水泥厂质量管理系统及其优化管理平台的设计与实现
|
摘要
产品质量是企业的生命线，是企业发展壮大的源泉。产品质量，就其本质
而言，它是产品所具有的某种“能力”的属性，就是这种“能力”，才使得产
品能满足社会的各种各样的需要。产品的质量概念包含结果，也包含质量的形
成和实现过程的工作质量。要用过程的工作质量来保证产品的特征和特性的实
现，就需要在生产中狠抓质量环节中各个工序的质量。而保证过程工作质量和
产品质量的关键在于优质质量管理和控制方针的正确、有效及切实落实和施
行。所以，质量管理和控制是一项系统工程。
本文以水泥厂的质量管理系统的研发项目为应用背景，构建了基于
MIDAS的多层分布为主与基于Web的B／S为辅的混合架构的体系结构；同
时，在此基础上，设计了一个以多种优化算法为核心，集成生产过程中的质量
信息，利用一些比较先进的数学算法、数据库技术、网络技术搭建一个工艺优
化平台，并且将该优化平台嵌入到生产过程中去。不仅能优化水泥厂的生产工
艺，设计出生产过程中重要的生产工艺的最优方案，还可以将其设计的结果，
直接送入下位机作为整个生产过程的控制目标，当生产条件发生变化时，能实
时进行现场调控，保证优化工艺设计的目标要求能在生产过程中得到稳定的实
四川大学硕士学位论文
现;而且在原燃料、能源的价格控制、出厂水泥的成本控制(矿渣掺入量)方
面也能提供优化方案，在保证质量合格的前提下，尽可能降低企业生产成本。
论文介绍了“水泥厂质量管理系统及其优化管理平台”的系统分析、设计
原理、组成及实现方法，重点阐述了正交试验设计与分析、线性规划、逐步多
元回归软件的设计思想和实现技术，还给出了三个专用优化系统的实例，对其
优化结果进行了讨论。
The product quality is the lifeline and development headspring of enterprise. As far as essence be concerned, it is certain attribution of ability what the product possesses , and just the ability makes the product fill a wide variety of needs of society. The concept of product quality includes the result, quality formation and work quality of realization process. To ensure the realization of characteristic and property of product by controlling the work quality of process, to control hard all kinds of working procedure quality is needed in the whole quality link. Moreover, to ensure good work quality of process and product quality, the key lies in to effectively and rightly put the project of optimization quality control and management into effect. Therefore, the quality management and control is a system project.
In this paper, a mixed distributed multilayer system architecture which base on dominating MIDAS(C/S) and subordinate Web
(B/S) techniques is built by the application background of research and the development project of the quality control system of cement factory . At the same time, a optimization technology platform which will be embedded in the production process is put up with several advanced optimization mathematics algorithms integrated quality information in production process, database technique and net technique. It not only optimizes the production technology of cement factory and designs the preferred plan of main production technology in the industrial production process, but also sends the result of design to the sub-dominative computer as the control target of the whole production process in order to adjust some online control information of fields and ensure the design demand of optimization technology to be stably realized in the production process while the production condition is changed. Moreover, it can provides optimization plans to control the price of the raw material, energy and cement and reduces cost
.
In this paper, the analysis, design and implement method of system are introduced firstly. Then the design method and implement technique of the orthogonal experiment plan and analysis, linear programming and multiple regression are expounded mainly. At last, three instances of the optimization custom-built system are given and then the optimization result of instances are discussed.
