基于产品质量基因的水泥装备制造过程质量诊断方法研究
|
摘要
随着科学技术的进步和经济全球化的发展,市场竞争日益激烈,优质的产品质量是企业提高核心竞争力的关键因素。面对水泥装备制造业的发展机遇,企业必须提高制造过程中的质量诊断方法,提供最优质的产品,才能获得并保持有利的市场地位。现有的一些质量诊断的方法和模式,如统计分析建模,专家诊断系统、模糊诊断系统、人工神经网络诊断系统等分别从不同的侧面致力于提高质量诊断方法的正确性和方便性。
水泥装备制造行业质量诊断分析是综合了产品工艺设计、加工和专家经验等的一个复杂过程,需要考虑产品质量信息的获取和提炼,分析质量信息间的关联性等。基因工程技术目前在设计、制造领域有了广泛的应用,本课题将其应用进一步扩展,应用在了产品制造过程的质量诊断上,提出基于产品质量基因的质量诊断模式：通过设计、存储、提取产品质量基因编码,分析产品质量基因的遗传、变异过程,优化质量诊断流程来达到质量诊断的目的。本文从以下几个方面对水泥装备产品加工过程中的质量诊断方法进行研究：
(1)分析了产品质量基因的产生,提出了产品质量基因模型,确定了产品质量基因组成,结合水泥装备加工企业质量信息的复杂性,提出了产品质量基因(PQG-Code)编码方式和产品质量基因XML存储模式(PQG-XML),并实例分析了产品质量基因编码方法。
(2)分析了水泥装备产品质量基因的形成与进化、水泥装备产品加工的检测过程。根据其质量基因进化与检测的特点,深入剖析了产品质量基因的各种遗传与变异,并实例进行了分析；针对产品遗传与变异特性,提出了水泥装备产品的质量基因诊断模型。
(3)根据质量基因诊断模型,提出了质量基因相似度定义,并分析了基于质量基因相似度的质量诊断流程,即通过比较待检测产品质量基因与产品质量基因诊断知识库中的案例基因的相似度与给定阀值的大小,来判断产品在加工工序中,质量的合格性,并获取产品加工过程中的质量诊断结果。
(4)研究了基于产品质量基因的质量诊断系统的需求分析及总体结构,并使用ASP.NET (C#)和SQL Server2005,开发出质量诊断的各模块,最后在唐山某水泥装备制造企业进行了实施和应用,取得良好的效果。
最后对全文的研究工作进行了总结,分析存在的不足,并对课题的后续研究工作进行了展望。
With the rapid development of scientific technology and economic globalization, increasingly fierce market competition, the high-quality products is the key factors for enterprise to improve the core competitiveness. Facing the development opportunity of cement equipment manufacturing industry, improving the quality diagnosis methods in manufacturing process and providing the best quality products is the best way for enterprises to obtain and keep a favorable market position. Some existing quality diagnosis methods and modes, such as statistical analysis modeling, expert diagnosis system, fuzzy diagnosis system, and artificial neural network diagnosis system and so on are used to improve the exactness and convenience of quality diagnosis methods from different sides, respectively.
Cement equipment manufacturing industry quality diagnosis analysis is a complicated process with comprehensive factors of product design, process and expertise and so on, which needs to consider the acquisition and refining of product quality information and analyse the relevance between the quality information. At present, genetic engineering technology has been widely used in design and manufacturing fields, this study further expands its application, and it is used in the quality diagnosis of manufacturing process. The product quality gene quality diagnosis model is putting forward to achieve the purpose of quality diagnosis through designing, storing, collecting product quality gene code, analysing heredity and variation process of quality gene, optimizing the quality diagnosis process. This study analyzes the quality diagnosis methods in cement equipment product processing from the following aspects:
(1) Product quality genetic model is putted forward through analyzing the origin of product quality gene and the composition of product quality gene.The product quality gene (PQG-Code) coding method and product quality gene XML storage modes is putted forward combined with the complexity of cement equipment processing enterprise quality information.The product quality gene coding method is analysed in one enterprise.
