、硬件设备的选型、实时/历史数据库平台的搭建及应用系统的开发;
本课题所研究的基于生产周期特征的氧化铝生产调度系统,较好地解决了上层管理与底层控制之间的“信息断层”和底层控制之间的“信息孤岛”问题,实现了生产管理静态和动态的统一、管理和控制的全方位统一。作为应用实例,本文所提出的基于生产周期特征的氧化铝生产调度系统已经在中国铝业贵州某分公司投入运行并取得了良好的运行效果。
On the background of production process of alumina in Guizhou Branch China Aluminium, the characteristics and existing problems, and through the analysis of the production process, the integrated optimization and control system of alumina production and the bottom layer between the upper management system and control system of information, based on the production cycle of alumina scheduling optimization system solutions was put forward, we design and implement the optimization of alumina process, and the basic framework of the system in database as the core.
On this topic, this paper mainly completed the following several aspects of the research work:
①Analysis the requirements and characteristics of alumina production scheduling system, and summarizes the current alumina production process problems and the faults, based on the production cycle of alumina scheduling optimization system solutions was put forward.
②Completing alkali ratio, calcium, water quality index predictive control charts. Mainly include ingredients of the analysis of the technology and the problems existing in the course of quality, the mechanism of the model and based on the theory of fuzzy raw mix slurry intelligent correction model were set;
③Realizing on-line accurate detection of ratio of soda to alumina(RSA) and leaching rate(LR). Mainly including high-pressure dissolving mechanism analysis, the mechanism model of RSA and LR.and based on principal component analysis of the neural network model was set,the research of composite neural network algorithm, the effect of the model in the commissioning;
④Designing the basic framework of the alumina scheduling optimization system. Mainly including the design of network system, the selection of hardware equipment, platform of real-time/historical database and the development of application system。
This topic is mainly research based on the production cycle of alumina production scheduling system,A good solution to the "information ground fault" between the upper management and the bottom control,and the "information island" problem"of the bottom control systems.Realizing the static and dynamic unification of the company production management and the comprehensive unification of the management and control,As the application example,Based on the alumina production scheduling system has been put into operation in guizhou branch, China Aluminium and achieved good effect.
围绕该课题,本文主要完成了如下几方面的研究工作:
①分析氧化铝生产调度优化系统的需求和特点,总结了目前氧化铝生产过程中存在问题和不足,提出了基于生产周期特征的氧化铝调度优化系统的解决方案;
②完成碱比、钙比、水分的生料浆质量指标预测控制结构图。主要包括配料工艺的分析及配料过程中存在的问题、生料浆质量的机理模型的搭建、基于模糊理论的生料浆智能修正模型得搭建;
③实现苛性比值和溶出率的精确在线检测。主要包括高压溶出机理的分析、二者机理模型和基于主元分析法的神经网络模型的搭建、复合神经网络算法的研究;模型在现场调试的效果;
④完成氧化铝调度优化系统基本框架的设计。主要包括系统网络设计、硬件设备的选型、实时/历史数据库平台的搭建及应用系统的开发;
本课题所研究的基于生产周期特征的氧化铝生产调度系统,较好地解决了上层管理与底层控制之间的“信息断层”和底层控制之间的“信息孤岛”问题,实现了生产管理静态和动态的统一、管理和控制的全方位统一。作为应用实例,本文所提出的基于生产周期特征的氧化铝生产调度系统已经在中国铝业贵州某分公司投入运行并取得了良好的运行效果。
On the background of production process of alumina in Guizhou Branch China Aluminium, the characteristics and existing problems, and through the analysis of the production process, the integrated optimization and control system of alumina production and the bottom layer between the upper management system and control system of information, based on the production cycle of alumina scheduling optimization system solutions was put forward, we design and implement the optimization of alumina process, and the basic framework of the system in database as the core.
On this topic, this paper mainly completed the following several aspects of the research work:
①Analysis the requirements and characteristics of alumina production scheduling system, and summarizes the current alumina production process problems and the faults, based on the production cycle of alumina scheduling optimization system solutions was put forward.