而言，它是产品所具有的某种“能力”的属性，就是这种“能力”，才使得产
品能满足社会的各种各样的需要。产品的质量概念包含结果，也包含质量的形
成和实现过程的工作质量。要用过程的工作质量来保证产品的特征和特性的实
现，就需要在生产中狠抓质量环节中各个工序的质量。而保证过程工作质量和
产品质量的关键在于优质质量管理和控制方针的正确、有效及切实落实和施
行。所以，质量管理和控制是一项系统工程。
本文以水泥厂的质量管理系统的研发项目为应用背景，构建了基于
MIDAS的多层分布为主与基于Web的B／S为辅的混合架构的体系结构；同
时，在此基础上，设计了一个以多种优化算法为核心，集成生产过程中的质量
信息，利用一些比较先进的数学算法、数据库技术、网络技术搭建一个工艺优
化平台，并且将该优化平台嵌入到生产过程中去。不仅能优化水泥厂的生产工
艺，设计出生产过程中重要的生产工艺的最优方案，还可以将其设计的结果，
直接送入下位机作为整个生产过程的控制目标，当生产条件发生变化时，能实
时进行现场调控，保证优化工艺设计的目标要求能在生产过程中得到稳定的实
四川大学硕士学位论文
现;而且在原燃料、能源的价格控制、出厂水泥的成本控制(矿渣掺入量)方
面也能提供优化方案，在保证质量合格的前提下，尽可能降低企业生产成本。
论文介绍了“水泥厂质量管理系统及其优化管理平台”的系统分析、设计
原理、组成及实现方法，重点阐述了正交试验设计与分析、线性规划、逐步多
元回归软件的设计思想和实现技术，还给出了三个专用优化系统的实例，对其
优化结果进行了讨论。
The product quality is the lifeline and development headspring of enterprise. As far as essence be concerned, it is certain attribution of ability what the product possesses , and just the ability makes the product fill a wide variety of needs of society. The concept of product quality includes the result, quality formation and work quality of realization process. To ensure the realization of characteristic and property of product by controlling the work quality of process, to control hard all kinds of working procedure quality is needed in the whole quality link. Moreover, to ensure good work quality of process and product quality, the key lies in to effectively and rightly put the project of optimization quality control and management into effect. Therefore, the quality management and control is a system project.
In this paper, a mixed distributed multilayer system architecture which base on dominating MIDAS(C/S) and subordinate Web
(B/S) techniques is built by the application background of research and the development project of the quality control system of cement factory . At the same time, a optimization technology platform which will be embedded in the production process is put up with several advanced optimization mathematics algorithms integrated quality information in production process, database technique and net technique. It not only optimizes the production technology of cement factory and designs the preferred plan of main production technology in the industrial production process, but also sends the result of design to the sub-dominative computer as the control target of the whole production process in order to adjust some online control information of fields and ensure the design demand of optimization technology to be stably realized in the production process while the production condition is changed. Moreover, it can provides optimization plans to control the price of the raw material, energy and cement and reduces cost
.
In this paper, the analysis, design and implement method of system are introduced firstly. Then the design method and implement technique of the orthogonal experiment plan and analysis, linear programming and multiple regression are expounded mainly. At last, three instances of the optimization custom-built system are given and then the optimization result of instances are discussed.
引文
[1] Alexei Bespalov. Orthogonal systems of singular functions and numerical treatment of problems with degeneration of data[J]. Advances in Computational Mathematics, 2003, 19: 159-182.
[2] Xiong M, Wang J, Wang P. Differential-Algebraic Approach to linear programming[J]. Journal of Optimization Theory and Applications, 2002, 114(2): 443-470.
[3] J Shen, R. Li, F Zhang, Z Rengel, C Tang. Orthogonal polynomial models to describe yield response of rice to nitrogen and phosphorus at different levels of soil fertility[J]. Nutrient Cycling in Agroecosystems, 2003, 65: 243-251.
[4] Chang T, Sadachi J, Wang X, Chen R.. On the Equivalence of the Simplex Methods and a Multiplier-Alike Method for linear programming[J]. Journal of Optimization Theory and Applications, 2002, 113(3): 487-512.
[5] J Tobin. A Statistical Demand Function for food in the U.S.A[J]. Journal of Royal Statistical society, Ser. A: 113-141.
[6] Liu C, Green S M, Watkins ND, Gregg P J, McCaskie A W. Effect of restraint on the creep behavior of clinical bone cement[J]. Journal of Materials Science Letters, 2002, 21(12): 959-962.
[7] Mac Nally, Ralph. Multiple regression and inference in ecology and conservation biology: further comments on identifying important predictor variables[J]. Biodiversity and Conservation, 2002, 11(8): 1397-1401.
[8] Duffy, Nigel, Helmbold, David. Boosting Methods for regression[J]. Machine Learning, 2002, 47(2/3): 153-200.
[9] Romas&apos, ko V S, Chernyavs&apos, kyi V L.. On "Unpredictable" Structure-Changing Effects in cement Materials[J]. Materials Science, 2002, 38(3): 413-417.
[10] Granichin O N. Estimating the Parameters of Linear regression in an Arbitrary Noise[J]. Automation and Remote Control, 2002, 63(1): 25-35.