(2) Analyzing the formation and evolution of cement equipment product quality gene and the testing process of cement equipment product. Dissecting various kinds of heredity and variation of product quality gene and analyzing the instance; Cement equipment product quality gene diagnostic model is putted forward by the characteristics of product heredity and variation.
(3) According to the quality gene diagnostic model, the definition of quality gene similarity is putted forward, and also quality diagnosis process based on quality gene similarity is analysed, that is whether the quality is normal or not can be judged by the given threshold and the similarity between the detected product quality gene and case gene of product quality genetic diagnosis knowledge base.
(4) The Demand analysis and general structure of quality gene diagnosis system based on the product quality gene is studied, and each module of quality diagnosis is developed by asp.net (C#) and SQL Server2005, and finally it is used in a cement equipment processing enterprise in Tangshan and achieves good effect.
Finally, the text summarizes the research work, analyze the deficiencies, and prospect follow-up study subjects.
水泥装备制造行业质量诊断分析是综合了产品工艺设计、加工和专家经验等的一个复杂过程,需要考虑产品质量信息的获取和提炼,分析质量信息间的关联性等。基因工程技术目前在设计、制造领域有了广泛的应用,本课题将其应用进一步扩展,应用在了产品制造过程的质量诊断上,提出基于产品质量基因的质量诊断模式：通过设计、存储、提取产品质量基因编码,分析产品质量基因的遗传、变异过程,优化质量诊断流程来达到质量诊断的目的。本文从以下几个方面对水泥装备产品加工过程中的质量诊断方法进行研究：
(1)分析了产品质量基因的产生,提出了产品质量基因模型,确定了产品质量基因组成,结合水泥装备加工企业质量信息的复杂性,提出了产品质量基因(PQG-Code)编码方式和产品质量基因XML存储模式(PQG-XML),并实例分析了产品质量基因编码方法。
(2)分析了水泥装备产品质量基因的形成与进化、水泥装备产品加工的检测过程。根据其质量基因进化与检测的特点,深入剖析了产品质量基因的各种遗传与变异,并实例进行了分析；针对产品遗传与变异特性,提出了水泥装备产品的质量基因诊断模型。
(3)根据质量基因诊断模型,提出了质量基因相似度定义,并分析了基于质量基因相似度的质量诊断流程,即通过比较待检测产品质量基因与产品质量基因诊断知识库中的案例基因的相似度与给定阀值的大小,来判断产品在加工工序中,质量的合格性,并获取产品加工过程中的质量诊断结果。
(4)研究了基于产品质量基因的质量诊断系统的需求分析及总体结构,并使用ASP.NET (C#)和SQL Server2005,开发出质量诊断的各模块,最后在唐山某水泥装备制造企业进行了实施和应用,取得良好的效果。
最后对全文的研究工作进行了总结,分析存在的不足,并对课题的后续研究工作进行了展望。
With the rapid development of scientific technology and economic globalization, increasingly fierce market competition, the high-quality products is the key factors for enterprise to improve the core competitiveness. Facing the development opportunity of cement equipment manufacturing industry, improving the quality diagnosis methods in manufacturing process and providing the best quality products is the best way for enterprises to obtain and keep a favorable market position. Some existing quality diagnosis methods and modes, such as statistical analysis modeling, expert diagnosis system, fuzzy diagnosis system, and artificial neural network diagnosis system and so on are used to improve the exactness and convenience of quality diagnosis methods from different sides, respectively.
Cement equipment manufacturing industry quality diagnosis analysis is a complicated process with comprehensive factors of product design, process and expertise and so on, which needs to consider the acquisition and refining of product quality information and analyse the relevance between the quality information. At present, genetic engineering technology has been widely used in design and manufacturing fields, this study further expands its application, and it is used in the quality diagnosis of manufacturing process. The product quality gene quality diagnosis model is putting forward to achieve the purpose of quality diagnosis through designing, storing, collecting product quality gene code, analysing heredity and variation process of quality gene, optimizing the quality diagnosis process. This study analyzes the quality diagnosis methods in cement equipment product processing from the following aspects:
(1) Product quality genetic model is putted forward through analyzing the origin of product quality gene and the composition of product quality gene.The product quality gene (PQG-Code) coding method and product quality gene XML storage modes is putted forward combined with the complexity of cement equipment processing enterprise quality information.The product quality gene coding method is analysed in one enterprise.