②Completing alkali ratio, calcium, water quality index predictive control charts. Mainly include ingredients of the analysis of the technology and the problems existing in the course of quality, the mechanism of the model and based on the theory of fuzzy raw mix slurry intelligent correction model were set;
③Realizing on-line accurate detection of ratio of soda to alumina(RSA) and leaching rate(LR). Mainly including high-pressure dissolving mechanism analysis, the mechanism model of RSA and LR.and based on principal component analysis of the neural network model was set,the research of composite neural network algorithm, the effect of the model in the commissioning;
④Designing the basic framework of the alumina scheduling optimization system. Mainly including the design of network system, the selection of hardware equipment, platform of real-time/historical database and the development of application system。
This topic is mainly research based on the production cycle of alumina production scheduling system,A good solution to the "information ground fault" between the upper management and the bottom control,and the "information island" problem"of the bottom control systems.Realizing the static and dynamic unification of the company production management and the comprehensive unification of the management and control,As the application example,Based on the alumina production scheduling system has been put into operation in guizhou branch, China Aluminium and achieved good effect.
引文
[1]毕诗文.氧化铝生产工艺[M].北京:化学工业出版社, 2006.
[2]毕诗文,于海燕等.拜耳法生产氧化铝[M].北京::冶金工业出版社, 2007.
[3]王冠,郑秀萍,张陶红,柴天佑.氧化铝生产过程的MES体系结构[J].东北大学自动化研究中心, 2003, 7:410-413.
[4]尹中林,顾松青.并行法氧化铝新工艺-我国混联法氧化铝生产工艺的发展方向[J].世界有色金属, 2001, 7:4-7.
[5]郑秀萍,王冠,柴天佑.氧化铝生产过程综合自动化体系结构[J].东北大学学报(自然科学版), 2003, 4:310-313.
[6]王寿和.面向新世纪的我国氧化铝生产过程自动化[J].铝镁通讯, 2003, 2:1-2.
[7]张红亮,陈湘涛,邹忠.我国氧化铝工业自动控制及系统的研究进展[J].冶金自动化, 2006, 8(03):59-62.
[8]林景栋.中国铝业股份有限公司遵义分公司生产调度系统技术方案[J].2008.
[9]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].中南大学,2004.
[10]么海青.氧化铝生产企业标准化管理系统研究[D].中南大学,2006.
[11] Trentesaux,D,Pesin,P.andTahon,C.Distributed artificial intelligence for FMS scheduling,control and design support.Journal of Intelligent Manufacturing[J].2000,11(6):573-589.
[12] M.A.S.Monfared and S.J.Steiner.Fuzzy adaptive scheduling and control systems.Fuzzy Sets and Systems[J].2000,115(2):231-246.
[13]魏海坤.神经网络结构设计的理论与方法[M].北京:国防工业出版社, 2005.
[14]王志,毕诗文等.碳酸化分解的机理研究与进展[J].轻金属, 2001, 12:12-14.
[15]任云翔.提高碳分氢氧化铝质量探讨与实践[J].世界有色金属, 2001, 2:16-19.
[16] Wang Zhi.Influence of mass concentration on carbonation decomposition in sodium aluninate solution [J].Journal of Northeastern University,2002,23(6): 560-562.
[17]周宗科,陈红武.烧结法氧化铝生料浆配料方法新研究[J].世界有色金属, 2004, 6(04):12-14.
[18]温丽敏.氧化铝生料浆调配过程的智能满意优化方法及应用[D].中南大学, 2008.
[19]范开俊.工业生产配料过程控制策略研究[D].浙江大学, 2008.
[20]张斯科.关于对氧化铝生产调度系统的探讨[J].太原科技, 2003, 4:56-58.
[21]芦东.浅谈氧化铝生产调度管理职能.世界有色金属[J]. 1999, 9:35-36.
[22]蒋凌燕,楼佩煌.基于B/S结构的生产调度系统实现技术研究[N].现代企业与先进制造技术高层论坛,2002, 3-6.
[23]柳树春,廖孟杨等.Browser/server模式管理信息系统设计与实现[J].计算机工程与应用, 2000, 36(6):57-58,63.
[24]王成光.流程工业大型实时数据库理论、技术与应用[D].浙江大学,2003.
[25]张志檩.实时数据库原理及应用[M].北京:中国石化出版社, 2001.