[11] Perraki M, Perraki T, Kolovos K., Tsivilis S, Kakal G. Secondary raw materials in cement industry[J]. Journal of Thermal Analysis and Calorimetry, 2002, 70(1): 143-150.
[12] Kivinen J, Warmuth M K. Relative Loss Bounds for Multidimensional regression Problems[J]. Machine Learning, 2001, 45(3): 301-329.
[13] Evdokimova N G, Murtazin F R., Tselishchev Ⅳ. Regression Analysis of Operation of a Modified Paving Asphalt Production Plant[J]. Chemistry and Technology of Fuels and Oils, 2001, 37(3): 163-166.
[14] Shubin V&period, I&period. The Effect of Temperature on the Lining of Rotary cement Kilns[J]. Refractories and Industrial Ceramics, 2001, 42(5/6): 216-221.
[15] Tyurin A N. The Development of a regression Model for Estimating the Stability of a System for Transferring the Dimensions of the Units of Physical Quantities[J]. Measurement Techniques, 2001, 44(2): 174-180.
[16] Walczuch Rita M, Watson, Richard T. Analyzing Group Data in MIS Research: Including the Effect of the Group[J]. Group Decision and Negotiation, 2001, 10(1): 83-94.
[17] Zlobec S. Stability in linear programming Models: An Index Set Approach[J]. Annals of Operations Research, 2001, 101(1/4): 363-382.
[18] Mac Nally, Ralph. Regression and model-building in conservation biology, biogeography and ecology[J]. Biodiversity and Conservation, 2000, 9(5): 655-671.
[19] Guo Shenghui, Sun Yufang. Information system design based on the dictionary database[J]. JISUANJI XUEBAO, 2000, 23(4): 414-418.
[20] Bohner M, Merkle H P, Landuyt P Van, Trophardy G, Lemaitre J. Effect of several additives and their admixtures on the physico-chemical properties of a calcium phosphate cement[J]. Journal of Materials Science: Materials in Medicine, 2000, 11(2): 111-116.
[21] FIESTRAS-JANEIRO M G, GARCIA-JURADO I, PUERTO J. The Concept of Proper Solution in linear programming[J]. Journal of Optimization Theory and Applications, 2000, 106(3): 511-525.
[22] Glazebrook K. D, Ni(?)o&dash, Mora J. A linear programming approach to stability, optimisationand performance analysis for Markovian multiclassqueueing networks[J]. Annals of Operations Research, 2000, 92(1): 1-18.
[23] A W Brown(Ed.). Component-Based Software Engineering[R]. Newyork IEEE Computer Soc, 1997, 12-23.
[24] http://www.opcconnect.com/delphi.php
[25] H NIEMANN, D SCHMIDT, U MAAS. An efficient storage scheme for reduced chemical kinetics based on orthogonal polynomials[J]. Journal of Engineering Mathematics, 1997, 31: 131-142.
[26] Steve Teixeira, Xavier Pacheco. Delphi 5开发人员指南[M]．机械工业出版社，2000．
[27] 李维．Delphi 5.x分布式多层应用系统篇[M]．机械工业出版社，2000．