(2) Analyzing the formation and evolution of cement equipment product quality gene and the testing process of cement equipment product. Dissecting various kinds of heredity and variation of product quality gene and analyzing the instance; Cement equipment product quality gene diagnostic model is putted forward by the characteristics of product heredity and variation.
(3) According to the quality gene diagnostic model, the definition of quality gene similarity is putted forward, and also quality diagnosis process based on quality gene similarity is analysed, that is whether the quality is normal or not can be judged by the given threshold and the similarity between the detected product quality gene and case gene of product quality genetic diagnosis knowledge base.
(4) The Demand analysis and general structure of quality gene diagnosis system based on the product quality gene is studied, and each module of quality diagnosis is developed by asp.net (C#) and SQL Server2005, and finally it is used in a cement equipment processing enterprise in Tangshan and achieves good effect.
Finally, the text summarizes the research work, analyze the deficiencies, and prospect follow-up study subjects.
引文
[1]苗瑞,王东鹏,姚英学.CIMS环境下大型单件小批生产的质量管理信息系统研究[J].计算机工程与应用,2002,(8)：28-30.
[2]李国庆.建材企业提高产品质量的十大举措[J].云南建材,2000(3)：12-13.
[3]程元军.项目质量管理[M].北京：机械工业出版社,2007.2.
[4]张根保,曾海峰,王国强等.基于广义信息熵测度的制造过程质量评估[J].中国机械工程,2010,21(20)：2451-2458.
[5]CHIN K S, DUAN G, TANG X. A computer-integrated framework for global quality chain management [J]. The International Journal of Advanced Manufacturing Technology,2006,27(2): 547-560.
[6]刘娜,关玲永.全面质量管理的理论分析[J].北京建筑工程学院学报,2008,2：71-73.
[7]李益兵.光电子产品生命周期质量管理关键技术研究[D].武汉理工大学博士学位论文,2009.
[8]李延来,唐加福,姚建明,等.质量功能展开中选择工程特性的多目标决策方法[J].计算机集成制造系统,2008,14(7)：1363-1369.
[9]李延来,唐加福,姚建明等.质量屋构建的研究进展[J].机械工程学报,2009,45(3)：57-70.
[10]徐显龙,同淑荣,张新卫.基于产品结构树的制造质量信息分类研究[J].制造业自动化,2006,128(12)：37-41.
[11]Gao Huanming, Guo Mingzhe, Zhang Yu, etal. Research on process knowledge management methods based on ontology[J]. Mechanical Engineer,2009(3):68-71.
[12]Yu Honglei, Wang Xiulun, Yang Zhigang, et al. Application of XML to process knowledge management system[J]. Machinery Manufacturing Engineer,2005(6):22-24.
[13]Xu X L,Tong S R,Li B. A conceptual model of manufacturing quality information based on product structure tree[C].The 3rd IEEE International Conference on Wireless Communications, Networking and Mobile Computing (WICOM), Shanghai, China,2007,5129-5132.
[14]李晖,梁工谦,张新卫.一种面向BOM的制造质量信息模型[J].机床与液压,2008,36(1)：608-610.
[15]Nan Xie, Lin Chen and Aiping Li. Fault diagnosis of multistage manufacturing systems based on rough set approach [J]. The International Journal of Advanced Manufacturing Technology,2010,48:1239-1247.
[16]张赤斌,史金飞,易红.基于偏最小二乘回归的工序质量建模[J].东南大学学报,2005,35(5)：702-705.
[17]陈冰,李山,吴宝海,王明微.基于尺寸关联和偏最小二乘回归的多工序质量分析与预测[J].计算机集成制造系统.2009,15(2),389-390.
[18]刘道玉,江平宇.面向多工序制造过程的e-质量控制体系结构研究[J].计算机集成制造系统,2004,13(4)：782-790.
[19]黄诒森,张光毅.生物化学与分子生物学[M].北京：科学出版社,2003.
[20]贺淹才.简明基因工程原理[M].北京：科学出版社,2005.