[26]张吉礼.模糊-神经网络控制原理与工程应用[M].哈尔滨:哈尔滨工业大学出版社, 2004.
[27]李勇刚,桂卫华,陈峰.基于因素分析的复合神经网络及其在软测量中的应用[J].信息与控制, 2001, 31(5):812-816.
[28]魏海坤,徐祠鑫,宋文忠.神经网络的泛化理论和泛化方法[J].自动化学报, 2001, 27(6):805-815.
[29]王雅林.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].中南大学,2001.
[30]袁曾任.人工神经网络及其应用[M].北京:清华大学出版社, 1999.
[31]应苏如.氧化铝生产过程中的苛性比值与溶出率智能集成预测模型的研究与应用[D].中南大学,2004.
[32]杨巧芳,赵清杰.我国氧化铝工业现状及世纪初发展战略探讨[J].世界有色金属, 2001, 6:24-26.
[33]王能斌.数据库系统原理[M].北京:电子工业出版社, 2002.
[34]陆钦芳.关于我国氧化铝工业竞争力和发展对策的探讨[J].轻金属,2001,5:3-5.
[35]白锐,柴天佑,马恩杰生料浆配料过程综合自动化系统[J].系统仿真学报,2007,17(11):2596~2598.
[36]彭怀文.氧化铝熟料烧结系统的工艺技术优化[D].中南大学,2006.
[37] Donald E.Shobrys and Douglas C. White.Planning,scheduling and control systems:why cannot they work together[J].Computer &Chemical Engineering.2002,26(2):148-160.
[38]王龙生.面向MES的作业车间调度与控制系统的研究与开发[D].中南大学,2005.
[39]温艳芬,李化民,贾九林等.氧化铝生产过程制造执行系统研究[J].控制工程,2004, 5(11):124-125.
[40]刘远生.计算机网络安全[M].北京:清华大学出版社, 2006.
[41]罗焕佐,宋国宁,王晓峰等.流程企业智能计划调度技术[M].沈阳:东北大学出版社, 2004.
[42] Yang Huizhong,Li Danjing,Tao Zhenlin,Zhang Suzhen.Variable-step BP training algorithm with an adaptive momentum term[J]. Intelligent Control and Automation. 2002, 3(10):196-200.
[43] Jim Y.f.Yam,Tommy W.S.Chow.A weight initialization method for improving training speed in feedforword neural network[J]. Neuralcomputing, 2000, 8:219-232.
[44] Paulito P.Palmes,Yaichi Hayasaka,Shiro Usui.Mutation-Based Genetic Neural network.[J]. IEEE Transactions on Neural Networks, 2005, 16(3):46-49.
[2]毕诗文,于海燕等.拜耳法生产氧化铝[M].北京::冶金工业出版社, 2007.
[3]王冠,郑秀萍,张陶红,柴天佑.氧化铝生产过程的MES体系结构[J].东北大学自动化研究中心, 2003, 7:410-413.
[4]尹中林,顾松青.并行法氧化铝新工艺-我国混联法氧化铝生产工艺的发展方向[J].世界有色金属, 2001, 7:4-7.
[5]郑秀萍,王冠,柴天佑.氧化铝生产过程综合自动化体系结构[J].东北大学学报(自然科学版), 2003, 4:310-313.
[6]王寿和.面向新世纪的我国氧化铝生产过程自动化[J].铝镁通讯, 2003, 2:1-2.
[7]张红亮,陈湘涛,邹忠.我国氧化铝工业自动控制及系统的研究进展[J].冶金自动化, 2006, 8(03):59-62.
[8]林景栋.中国铝业股份有限公司遵义分公司生产调度系统技术方案[J].2008.
[9]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].中南大学,2004.
[10]么海青.氧化铝生产企业标准化管理系统研究[D].中南大学,2006.
[11] Trentesaux,D,Pesin,P.andTahon,C.Distributed artificial intelligence for FMS scheduling,control and design support.Journal of Intelligent Manufacturing[J].2000,11(6):573-589.
[12] M.A.S.Monfared and S.J.Steiner.Fuzzy adaptive scheduling and control systems.Fuzzy Sets and Systems[J].2000,115(2):231-246.
