陶瓷干燥窑热工诊断专家系统
摘要
本论文的研究工作以陶瓷生产干燥过程中的热工参数诊断为对象，建立陶瓷干燥窑热工诊断专家系统。干燥在陶瓷生产中起着生死攸关的作用，一旦干燥系统发生故障，由于故障的相关性、不确定性以及诊断信息的多样性，使得它的故障诊断变得很复杂。而采用专家系统进行诊断，是基于知识的诊断，具有科学性和快速性。
本文首先分析了干燥过程中经常出现的缺陷形式及其产生机理，并总结了解决的方法。这样就得到了领域知识，也是建立合理的知识库体系的基础。接下来本文着重讨论、研究了干燥窑热工诊断专家系统的知识库和推理机制。
干燥过程的原理和缺陷的产生原因，决定了诊断知识的特性，由此建立了关系型知识库体系。在知识的表示中，集面向对象的表示和产生式表示为一体，并对不同深度的知识予以灵活处理。系统推理机的实现，主要取决于基于知识库体系的诊断模型。推理过程是一个基于产生式规则的机制，运用正向推理过程辅以人机交互接口直至将产生干燥问题的原因得出，并给出专家系统的调试建议。
本文最后，简要介绍了在Delphi 5可视化集成开发环境下，实现专家系统的各个模块完成专家系统的设计。
In this paper, the thermal status parameters of the desiccation procedure are worked over in the ceramic production, and setting up the expert system for thermal status diagnose. Desiccation acts a very important role in the ceramic production, once the desiccation system works abnormality, because of the relativity and uncertainty of the faults and the diversity of the diagnoses information it makes complexity to diagnose the faults. When we use expert system based on knowledge to diagnose will be scientific and quick.
The fist, we analyze the defects often appeared and how the defaults coming into being in the desiccation procedure, and summarization the solution. So we ?got the field knowledge, it's also the base of setting up the architecture of repository. The followings are discussing and studying the reasoning mechanism and repository of the expert system.
The characteristics of the diagnoses knowledge are decided by the principle of the desiccation procedure and the cause of formation for the defect., according to above we set up the relation architecture repository. When expression the knowledge we integrate object oriented expression and produce formula expression, and flexible deal with knowledge with different profundities. The realization of the reasoning machine mainly lies on the diagnoses model, which based on repository architecture. Reasoning procedure is mechanism, which based on produce formula, using positive direction reasoning with the help of
man-machine conversation interface up to get the reason of the desiccation problem, followed with the debugging suggestions given by the expert system.
At the end of this paper, we make a brief introduction on how to realization all modules of the expert system in environment, which is visible, integrated of Delphi 5.
本文首先分析了干燥过程中经常出现的缺陷形式及其产生机理，并总结了解决的方法。这样就得到了领域知识，也是建立合理的知识库体系的基础。接下来本文着重讨论、研究了干燥窑热工诊断专家系统的知识库和推理机制。
干燥过程的原理和缺陷的产生原因，决定了诊断知识的特性，由此建立了关系型知识库体系。在知识的表示中，集面向对象的表示和产生式表示为一体，并对不同深度的知识予以灵活处理。系统推理机的实现，主要取决于基于知识库体系的诊断模型。推理过程是一个基于产生式规则的机制，运用正向推理过程辅以人机交互接口直至将产生干燥问题的原因得出，并给出专家系统的调试建议。
本文最后，简要介绍了在Delphi 5可视化集成开发环境下，实现专家系统的各个模块完成专家系统的设计。
In this paper, the thermal status parameters of the desiccation procedure are worked over in the ceramic production, and setting up the expert system for thermal status diagnose. Desiccation acts a very important role in the ceramic production, once the desiccation system works abnormality, because of the relativity and uncertainty of the faults and the diversity of the diagnoses information it makes complexity to diagnose the faults. When we use expert system based on knowledge to diagnose will be scientific and quick.