[21]冯培恩,陈泳,张帅.基于产品基因的概念设计[J].机械工程学报,2002,38(10)：1-6.
[22]董晔弘,向东,龙旦风,刘畅,段广洪.基于贝叶斯网的制造工艺质量建模方法研究.2010全国现代制造集成技术学术会议论文集,2010,398-498.
[23]Teng,Ruming,Cao,Xuyang; Gao,Shunde;Wang,Xin.Hybrid integrated design and realizable strategy of database of mechanical product[J]. Proceedings-2010 6th International Conference on Natural Computation,2010,8:4039-4044.
[24]刘植惠.知识基因探索[J].情报理论与实践.1998,21(1)：62-64.
[25]李伯文.论科学的“遗传”和“变异”[J].科学学与科学技术管理,1985(10)：21-25.
[26]陈洪武,曹守启,许哲,赵波.机械产品基因研究[J].机床与液压,2009,37(4)：155-158.
[27]Hideki Innan, Fyodor Kondrashov.The evolution of gene duplications:classifying and distinguishing between models [J]. Nature Reviews Genetics,2010,11,97-108.
[28]赵秀燕,赵婷婷,魏小鹏,陈微微,许春霞.变型设计中的基因模型与产品优化方法实现[J].计算机工程与应用,2009,45(19)：68-71.
[29]顾新建,祁国宁.过程信息中的基因模型[J].中国机械工程,1998,9(11)：80-85.
[30]陈昆昌,陈子辰.大批量定制产品仿生建模基础[J].机电工程,2005,22(7)：36-41.
[31]John S Gero. Extensions to evolutionary systems in design from genetic engineering and developmental biology[C]. Proceedings 1999 Congress on Evolutionary Computation-CEC99, IEEE, Piscataway, New Jersey,1999:474-479.
[32]John S Gero, Vladimir Kazakov. A genetic engineering extension to genetic algorithms [J]. Evolutionary Systems,2001,9 (1):71-92.
[33]John S Gero. Developments in computer-aided design[C]. In H.Li, Q. Shen, D. Scott and P. Love, eds. INCITE 2000, HKPU Press, Hung Hom, Kowloon, Hong Kong,2000:16-24.
[34]John S Gero. Vladimir Kazakov. Evolving design genes in space layout problems [J]. Artificial Intelligence in Engineering.1998,12(3):163-176.
[35]Chen Yong, Feng Peien, Lin Zhongqin. A genetics-based approach for the principle conceptual design of mechanical products [J].The International Journal of Advanced Manufacturing Technology,2005,27:225-233.
[36]周生祥.基于基因工程思想的产品建模方法[J].机械科学与技术,2000,19(5)：815-820.
[37]GONG Yun-qi, LV Min, WANG Gang, ZHOU Ke. Research on Process Knowledge Management Based on Ontology [J].2007 International Conference on[C].2007.
[38]Samini Subramaniam, Su-Cheng Haw, Poo Kuan Hoong. s-XML: An efficient mapping scheme to bridge XML and relational database[J]. Knowledge-Based Systems,2012,27:369-380.
[39]http://www. w3school. com. cn/xml/xml tree. asp. XML树结构.
[40]孙啸,陆祖宏,谢建明.生物信息学基础,清华大学出版社,2005.
[41]潘旭伟,顾新建,仇元福,程耀东.面向知识管理的知识建模技术[J].2003,9(7)：517-521.
[42]邓聚龙.灰色控制系统.华中理工大学出版社,1985.
[43]王兴元,杨旭.质量预测控制与诊断灰色系统方法[J].系统工程理论与实践,1997,5(5)：105-108.
[44]邓聚龙.灰色系统基本方法[M].武汉：华中理工大学出版社,1987.
[45]严智渊.灰色关联分析与应用[M].南京：江苏科学技术出版社,1989.
[46]邓聚龙.灰色系统理论[M].武汉：华中科技大学出版社,1989.
[47]Wong, C. C., & Lai, H. R. A new grey relational measurement. Journal of Grey System,2000, 4,341-346.
[48]Amani, H., Mochiduki, N., & Yamada, H. (2006). A model for detecting cost-prone classes based on Mahalanobis-Taguchi method. IEICE Transactions on Information and Systems,4, 1347-1358.