[13]魏海坤.神经网络结构设计的理论与方法[M].北京:国防工业出版社, 2005.
[14]王志,毕诗文等.碳酸化分解的机理研究与进展[J].轻金属, 2001, 12:12-14.
[15]任云翔.提高碳分氢氧化铝质量探讨与实践[J].世界有色金属, 2001, 2:16-19.
[16] Wang Zhi.Influence of mass concentration on carbonation decomposition in sodium aluninate solution [J].Journal of Northeastern University,2002,23(6): 560-562.
[17]周宗科,陈红武.烧结法氧化铝生料浆配料方法新研究[J].世界有色金属, 2004, 6(04):12-14.
[18]温丽敏.氧化铝生料浆调配过程的智能满意优化方法及应用[D].中南大学, 2008.
[19]范开俊.工业生产配料过程控制策略研究[D].浙江大学, 2008.
[20]张斯科.关于对氧化铝生产调度系统的探讨[J].太原科技, 2003, 4:56-58.
[21]芦东.浅谈氧化铝生产调度管理职能.世界有色金属[J]. 1999, 9:35-36.
[22]蒋凌燕,楼佩煌.基于B/S结构的生产调度系统实现技术研究[N].现代企业与先进制造技术高层论坛,2002, 3-6.
[23]柳树春,廖孟杨等.Browser/server模式管理信息系统设计与实现[J].计算机工程与应用, 2000, 36(6):57-58,63.
[24]王成光.流程工业大型实时数据库理论、技术与应用[D].浙江大学,2003.
[25]张志檩.实时数据库原理及应用[M].北京:中国石化出版社, 2001.
[26]张吉礼.模糊-神经网络控制原理与工程应用[M].哈尔滨:哈尔滨工业大学出版社, 2004.
[27]李勇刚,桂卫华,陈峰.基于因素分析的复合神经网络及其在软测量中的应用[J].信息与控制, 2001, 31(5):812-816.
[28]魏海坤,徐祠鑫,宋文忠.神经网络的泛化理论和泛化方法[J].自动化学报, 2001, 27(6):805-815.
[29]王雅林.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].中南大学,2001.
[30]袁曾任.人工神经网络及其应用[M].北京:清华大学出版社, 1999.
[31]应苏如.氧化铝生产过程中的苛性比值与溶出率智能集成预测模型的研究与应用[D].中南大学,2004.
[32]杨巧芳,赵清杰.我国氧化铝工业现状及世纪初发展战略探讨[J].世界有色金属, 2001, 6:24-26.
[33]王能斌.数据库系统原理[M].北京:电子工业出版社, 2002.
[34]陆钦芳.关于我国氧化铝工业竞争力和发展对策的探讨[J].轻金属,2001,5:3-5.
[35]白锐,柴天佑,马恩杰生料浆配料过程综合自动化系统[J].系统仿真学报,2007,17(11):2596~2598.
[36]彭怀文.氧化铝熟料烧结系统的工艺技术优化[D].中南大学,2006.
[37] Donald E.Shobrys and Douglas C. White.Planning,scheduling and control systems:why cannot they work together[J].Computer &Chemical Engineering.2002,26(2):148-160.
[38]王龙生.面向MES的作业车间调度与控制系统的研究与开发[D].中南大学,2005.
[39]温艳芬,李化民,贾九林等.氧化铝生产过程制造执行系统研究[J].控制工程,2004, 5(11):124-125.
[40]刘远生.计算机网络安全[M].北京:清华大学出版社, 2006.
[41]罗焕佐,宋国宁,王晓峰等.流程企业智能计划调度技术[M].沈阳:东北大学出版社, 2004.
[42] Yang Huizhong,Li Danjing,Tao Zhenlin,Zhang Suzhen.Variable-step BP training algorithm with an adaptive momentum term[J]. Intelligent Control and Automation. 2002, 3(10):196-200.
[43] Jim Y.f.Yam,Tommy W.S.Chow.A weight initialization method for improving training speed in feedforword neural network[J]. Neuralcomputing, 2000, 8:219-232.
[44] Paulito P.Palmes,Yaichi Hayasaka,Shiro Usui.Mutation-Based Genetic Neural network.[J]. IEEE Transactions on Neural Networks, 2005, 16(3):46-49.