基于神经网络专家系统的胎面生产线故障诊断的研究
|
摘要
本文应用神经网络及相应的数学工具结合专家系统对胎面生产线相关的设备进行准确的故障诊断,使该故障诊断系统既具有专家系统的逻辑思维能力,又具有神经网络的经验思维能力,它不仅可以减少和防止故障对生产设备造成的影响,使系统尽快恢复正常运行,减少经济损失,而且对确保轮胎生产的质量、产量和设备人员安全也有重要的意义。
本文以轮胎胎面复合挤出联动线PLC控制系统为研究背景,概述了胎面生产线的工艺流程,根据其控制系统的总体结构和性能特点,分析了轮胎胎面复合挤出联动线PLC控制系统故障的类型、范围及产生的原因,从中选取了三类故障作为本文的实验对象进行简单分析。
论文对智能故障诊断系统的产生和发展现状作了综述,分析了神经网络与专家系统的结构和特点,以及在故障诊断中的运用,讨论了神经网络与专家系统单独应用于故障诊断的不足,并确立了神经网络与专家系统结合的故障诊断体系。同时对神经网络专家系统的基本结构以及知识库、推理机、解释器以及人机界面等模块的建立进行的详细论述。
最后对该系统进行了仿真,通过Matlab编程软件对现场采集的螺杆挤出机的转速,压力,主机电流等数据进行仿真,通过基于MATLAB的神经网络故障判断的仿真实验,证明了本文研究的有效性,基本建立了基于神经网络专家系统的胎面生产线故障诊断的工作模式。为后续的故障诊断研究提供了一定的理论基础,为实际开发、生产提供了一个参考方向。
In this paper, neural network and the corresponding mathematical tools are used to accurately diagnose the fault of tread production line and related equipments, Fault diagnosis system not only have the expert system logic ability, but also has the experience of neural networks. It can reduce and prevent the failure of the impact of production equipment, allowing the system to resume normal operation as soon as possible, and to reduce economic losses, it is also important to ensure the quality of tire production, production and equipment and personnel safety.
In the background of PLC controlling system of the tire extruding line and according to system population structure and function characteristic, this paper analyses the types, range and causes of the faults in the line. This article outlines the tread production line process and fault diagnosis, based on a common approach and the analysis process, the classification of the common faults. In this paper, three types of failure are selected as the subjects. Discusses the neural network and expert system fault diagnosis applied to the shortcomings and established the combination of neural network and expert system fault diagnosis system and based on neural network expert system's basic structure and establish the modules of a knowledge base, inference engine and explain.
Finally the system is simulated, based on MATLAB, and do an experiment of neural network to diagnose the fault, it is proved that the simulation experiments this study is the validity to diagnose the fault.Based on neural network expert system, established fault diagnosis of tread production line operating mode. For the follow-up study, Fault diagnosis provides a theoretical basis for the actual development, production and provides a reference direction.
本文以轮胎胎面复合挤出联动线PLC控制系统为研究背景,概述了胎面生产线的工艺流程,根据其控制系统的总体结构和性能特点,分析了轮胎胎面复合挤出联动线PLC控制系统故障的类型、范围及产生的原因,从中选取了三类故障作为本文的实验对象进行简单分析。
论文对智能故障诊断系统的产生和发展现状作了综述,分析了神经网络与专家系统的结构和特点,以及在故障诊断中的运用,讨论了神经网络与专家系统单独应用于故障诊断的不足,并确立了神经网络与专家系统结合的故障诊断体系。同时对神经网络专家系统的基本结构以及知识库、推理机、解释器以及人机界面等模块的建立进行的详细论述。
最后对该系统进行了仿真,通过Matlab编程软件对现场采集的螺杆挤出机的转速,压力,主机电流等数据进行仿真,通过基于MATLAB的神经网络故障判断的仿真实验,证明了本文研究的有效性,基本建立了基于神经网络专家系统的胎面生产线故障诊断的工作模式。为后续的故障诊断研究提供了一定的理论基础,为实际开发、生产提供了一个参考方向。
In this paper, neural network and the corresponding mathematical tools are used to accurately diagnose the fault of tread production line and related equipments, Fault diagnosis system not only have the expert system logic ability, but also has the experience of neural networks. It can reduce and prevent the failure of the impact of production equipment, allowing the system to resume normal operation as soon as possible, and to reduce economic losses, it is also important to ensure the quality of tire production, production and equipment and personnel safety.