[49]Nagao, M., Yamamoto, M., Suzuki, K., & Ohuchi, A. (2001). A face identification system based on the Mahalanobis-Taguchi system.International Transactions in Operational Research,8, 31-45.
[50]Yong-Huang Lin, Pin-Chan Lee, Ta-Peng Chang. Practical expert diagnosis model based on the grey relational analysis technique. Expert Systems with Applications 36 (2009) 1523-1528.
[51]Chang, W. C. A comprehensive study of grey relational generating. Journal of Grey System, 2000(3),53-62.
[2]李国庆.建材企业提高产品质量的十大举措[J].云南建材,2000(3)：12-13.
[3]程元军.项目质量管理[M].北京：机械工业出版社,2007.2.
[4]张根保,曾海峰,王国强等.基于广义信息熵测度的制造过程质量评估[J].中国机械工程,2010,21(20)：2451-2458.
[5]CHIN K S, DUAN G, TANG X. A computer-integrated framework for global quality chain management [J]. The International Journal of Advanced Manufacturing Technology,2006,27(2): 547-560.
[6]刘娜,关玲永.全面质量管理的理论分析[J].北京建筑工程学院学报,2008,2：71-73.
[7]李益兵.光电子产品生命周期质量管理关键技术研究[D].武汉理工大学博士学位论文,2009.
[8]李延来,唐加福,姚建明,等.质量功能展开中选择工程特性的多目标决策方法[J].计算机集成制造系统,2008,14(7)：1363-1369.
[9]李延来,唐加福,姚建明等.质量屋构建的研究进展[J].机械工程学报,2009,45(3)：57-70.
[10]徐显龙,同淑荣,张新卫.基于产品结构树的制造质量信息分类研究[J].制造业自动化,2006,128(12)：37-41.
[11]Gao Huanming, Guo Mingzhe, Zhang Yu, etal. Research on process knowledge management methods based on ontology[J]. Mechanical Engineer,2009(3):68-71.
[12]Yu Honglei, Wang Xiulun, Yang Zhigang, et al. Application of XML to process knowledge management system[J]. Machinery Manufacturing Engineer,2005(6):22-24.
[13]Xu X L,Tong S R,Li B. A conceptual model of manufacturing quality information based on product structure tree[C].The 3rd IEEE International Conference on Wireless Communications, Networking and Mobile Computing (WICOM), Shanghai, China,2007,5129-5132.
[14]李晖,梁工谦,张新卫.一种面向BOM的制造质量信息模型[J].机床与液压,2008,36(1)：608-610.
[15]Nan Xie, Lin Chen and Aiping Li. Fault diagnosis of multistage manufacturing systems based on rough set approach [J]. The International Journal of Advanced Manufacturing Technology,2010,48:1239-1247.
[16]张赤斌,史金飞,易红.基于偏最小二乘回归的工序质量建模[J].东南大学学报,2005,35(5)：702-705.
[17]陈冰,李山,吴宝海,王明微.基于尺寸关联和偏最小二乘回归的多工序质量分析与预测[J].计算机集成制造系统.2009,15(2),389-390.
[18]刘道玉,江平宇.面向多工序制造过程的e-质量控制体系结构研究[J].计算机集成制造系统,2004,13(4)：782-790.
[19]黄诒森,张光毅.生物化学与分子生物学[M].北京：科学出版社,2003.
[20]贺淹才.简明基因工程原理[M].北京：科学出版社,2005.
[21]冯培恩,陈泳,张帅.基于产品基因的概念设计[J].机械工程学报,2002,38(10)：1-6.
[22]董晔弘,向东,龙旦风,刘畅,段广洪.基于贝叶斯网的制造工艺质量建模方法研究.2010全国现代制造集成技术学术会议论文集,2010,398-498.
[23]Teng,Ruming,Cao,Xuyang; Gao,Shunde;Wang,Xin.Hybrid integrated design and realizable strategy of database of mechanical product[J]. Proceedings-2010 6th International Conference on Natural Computation,2010,8:4039-4044.
[24]刘植惠.知识基因探索[J].情报理论与实践.1998,21(1)：62-64.