[28] 李维．Delphi 5.x ADO/MTS/COM+高级程序设计篇[M]．机械工业出版社，2000．
[29] 刘艺．Delphi 6企业级解决方案及应用[M]．机械工业出版社，2002．
[30] Ivar Jacobson,Grady Booch,James Rumbaugh.统一软件开发过程[M]．机械工业出版社，2002．
[31] 陈京民．数据仓库与数据挖掘技术[M]．电子工业出版社，2002．
[32] 刘龙．水泥工厂实用新技术[M]．武汉工业大学出版社，2000．
[33] 李明豫，丁卫东．地方水泥企业化验室工作手册[M]．中国建材工业出版社，1994．
[34] Hichael Otey, Paul Conte.SQL Server 2000开发指南[M]．清华大学出版社，2002．
[35] 周德泽，袁南儿，应英．计算机智能监测控制系统的设计及应用[M]．清华大学出版社，2002．
[36] 陈魁．试验设计与分析[M]．清华大学出版社，1996．
[37] 甘应爱，田丰等．运筹学[M]．清华大学出版社，2002．
[38] Don Box.COM本质论[M]．中国电力出版社，2001．
[39] Jiawei Han,Micheline Kamber．数据挖掘与技术[M]．机械工业出版社，2001．
[40] 陈松乔，任胜兵，王国军．现代软件工程[M]．北方交通大学出版社，2002．
[41] 夏翔，范轶翔，吴庆星．水泥厂设备管理多层分布系统及优化设计的实现[J]．四川大学学报(自然科学版)，2003，40(2)：251～255．
[42] 张华雨，韩伯棠，覃云．Intranet环境下基于数据仓库技术的新型DSS[J]．计算机应用，2002，22(2)．
[43] 陈希孺．数理统计引论[M]．科学出版社，1999．
[44] 张捍东，许宝栋等．数据挖掘技术在管理与决策中的应用[J]．华中科技大学学报(自然科学版)，2002，30(6)：73～75．
[45] 高方君，刘小冬．线性规划问题的一种新的摄动法．运筹与管理[J]，2002，56～59．
[46] 沈文杰，葛运旺．利用PLC组成的水泥厂烧成车间控制系统DCS[J]．洛阳工业高等专科学校学报，2001，6～8．
[47] 吴凤玲．C／S和B／S模式下的教务管理系统的开发[J]．武汉理工大学学报(信息与管理工程版)，2003，25(1)．
[48] 张丽、张艳．从C／S到B／S，再到三层(多层)结构——论体系结构的发展[J]．河南师范大学学报(自然科学版)，2002，30(3)．
[49] 黄宁，张俊，范轶翔．成份配料算法的研究与实现[J]．四川大学学报(自然科学版)，2002，39(2)．
[50] 刘晓莉．试验设计中多元回归分析方法的研究[J]．数理统计与管理，2001，4：14～16．
[51] 冉林仓，王建华．基于COM组件的不等式约束线性规划求解[J]．北京机械工业学院学报，2002，6：6～9．
[52] 吕剑亮，朱坤．运筹学线性规划模型求解的计算机应用[J]．长春工程学院学报(自然科学版)，2001，2(3)：59～62．
[53] 吴庆星，叶思敏．基于分布式多层结构的水泥厂MIS系统[J]．四川大学学报(工程科学版)，2002，34(6)：105～108．
[54] 胡纫兰，等．湿法水泥生料浆自动配库控制系统的研制[J]．四川大学学报(自然科学版)，1998，35(3)：358．
[55] 胡纫兰，等．支持工艺优化决策的水泥厂质量信息管理系统的设计和实现[J]．四川大学学报(自然科学版)，1998，35(2)：183-186．
[56] 王亚杰．基于C／S和B／S的信息系统开发与研究[J]．沈阳大学学报，2002，14(2)．
[57] 张红云，马恒．数据挖掘与决策支持系统的关系[J]．鞍山师范学院学报，2001，8(3)：86～89．
[58] 周汝宗，王翔．查询与发现：数据挖掘技术分析[J]．计算机工程，2002，28(11)：281～283．
[59] 李延，黄宁，范轶翔．基于Web 3层客户／服务器(B/S)模式的信息网与设备网无缝互连的研究[J]．四川大学学报(自然科学版)，2002，39(1)：69-72．
[60] 查卫翔，谭南林．ActiveX控件在基于B／S结构的远层监控中的应用[J]．北方交通大学学报，2002，26(1)．
[61] 李占贤，黄金凤．利用多元回归分析方法控制球磨机负荷稳定性[J]．矿山机械，2002，37～38．
[2] Xiong M, Wang J, Wang P. Differential-Algebraic Approach to linear programming[J]. Journal of Optimization Theory and Applications, 2002, 114(2): 443-470.
[3] J Shen, R. Li, F Zhang, Z Rengel, C Tang. Orthogonal polynomial models to describe yield response of rice to nitrogen and phosphorus at different levels of soil fertility[J]. Nutrient Cycling in Agroecosystems, 2003, 65: 243-251.
[4] Chang T, Sadachi J, Wang X, Chen R.. On the Equivalence of the Simplex Methods and a Multiplier-Alike Method for linear programming[J]. Journal of Optimization Theory and Applications, 2002, 113(3): 487-512.
[5] J Tobin. A Statistical Demand Function for food in the U.S.A[J]. Journal of Royal Statistical society, Ser. A: 113-141.