In the background of PLC controlling system of the tire extruding line and according to system population structure and function characteristic, this paper analyses the types, range and causes of the faults in the line. This article outlines the tread production line process and fault diagnosis, based on a common approach and the analysis process, the classification of the common faults. In this paper, three types of failure are selected as the subjects. Discusses the neural network and expert system fault diagnosis applied to the shortcomings and established the combination of neural network and expert system fault diagnosis system and based on neural network expert system's basic structure and establish the modules of a knowledge base, inference engine and explain.
Finally the system is simulated, based on MATLAB, and do an experiment of neural network to diagnose the fault, it is proved that the simulation experiments this study is the validity to diagnose the fault.Based on neural network expert system, established fault diagnosis of tread production line operating mode. For the follow-up study, Fault diagnosis provides a theoretical basis for the actual development, production and provides a reference direction.
引文
[1]任康.基于神经网络的电器故障诊断研究.2006.04
[2]李福进,安逸.基于BP神经网络的冷轧薄板生产线故障诊断系统研究.河北冶金.2008.06
[3]Francis E H, Shen L X. Fault diagnosis based on rough set t theory [J] Engineering Applacations of A artificial In tally gence,2003,16 (1):39 43.
[4]吴静,柳世考,邓堃.基于改进BP神经网络的故障诊断方法.工业仪表与自动化装置.2007.03
[5]刘锋,夏春先,黄振和.基于人工神经网络的故障诊断专家系统.国外电子测量技术.2004.04
[6]Patton R J, Chen J, Siew T M. Fault diagnosis in non21inear dynamic systems via neural networks [C]//Proc. IEEE In. Conf. Control 94, Covent ry, U K,1994,1:1346-1351.
[7]刘爱元,杜鑫.基于神经网络的实时故障诊断系统设计.2002.05
[8]张荣.基于神经网络的智能故障诊断技术.控制理论与应用.2003.
[9]谈理,刘谨,叶赛蓬.生产线故障诊断知识引擎系统的集成推理方法.哈尔滨工业大学学报.2009.01
[10]Guardado J L, Naredo J L. A comparative study of neural network efficiency in power transformers diagnosis using dissolved gas analysis [J]. I EEE Trans. on Power Deli very,2001,16(4):643-647.
[11]迭华,张进武.、半钢子午线轮胎胎面生产线控制系统常见故障及解决措施.轮胎工业.2004
[12]魏晓宾,马小平,李亚朋.故障诊断技术综述.煤矿机电.2009
[13]李安,胡柏青,赵济民.故障诊断专家系统的可视化设计及实现.海军工程学院学报.1997年第3期.
[14]时圣革,常文兵,李祥臣.故障诊断专家系统及发展趋势.机械工程师.2006年11期
[15]陈文清.基于ANNES的故障智能诊断技术研究.
[16]Ishibuchi H, Nii M. Numerical Analysis of the Learning of Fuzzified Neural Networks from Fuzzy If then Rules[J]. Fuzzy Set s and Systems,2001, 120(2):2812307.
[17]李萍,曾令可,税安泽,金雪莉,刘艳春,王慧.基于MATLAB的BP神经网络预测系统的设计.计算机应用与软件.2008.04
[18]范海锋,潘成胜,李航.基于专家系统与神经网络的远程故障诊断研究.沈阳工业学院学报.2003.09.
[19]刘云.基于联想记忆神经网络的故障诊断.湖南文理学院学(自然科学版).2008.03
[20]孙雅囡,杨晓东.基于模糊神经网络的故障诊断新方法.先进制造与管理.2008年第27卷第7期.
[21]Jiang,Minghu,Gielen,Georges,Zhang,Bo.Fast learning algorithm for feed forward neural networks [J]. Applied Intelligence, v18, January/February,2003:37-54
[22]吴宗彦,韩煜,张建军,张,利.基于模糊神经网络的自动生产线故障诊断方法研究.中国机械工程.2008.05
[23]许听,潘铭志,王晶禹,潘宏伙.基于神经网络的故障诊断方法的研究.机械管理开发.2007.04.
