摘要
旋转机械故障诊断是近年来国内外发展较快的一门新兴技术。旋转机械监测与诊断系统的研究与应用对于避免巨额的经济损失和灾难性事故的发生有着重要意义。专家系统是人工智能的一个分支。它实际上是一种基于知识的计算机软件系统,它拥有某个领域内众多专家的知识和经验,并能像专家那样运用这些知识,通过推理,做出智能决策。使用专家系统可以对大型旋转机械的故障进行比较准确的诊断,促进维修方式从预防性到预测性的转变,延长机组寿命,从而创造巨大的社会效益和经济效益。
本论文以风机旋转机械为研究对象,研究了风机旋转机械设备故障诊断专家系统的基本原理。并结合费舍尔判别分析法、粗糙集、模糊技术和蚁群算法,提出了将传统风机机械故障诊断专家系统知识自动化推理的技术。
论文提出的风机故障诊断系统分为前端在线简单监测和后端自动诊断专家系统两部分。在线简单监测部分的实现,运用了费舍尔判别分析法及其他的一些监测技术;自动故障诊断系统,综合运用了粗糙集、知识特征化、模糊算法和蚁群算法等技术。最后通过工厂采集的实际数据进行测试,依据本论文提出的技术建立的故障自动诊断专家系统可以进行风机机械状态识别和故障分析解释。
Nowadays, fault diagnosis of rotating machine is one of the quickly developed emerging technologies at home and abroad. Research and application on monitoring and analyzing system of the large rotating machine is significant to avoid substantive economic loss and disastrous accident. Expert system is a branch of artificial intelligence. Actually, it is a computer software system based on knowledge, owning much expert's knowledge and experience in one field and using knowledge like expert to make intelligent decision by reasoning. Expert system can diagnose fault of the large rotating machine more accurate, promote change of maintenance way from prevention to forecast and extend machine's life. It creates a great societal and economical benefit.
In this paper, ventilator rotating machine is the object of research, and the basic theorem for expert system of fault diagnosis to ventilator rotating machine is researched. Combining fisher's discriminant analysis, rough set, fuzzy and ant algorithm, technology of knowledge auto-reason has been presented, which can change the traditional fault diagnosis expert system for ventilator machine to automatic diagnosis.
The system of ventilator fault diagnosis has been presented that includes two parts: the fore is an online simple monitor and the end is an expert system of automatic diagnosis. The implementation of online simple monitor contains fisher's discriminant analysis and other monitor technology; the expert system of automatic diagnosis uses rough set, characteristic knowledge, fuzzy and ant algorithm. Finally, the actual data is applied to test the expert system of fault auto-diagnosis. The result is that the expert system of fault auto-diagnosis can discriminate ventilator machine in good state from bad state and explain the reason of fault.
引文
[1]A.W.LEES,M.I.FRISWELL.THE EVALUATION OF ROTOR IMBALANCE IN FLEXIBLY MOUNTED MACHINES[J].Journal of Sound and Vibration,1997,208(5):671-683
[2]史天运,王信义,张之敬.FMS故障诊断专家系统的研究与开发[J].北京理工大学学报,1998,18(3):290-295
[3]虞和济.故障诊断的基本原理[M].北京:冶金工业出版社,1988
[4]袁洪芳.FMS故障诊断与预维修技术研究[D].北京:北京理工大学机械工程与自动化学院,2001
[5]刘相臣张秉淑.化工装备事故分析与预防[M].北京:化学工业出版社,1994
[6]M K Hu.Visual pattern recognition by moment invariants[J].IEEE trans Information Theory,1962,8:179-187
[7]陈书凯.转子系统振动故障分析与诊断[D].南京:南京航空航天大学,2005
[8]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].北京:清华大学出版社,2002
[9]刘纪锋.基于知识的航空发动机转子动力学分析及诊断方法研究[D].南京:南京航空航天大学,1995