[6] Liu C, Green S M, Watkins ND, Gregg P J, McCaskie A W. Effect of restraint on the creep behavior of clinical bone cement[J]. Journal of Materials Science Letters, 2002, 21(12): 959-962.
[7] Mac Nally, Ralph. Multiple regression and inference in ecology and conservation biology: further comments on identifying important predictor variables[J]. Biodiversity and Conservation, 2002, 11(8): 1397-1401.
[8] Duffy, Nigel, Helmbold, David. Boosting Methods for regression[J]. Machine Learning, 2002, 47(2/3): 153-200.
[9] Romas&apos, ko V S, Chernyavs&apos, kyi V L.. On "Unpredictable" Structure-Changing Effects in cement Materials[J]. Materials Science, 2002, 38(3): 413-417.
[10] Granichin O N. Estimating the Parameters of Linear regression in an Arbitrary Noise[J]. Automation and Remote Control, 2002, 63(1): 25-35.
[11] Perraki M, Perraki T, Kolovos K., Tsivilis S, Kakal G. Secondary raw materials in cement industry[J]. Journal of Thermal Analysis and Calorimetry, 2002, 70(1): 143-150.
[12] Kivinen J, Warmuth M K. Relative Loss Bounds for Multidimensional regression Problems[J]. Machine Learning, 2001, 45(3): 301-329.
[13] Evdokimova N G, Murtazin F R., Tselishchev Ⅳ. Regression Analysis of Operation of a Modified Paving Asphalt Production Plant[J]. Chemistry and Technology of Fuels and Oils, 2001, 37(3): 163-166.
[14] Shubin V&period, I&period. The Effect of Temperature on the Lining of Rotary cement Kilns[J]. Refractories and Industrial Ceramics, 2001, 42(5/6): 216-221.
[15] Tyurin A N. The Development of a regression Model for Estimating the Stability of a System for Transferring the Dimensions of the Units of Physical Quantities[J]. Measurement Techniques, 2001, 44(2): 174-180.
[16] Walczuch Rita M, Watson, Richard T. Analyzing Group Data in MIS Research: Including the Effect of the Group[J]. Group Decision and Negotiation, 2001, 10(1): 83-94.
[17] Zlobec S. Stability in linear programming Models: An Index Set Approach[J]. Annals of Operations Research, 2001, 101(1/4): 363-382.
[18] Mac Nally, Ralph. Regression and model-building in conservation biology, biogeography and ecology[J]. Biodiversity and Conservation, 2000, 9(5): 655-671.
[19] Guo Shenghui, Sun Yufang. Information system design based on the dictionary database[J]. JISUANJI XUEBAO, 2000, 23(4): 414-418.
[20] Bohner M, Merkle H P, Landuyt P Van, Trophardy G, Lemaitre J. Effect of several additives and their admixtures on the physico-chemical properties of a calcium phosphate cement[J]. Journal of Materials Science: Materials in Medicine, 2000, 11(2): 111-116.
[21] FIESTRAS-JANEIRO M G, GARCIA-JURADO I, PUERTO J. The Concept of Proper Solution in linear programming[J]. Journal of Optimization Theory and Applications, 2000, 106(3): 511-525.
[22] Glazebrook K. D, Ni(?)o&dash, Mora J. A linear programming approach to stability, optimisationand performance analysis for Markovian multiclassqueueing networks[J]. Annals of Operations Research, 2000, 92(1): 1-18.
[23] A W Brown(Ed.). Component-Based Software Engineering[R]. Newyork IEEE Computer Soc, 1997, 12-23.
[24] http://www.opcconnect.com/delphi.php
[25] H NIEMANN, D SCHMIDT, U MAAS. An efficient storage scheme for reduced chemical kinetics based on orthogonal polynomials[J]. Journal of Engineering Mathematics, 1997, 31: 131-142.