The fist, we analyze the defects often appeared and how the defaults coming into being in the desiccation procedure, and summarization the solution. So we ?got the field knowledge, it's also the base of setting up the architecture of repository. The followings are discussing and studying the reasoning mechanism and repository of the expert system.
The characteristics of the diagnoses knowledge are decided by the principle of the desiccation procedure and the cause of formation for the defect., according to above we set up the relation architecture repository. When expression the knowledge we integrate object oriented expression and produce formula expression, and flexible deal with knowledge with different profundities. The realization of the reasoning machine mainly lies on the diagnoses model, which based on repository architecture. Reasoning procedure is mechanism, which based on produce formula, using positive direction reasoning with the help of
man-machine conversation interface up to get the reason of the desiccation problem, followed with the debugging suggestions given by the expert system.
At the end of this paper, we make a brief introduction on how to realization all modules of the expert system in environment, which is visible, integrated of Delphi 5.
引文
[1] 李桢．我国干燥技术的进展概况．中国化工学会，化工机械学会年会，1989
[2] A·φ·契津斯基(苏)．陶瓷原料与制品干燥．中国建筑工业出版社，1980
[3] 陶瓷原料的干燥(法)．砖瓦．No．4，1978
[4] 大块砖和条行砖无变形干燥的条件(德、法)．砖瓦．No．2，1978
[5] 陈世福，陈兆乾．人工智能与知识工程．南京大学出版社，1997
[6] 冯博琴．实用专家系统．电子工业出版社，1992
[7] Baklouti S, Chartier T. Mechanical Properties of dry-pressed ceramic green products[J]. J Am Ceram Soc, 1997,80(8):1992-1996
[8] Zheng J, Reed J S. Particle and granule parameters affecting compaction efficiency in dry pressing [J]. J Am Ceram Soc,1998,71(11)
[9] 陈敏恒等．化工原理下册．北京；化学工业出版社，1986．236
[10] 若利P等．木材干燥理论、实践和经济．宋闯译，中国林业出版社．1986，236
[11] 曾令可．陶瓷干燥过程原理分析．华南理工大学材料学院，2001
[12] R．B．基伊．干燥原理及其应用．清华大学出版社，1992
[13] 秦光，王志洁．干燥装置手册．上海科学技术出版社，1983
[14] As M ujumdar．干燥中的传热问题(第五届全国干燥技术交流会论文集)．1995，23
[15] R B Keey. Progress in Understanding Drying(第五届全国干燥会议论文集)．1995，33
[16] 刘向阳．调整干燥曲线和制度提高—次烧成釉面砖质量．中国建材，1999年第6期
[17] 刘三水．坯体干燥时产生的缺陷分析．景德镇陶瓷，第三期(总第77期)
[18] 马炳初．关于硅砖内部裂纹的探讨．河南冶金，第2期总第31期
[19] 刘世革，管可杰．精陶产品开裂缺陷分析及解决方法．江苏陶瓷，Vol．31．No．2(1998)
[20] 赵炳钧，曹仁忠．浅析釉面砖龟裂缺陷产生的原因及克服方法．山东陶瓷，Vol．20．No．1，Mar．1997
[21] 谯永林．釉面砖针孔、釉泡缺陷分析．陶瓷研究，Vol．13 No．3，Sep．1998
[22] 田盛丰等．人工智能原理与应用．北京理工大学出版社，1992
[23] 何新贵．知识处理与专家系统．国防工业出版社，1990
[24] 王永庆．人工智能原理与方法．西安交通大学出版社，1998