[24]D. whitley. An overview of evolutionary algorithm:practical issues and common pitfalls, Information and Soft reTechnology.2001(43):817-831
[25]霍志红,张志学,郭江,唐必光.基于神经网络的故障诊断研究.工业控制计算机.2001.
[26]吴凌云.基于神经网络的故障诊断专家系统.现代电子技术.2003.01.
[27]成曙,董程林,陈科吉,杜磊.基于神经网络的故障诊断专家系统研究.微计算机信息.2006年22卷11期.
[28]潘传友,陈幼平,史玉升,周祖德.基于神经网络的智能故障诊断系统的开发研究.机械与电子.2001.04.
[29]吴今培,肖建华.智能故障诊断与专家系统[M]北京：科学出版社,1997.1～163
[30]王致杰,王耀才等.现代大型设备故障智能诊断技术的现状与展望[J]煤矿机械,2003(07)：103
[31]M. Sampath, etal. Diagnosability of Discrete EventSystems.Automatica Control,1995,40(9):1555-1575.
[32]崔彦平,王秉仁等.基于神经网络的综合智能故障诊断专家系统[J]机电一体化,2003(04)：102～104
[33]Frank P M. Fault diagnosis in dynamic systems using analytical and knowledge2based redundancy 2 A surveys and some new re2sult s [J]. A automatic,1990,26 (3):459-474.
[34]雷勇.基于神经网络和专家系统的变电站故障诊断和处理系统的构想[J]福建电力与电工,2003(12)：22～24
[35]李海港,周一恒.神经网络故障诊断专家系统的结构设计.煤矿机械.2005.01.
[36]李大虞,王莉娟,向华.胎面生产线控制系统改造.微机应用与自动控制.2007.
[37]杨廷勇,施冲.基于经验思维和逻辑思维的故障诊断专家系统[J]电力自动化设备,2002(10)：24～26
[38]鞠万群,韩秋实.基于神经网络与规则库的故障诊断专家系统[J]北京机械工业学院学报,2001(03)：6～8
[39]常晓丽.基于Matlab的BP神经网络设计.机械工程与自动化.2006.08
[40]何新贵.知识处理与专家系统[M].北京：国防工业出版社,1995：10～40
[41]梁禹,褚俊霞,王娜.一种新的神经网络故障诊断方法.自动化与仪器仪表.2004.03.
[42]马成才,顾晓东.基于神经网络组与故障分级的故障诊断.系统工程与电子技术.2009.01.
[43]BemjerjAetal. A Neural] Networks Approach for Identification and Fault Diagnosis Damien Systems. IEEETrans. onIM,1994,43(6)
[44]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[J]杭州电子工业学院学报,2001(02)：30～32
[45]Narendra K, K Parthasarathy. Identification and control of dynamical systems using neural networks [J]. IEEE Trans on Neural Networks,1990, 1(1):4-27.
[46]于洋,周学伟,杨青,赵亚威.基于模糊神经网络的锅炉系统故障诊断研究.工业仪表与自动化装置。2008.02
[47]Lu Hongjun,Setiono Rudy, Liu Huan. Effective data mining using neural network [J]. IEEE Transactions on Knowledge and Data Engineering,1996, 8 (6):957-961.
[48]张松鹤,罗跃纲,岳玉莲.基于自适应神经网络的智能故障诊断研究.大连民族学院学报.2004.01
[49]Yang B, Ding H. Expert system of transformer fault diagnosis based on knowledge base[J]. Proceedings of the Chinese Society for Elect racial Engineering,2002,22 (10):122-124.
[50]廖建军.基于神经网络的故障诊断数据挖掘系统结构的研究.信息技术.2009.04.
[51]Aygen Z E, Seker S, Bagriyanik Metal. Fault section estimation in elect recall power systems using artificial neural network approach[C] IEEE Transmission and Distribution Conference,1999 (1):466-469.
[52]Wang M H. A novel extension neural network for power transformer fault diagnosis [J]. IEE Proceedings 2 Generation, Transmission and Distribution, 2003,150 (6):669-685.
[53]张毅,王红,朱永波,杨占才.一个基于神经网络故障诊断专家系统研究.测控技术.2004.