[25]李伯文.论科学的“遗传”和“变异”[J].科学学与科学技术管理,1985(10)：21-25.
[26]陈洪武,曹守启,许哲,赵波.机械产品基因研究[J].机床与液压,2009,37(4)：155-158.
[27]Hideki Innan, Fyodor Kondrashov.The evolution of gene duplications:classifying and distinguishing between models [J]. Nature Reviews Genetics,2010,11,97-108.
[28]赵秀燕,赵婷婷,魏小鹏,陈微微,许春霞.变型设计中的基因模型与产品优化方法实现[J].计算机工程与应用,2009,45(19)：68-71.
[29]顾新建,祁国宁.过程信息中的基因模型[J].中国机械工程,1998,9(11)：80-85.
[30]陈昆昌,陈子辰.大批量定制产品仿生建模基础[J].机电工程,2005,22(7)：36-41.
[31]John S Gero. Extensions to evolutionary systems in design from genetic engineering and developmental biology[C]. Proceedings 1999 Congress on Evolutionary Computation-CEC99, IEEE, Piscataway, New Jersey,1999:474-479.
[32]John S Gero, Vladimir Kazakov. A genetic engineering extension to genetic algorithms [J]. Evolutionary Systems,2001,9 (1):71-92.
[33]John S Gero. Developments in computer-aided design[C]. In H.Li, Q. Shen, D. Scott and P. Love, eds. INCITE 2000, HKPU Press, Hung Hom, Kowloon, Hong Kong,2000:16-24.
[34]John S Gero. Vladimir Kazakov. Evolving design genes in space layout problems [J]. Artificial Intelligence in Engineering.1998,12(3):163-176.
[35]Chen Yong, Feng Peien, Lin Zhongqin. A genetics-based approach for the principle conceptual design of mechanical products [J].The International Journal of Advanced Manufacturing Technology,2005,27:225-233.
[36]周生祥.基于基因工程思想的产品建模方法[J].机械科学与技术,2000,19(5)：815-820.
[37]GONG Yun-qi, LV Min, WANG Gang, ZHOU Ke. Research on Process Knowledge Management Based on Ontology [J].2007 International Conference on[C].2007.
[38]Samini Subramaniam, Su-Cheng Haw, Poo Kuan Hoong. s-XML: An efficient mapping scheme to bridge XML and relational database[J]. Knowledge-Based Systems,2012,27:369-380.
[39]http://www. w3school. com. cn/xml/xml tree. asp. XML树结构.
[40]孙啸,陆祖宏,谢建明.生物信息学基础,清华大学出版社,2005.
[41]潘旭伟,顾新建,仇元福,程耀东.面向知识管理的知识建模技术[J].2003,9(7)：517-521.
[42]邓聚龙.灰色控制系统.华中理工大学出版社,1985.
[43]王兴元,杨旭.质量预测控制与诊断灰色系统方法[J].系统工程理论与实践,1997,5(5)：105-108.
[44]邓聚龙.灰色系统基本方法[M].武汉：华中理工大学出版社,1987.
[45]严智渊.灰色关联分析与应用[M].南京：江苏科学技术出版社,1989.
[46]邓聚龙.灰色系统理论[M].武汉：华中科技大学出版社,1989.
[47]Wong, C. C., & Lai, H. R. A new grey relational measurement. Journal of Grey System,2000, 4,341-346.
[48]Amani, H., Mochiduki, N., & Yamada, H. (2006). A model for detecting cost-prone classes based on Mahalanobis-Taguchi method. IEICE Transactions on Information and Systems,4, 1347-1358.
[49]Nagao, M., Yamamoto, M., Suzuki, K., & Ohuchi, A. (2001). A face identification system based on the Mahalanobis-Taguchi system.International Transactions in Operational Research,8, 31-45.
[50]Yong-Huang Lin, Pin-Chan Lee, Ta-Peng Chang. Practical expert diagnosis model based on the grey relational analysis technique. Expert Systems with Applications 36 (2009) 1523-1528.
[51]Chang, W. C. A comprehensive study of grey relational generating. Journal of Grey System, 2000(3),53-62.