[25] Paul R. Cohen. Heuristic Reasoning about Uncertainty: An Artificial Intelligence Approach. Pitman Advanced Publishing program, 1985
[26] 刘大有．用于智能诊断的外推推理方法．模式识别与人工智能，第3期，1991．9
[27] 陆汝钤等．专家系统开发环境．科学出版社，1994
[28] 万沃华．Delphi 5编程指南．西北工业大学出版社，1997．4
[29] David J．kruglinsky．Delphi 5技术内幕．清华大学出版社，1995
[30] 吴长华，章少华．基于主动式知识库的专家系统建模和推理研究．哈尔谈工业大学学报，Vol．27，No．2 April．1995
[31] Harmon, P. & King, D. Expert Systems: Artificial Intelligence in Business. New York, John Wiley & Sons, Ins. 1985
[32] 程从从．一种基于模型推理的故障诊断专家系统．小型微型计算机系统，1997，18(2)
[33] Mickey Williamson. Artificial Intelligence for Micro-Computers, The guide for business decision makers. Brady Communication Company Inc, 1986
[2] A·φ·契津斯基(苏)．陶瓷原料与制品干燥．中国建筑工业出版社，1980
[3] 陶瓷原料的干燥(法)．砖瓦．No．4，1978
[4] 大块砖和条行砖无变形干燥的条件(德、法)．砖瓦．No．2，1978
[5] 陈世福，陈兆乾．人工智能与知识工程．南京大学出版社，1997
[6] 冯博琴．实用专家系统．电子工业出版社，1992
[7] Baklouti S, Chartier T. Mechanical Properties of dry-pressed ceramic green products[J]. J Am Ceram Soc, 1997,80(8):1992-1996
[8] Zheng J, Reed J S. Particle and granule parameters affecting compaction efficiency in dry pressing [J]. J Am Ceram Soc,1998,71(11)
[9] 陈敏恒等．化工原理下册．北京；化学工业出版社，1986．236
[10] 若利P等．木材干燥理论、实践和经济．宋闯译，中国林业出版社．1986，236
[11] 曾令可．陶瓷干燥过程原理分析．华南理工大学材料学院，2001
[12] R．B．基伊．干燥原理及其应用．清华大学出版社，1992
[13] 秦光，王志洁．干燥装置手册．上海科学技术出版社，1983
[14] As M ujumdar．干燥中的传热问题(第五届全国干燥技术交流会论文集)．1995，23
[15] R B Keey. Progress in Understanding Drying(第五届全国干燥会议论文集)．1995，33
[16] 刘向阳．调整干燥曲线和制度提高—次烧成釉面砖质量．中国建材，1999年第6期
[17] 刘三水．坯体干燥时产生的缺陷分析．景德镇陶瓷，第三期(总第77期)
[18] 马炳初．关于硅砖内部裂纹的探讨．河南冶金，第2期总第31期
[19] 刘世革，管可杰．精陶产品开裂缺陷分析及解决方法．江苏陶瓷，Vol．31．No．2(1998)
[20] 赵炳钧，曹仁忠．浅析釉面砖龟裂缺陷产生的原因及克服方法．山东陶瓷，Vol．20．No．1，Mar．1997
[21] 谯永林．釉面砖针孔、釉泡缺陷分析．陶瓷研究，Vol．13 No．3，Sep．1998
[22] 田盛丰等．人工智能原理与应用．北京理工大学出版社，1992
[23] 何新贵．知识处理与专家系统．国防工业出版社，1990
[24] 王永庆．人工智能原理与方法．西安交通大学出版社，1998
[25] Paul R. Cohen. Heuristic Reasoning about Uncertainty: An Artificial Intelligence Approach. Pitman Advanced Publishing program, 1985
[26] 刘大有．用于智能诊断的外推推理方法．模式识别与人工智能，第3期，1991．9
[27] 陆汝钤等．专家系统开发环境．科学出版社，1994
[28] 万沃华．Delphi 5编程指南．西北工业大学出版社，1997．4
[29] David J．kruglinsky．Delphi 5技术内幕．清华大学出版社，1995
[30] 吴长华，章少华．基于主动式知识库的专家系统建模和推理研究．哈尔谈工业大学学报，Vol．27，No．2 April．1995
[31] Harmon, P. & King, D. Expert Systems: Artificial Intelligence in Business. New York, John Wiley & Sons, Ins. 1985
[32] 程从从．一种基于模型推理的故障诊断专家系统．小型微型计算机系统，1997，18(2)
[33] Mickey Williamson. Artificial Intelligence for Micro-Computers, The guide for business decision makers. Brady Communication Company Inc, 1986