[54]周黄斌,周永华,朱丽娟.基于MATLAB的改进BP神经网络的实现与比较.计算技术与自动化.2008.03
[55]Young B, Moon Divers C Kenneth. A EW S:an integrated know ledge-based system with neural network s for reliability predict ion [J]. Computers in Industry,1998,35:101-108.
[56]李水祥,谢文武.MATLAB语言的神经网络工具箱及应用.高等函授学报(自然科学版).2007.02
[57]Haykin, SNeural Networks, U. S.A.,Macmillan College Publishing Company,1994:21.
[58]虞和济.基于神经网络的智能诊断[M].北京：冶金工业出版社
[2]李福进,安逸.基于BP神经网络的冷轧薄板生产线故障诊断系统研究.河北冶金.2008.06
[3]Francis E H, Shen L X. Fault diagnosis based on rough set t theory [J] Engineering Applacations of A artificial In tally gence,2003,16 (1):39 43.
[4]吴静,柳世考,邓堃.基于改进BP神经网络的故障诊断方法.工业仪表与自动化装置.2007.03
[5]刘锋,夏春先,黄振和.基于人工神经网络的故障诊断专家系统.国外电子测量技术.2004.04
[6]Patton R J, Chen J, Siew T M. Fault diagnosis in non21inear dynamic systems via neural networks [C]//Proc. IEEE In. Conf. Control 94, Covent ry, U K,1994,1:1346-1351.
[7]刘爱元,杜鑫.基于神经网络的实时故障诊断系统设计.2002.05
[8]张荣.基于神经网络的智能故障诊断技术.控制理论与应用.2003.
[9]谈理,刘谨,叶赛蓬.生产线故障诊断知识引擎系统的集成推理方法.哈尔滨工业大学学报.2009.01
[10]Guardado J L, Naredo J L. A comparative study of neural network efficiency in power transformers diagnosis using dissolved gas analysis [J]. I EEE Trans. on Power Deli very,2001,16(4):643-647.
[11]迭华,张进武.、半钢子午线轮胎胎面生产线控制系统常见故障及解决措施.轮胎工业.2004
[12]魏晓宾,马小平,李亚朋.故障诊断技术综述.煤矿机电.2009
[13]李安,胡柏青,赵济民.故障诊断专家系统的可视化设计及实现.海军工程学院学报.1997年第3期.
[14]时圣革,常文兵,李祥臣.故障诊断专家系统及发展趋势.机械工程师.2006年11期
[15]陈文清.基于ANNES的故障智能诊断技术研究.
[16]Ishibuchi H, Nii M. Numerical Analysis of the Learning of Fuzzified Neural Networks from Fuzzy If then Rules[J]. Fuzzy Set s and Systems,2001, 120(2):2812307.
[17]李萍,曾令可,税安泽,金雪莉,刘艳春,王慧.基于MATLAB的BP神经网络预测系统的设计.计算机应用与软件.2008.04
[18]范海锋,潘成胜,李航.基于专家系统与神经网络的远程故障诊断研究.沈阳工业学院学报.2003.09.
[19]刘云.基于联想记忆神经网络的故障诊断.湖南文理学院学(自然科学版).2008.03
[20]孙雅囡,杨晓东.基于模糊神经网络的故障诊断新方法.先进制造与管理.2008年第27卷第7期.
[21]Jiang,Minghu,Gielen,Georges,Zhang,Bo.Fast learning algorithm for feed forward neural networks [J]. Applied Intelligence, v18, January/February,2003:37-54
[22]吴宗彦,韩煜,张建军,张,利.基于模糊神经网络的自动生产线故障诊断方法研究.中国机械工程.2008.05
[23]许听,潘铭志,王晶禹,潘宏伙.基于神经网络的故障诊断方法的研究.机械管理开发.2007.04.
[24]D. whitley. An overview of evolutionary algorithm:practical issues and common pitfalls, Information and Soft reTechnology.2001(43):817-831
[25]霍志红,张志学,郭江,唐必光.基于神经网络的故障诊断研究.工业控制计算机.2001.
[26]吴凌云.基于神经网络的故障诊断专家系统.现代电子技术.2003.01.
[27]成曙,董程林,陈科吉,杜磊.基于神经网络的故障诊断专家系统研究.微计算机信息.2006年22卷11期.