[26] Steve Teixeira, Xavier Pacheco. Delphi 5开发人员指南[M]．机械工业出版社，2000．
[27] 李维．Delphi 5.x分布式多层应用系统篇[M]．机械工业出版社，2000．
[28] 李维．Delphi 5.x ADO/MTS/COM+高级程序设计篇[M]．机械工业出版社，2000．
[29] 刘艺．Delphi 6企业级解决方案及应用[M]．机械工业出版社，2002．
[30] Ivar Jacobson,Grady Booch,James Rumbaugh.统一软件开发过程[M]．机械工业出版社，2002．
[31] 陈京民．数据仓库与数据挖掘技术[M]．电子工业出版社，2002．
[32] 刘龙．水泥工厂实用新技术[M]．武汉工业大学出版社，2000．
[33] 李明豫，丁卫东．地方水泥企业化验室工作手册[M]．中国建材工业出版社，1994．
[34] Hichael Otey, Paul Conte.SQL Server 2000开发指南[M]．清华大学出版社，2002．
[35] 周德泽，袁南儿，应英．计算机智能监测控制系统的设计及应用[M]．清华大学出版社，2002．
[36] 陈魁．试验设计与分析[M]．清华大学出版社，1996．
[37] 甘应爱，田丰等．运筹学[M]．清华大学出版社，2002．
[38] Don Box.COM本质论[M]．中国电力出版社，2001．
[39] Jiawei Han,Micheline Kamber．数据挖掘与技术[M]．机械工业出版社，2001．
[40] 陈松乔，任胜兵，王国军．现代软件工程[M]．北方交通大学出版社，2002．
[41] 夏翔，范轶翔，吴庆星．水泥厂设备管理多层分布系统及优化设计的实现[J]．四川大学学报(自然科学版)，2003，40(2)：251～255．
[42] 张华雨，韩伯棠，覃云．Intranet环境下基于数据仓库技术的新型DSS[J]．计算机应用，2002，22(2)．
[43] 陈希孺．数理统计引论[M]．科学出版社，1999．
[44] 张捍东，许宝栋等．数据挖掘技术在管理与决策中的应用[J]．华中科技大学学报(自然科学版)，2002，30(6)：73～75．
[45] 高方君，刘小冬．线性规划问题的一种新的摄动法．运筹与管理[J]，2002，56～59．
[46] 沈文杰，葛运旺．利用PLC组成的水泥厂烧成车间控制系统DCS[J]．洛阳工业高等专科学校学报，2001，6～8．
[47] 吴凤玲．C／S和B／S模式下的教务管理系统的开发[J]．武汉理工大学学报(信息与管理工程版)，2003，25(1)．
[48] 张丽、张艳．从C／S到B／S，再到三层(多层)结构——论体系结构的发展[J]．河南师范大学学报(自然科学版)，2002，30(3)．
[49] 黄宁，张俊，范轶翔．成份配料算法的研究与实现[J]．四川大学学报(自然科学版)，2002，39(2)．
[50] 刘晓莉．试验设计中多元回归分析方法的研究[J]．数理统计与管理，2001，4：14～16．
[51] 冉林仓，王建华．基于COM组件的不等式约束线性规划求解[J]．北京机械工业学院学报，2002，6：6～9．
[52] 吕剑亮，朱坤．运筹学线性规划模型求解的计算机应用[J]．长春工程学院学报(自然科学版)，2001，2(3)：59～62．
[53] 吴庆星，叶思敏．基于分布式多层结构的水泥厂MIS系统[J]．四川大学学报(工程科学版)，2002，34(6)：105～108．
[54] 胡纫兰，等．湿法水泥生料浆自动配库控制系统的研制[J]．四川大学学报(自然科学版)，1998，35(3)：358．
[55] 胡纫兰，等．支持工艺优化决策的水泥厂质量信息管理系统的设计和实现[J]．四川大学学报(自然科学版)，1998，35(2)：183-186．
[56] 王亚杰．基于C／S和B／S的信息系统开发与研究[J]．沈阳大学学报，2002，14(2)．
[57] 张红云，马恒．数据挖掘与决策支持系统的关系[J]．鞍山师范学院学报，2001，8(3)：86～89．
[58] 周汝宗，王翔．查询与发现：数据挖掘技术分析[J]．计算机工程，2002，28(11)：281～283．
[59] 李延，黄宁，范轶翔．基于Web 3层客户／服务器(B/S)模式的信息网与设备网无缝互连的研究[J]．四川大学学报(自然科学版)，2002，39(1)：69-72．
[60] 查卫翔，谭南林．ActiveX控件在基于B／S结构的远层监控中的应用[J]．北方交通大学学报，2002，26(1)．
[61] 李占贤，黄金凤．利用多元回归分析方法控制球磨机负荷稳定性[J]．矿山机械，2002，37～38．