[28]潘传友,陈幼平,史玉升,周祖德.基于神经网络的智能故障诊断系统的开发研究.机械与电子.2001.04.
[29]吴今培,肖建华.智能故障诊断与专家系统[M]北京：科学出版社,1997.1～163
[30]王致杰,王耀才等.现代大型设备故障智能诊断技术的现状与展望[J]煤矿机械,2003(07)：103
[31]M. Sampath, etal. Diagnosability of Discrete EventSystems.Automatica Control,1995,40(9):1555-1575.
[32]崔彦平,王秉仁等.基于神经网络的综合智能故障诊断专家系统[J]机电一体化,2003(04)：102～104
[33]Frank P M. Fault diagnosis in dynamic systems using analytical and knowledge2based redundancy 2 A surveys and some new re2sult s [J]. A automatic,1990,26 (3):459-474.
[34]雷勇.基于神经网络和专家系统的变电站故障诊断和处理系统的构想[J]福建电力与电工,2003(12)：22～24
[35]李海港,周一恒.神经网络故障诊断专家系统的结构设计.煤矿机械.2005.01.
[36]李大虞,王莉娟,向华.胎面生产线控制系统改造.微机应用与自动控制.2007.
[37]杨廷勇,施冲.基于经验思维和逻辑思维的故障诊断专家系统[J]电力自动化设备,2002(10)：24～26
[38]鞠万群,韩秋实.基于神经网络与规则库的故障诊断专家系统[J]北京机械工业学院学报,2001(03)：6～8
[39]常晓丽.基于Matlab的BP神经网络设计.机械工程与自动化.2006.08
[40]何新贵.知识处理与专家系统[M].北京：国防工业出版社,1995：10～40
[41]梁禹,褚俊霞,王娜.一种新的神经网络故障诊断方法.自动化与仪器仪表.2004.03.
[42]马成才,顾晓东.基于神经网络组与故障分级的故障诊断.系统工程与电子技术.2009.01.
[43]BemjerjAetal. A Neural] Networks Approach for Identification and Fault Diagnosis Damien Systems. IEEETrans. onIM,1994,43(6)
[44]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[J]杭州电子工业学院学报,2001(02)：30～32
[45]Narendra K, K Parthasarathy. Identification and control of dynamical systems using neural networks [J]. IEEE Trans on Neural Networks,1990, 1(1):4-27.
[46]于洋,周学伟,杨青,赵亚威.基于模糊神经网络的锅炉系统故障诊断研究.工业仪表与自动化装置。2008.02
[47]Lu Hongjun,Setiono Rudy, Liu Huan. Effective data mining using neural network [J]. IEEE Transactions on Knowledge and Data Engineering,1996, 8 (6):957-961.
[48]张松鹤,罗跃纲,岳玉莲.基于自适应神经网络的智能故障诊断研究.大连民族学院学报.2004.01
[49]Yang B, Ding H. Expert system of transformer fault diagnosis based on knowledge base[J]. Proceedings of the Chinese Society for Elect racial Engineering,2002,22 (10):122-124.
[50]廖建军.基于神经网络的故障诊断数据挖掘系统结构的研究.信息技术.2009.04.
[51]Aygen Z E, Seker S, Bagriyanik Metal. Fault section estimation in elect recall power systems using artificial neural network approach[C] IEEE Transmission and Distribution Conference,1999 (1):466-469.
[52]Wang M H. A novel extension neural network for power transformer fault diagnosis [J]. IEE Proceedings 2 Generation, Transmission and Distribution, 2003,150 (6):669-685.
[53]张毅,王红,朱永波,杨占才.一个基于神经网络故障诊断专家系统研究.测控技术.2004.
[54]周黄斌,周永华,朱丽娟.基于MATLAB的改进BP神经网络的实现与比较.计算技术与自动化.2008.03
[55]Young B, Moon Divers C Kenneth. A EW S:an integrated know ledge-based system with neural network s for reliability predict ion [J]. Computers in Industry,1998,35:101-108.
[56]李水祥,谢文武.MATLAB语言的神经网络工具箱及应用.高等函授学报(自然科学版).2007.02
[57]Haykin, SNeural Networks, U. S.A.,Macmillan College Publishing Company,1994:21.
[58]虞和济.基于神经网络的智能诊断[M].北京：冶金工业出版社