[10]A V Tuzikov,S A Sheynin.Symmetry Measure Computation for Convex Polyhedra[J].Journal of Mathematical/maging and Vision.2002,16:41-56
[11]虞和济,陈长征,张省,等.基于神经网络的智能诊断[M].北京:冶金工业出版社,2000
[12]关守平,柴天佑.一种数字网络推理技术[J].控制与决策,1996,11(增刊Ⅰ):125-131
[13]杨叔子,丁洪.基于知识的诊断推理[M].北京:清华大学出版社,1993
[14]T.Asakura,T.Kobayashi,S.Hayashi.A study of fault diagnosis system using neural networks[J],Stochastic Systems Theory and Its Applications,1998,1,19-24
[15]虞和济,韩庆大,李沈,等.设备故障诊断工程[M].北京:冶金工业出版社,2001,1252-1254
[16]钟佑明,秦树人,汤宝平.一种振动信号新变换法的研究[J].振动工程学报,2002,15(2):233-238
[17]应怀樵.波形和频谱分析与随机数据处理[M].北京:中国铁道出版社,1983
[18]Gianpaolo Evangelista.Comb and multiplexed wavelet and their applications to signal processing[J].IEEE Transactions on Signal processing,1994,42(2):292-303
[19]Nurgun Erdol.Wavelet transform based adaptive filters:analysis and new results[J].IEEE Transaction on Signal Processing,1996,44(9):2163-2171
[20]曲庆文,马浩.油膜振荡的特征及判别方法[J].机械科学与技术,2000,19(1):91-93
[21]刘朝山,张继华.油膜涡动和油膜振荡的特征频率[J].风机技术,2001,6:52-54
[22]杨剑刚.一种新的旋转机械故障诊断系统[J],东南大学学报,1995,25(1),51-56
[23]李录平,韩守木,黄树红,等.旋转机械故障特征的定性提取[J].华中理工大学学报,1998,26(1):101-103.
[24]陈长征.风机不平衡故障的现场诊断与控制[J].东北大学学报,1997,18,56-69
[25]寇惠.故障诊断的振动理论基础[M].北京:冶金工业出版社,1992
[26]韩捷,张瑞林等.旋转机械故障机理及诊断技术[M].北京:机械工业出版社.1997, 91-92
[27]张正松.旋转机械振动监测及故障诊断[M].北京:机械工业出版社,1991
[28]沈庆根.化工机器故障诊断技术[M].杭州:浙江大学出版社,1994
[29]伍洋.用状态监测判断不对中故障[J].技术与应用,2004,3:47-48
[30]王殿武,张敬伟.不对中故障的分析与诊断技巧[J].设备管理与维修,2005,10:32-34
[31]李志荣,侯洪海,曾亚宏.转子不平衡的故障特征[J].洛阳工学院学报,1999,20(1):54-57
[32]韩延良.离心机不平衡故障的早期诊断[J].振动测试与诊断,1992,12(1):50-51
[33]Yong Xu,Jing-Yu Yang,Zhong Jin.A.Novel method for Fisher discriminant analysis[J].Pattern Recognition,2004,37(2):381-384
[34]蒋浩天,E.L.拉塞尔,R.D.布拉次.工业系统的故障检测与诊断[M].北京:机械工业出版社,2003,30-65
[35]Billings S A,Lee K L.Nonlinear Fisher discriminant analysis using a minimum squared error cost function and the orthogonal least squares algorithm[J].Neural Networks,2002,15(2):263-270
[36]L H Chiang,E L Russell,R D Braatz.Fault diagnosis in chemical processes using Fisher discriminant analysis,discriminant partial least squares,and principal component analysis[J].Chemometrics and Intelligent Laboratory Systems,2000,50(2):243-252
[37]Leo H Chiang,Mark E Kotanchek,Arthur K Kordon.Fault diagnosis based on Fisher discriminant analysis and support vector machines[J].Computers & Chemical Engineering,2004,28(8):1389-1401
[38]C.Angeli.Online expert system for fault diagnosis in hydraulic system[J].Expert System,1999,16,115-120
[39]R.Isermann.Supervision,fault-detection and fault-diagnosis method-Anintroduction[J].Control Engineering Practice,1997,5,639-652
[40]田盛丰.人工智能原理与应用[M].北京:北京理工大学出版社,1993
[41]钟秉林,颜延虎.智能化故障诊断理论与方法的研究现状和展望[J].东南大学学报,1993,增刊,1-11
[42]Dorigo M,Maniezzo V,Colomi A.Positive feedback as a search strategy[D].Italy:Politecnico di Milano,1991
[43]Dorigo M,Bonabeau Eric,Theraulaz Guy.Ant algorithms and stigmergy[J].Future Generation Computer System,2000,16(8):851-871
[44]Dorigo M,Di Caro G,Gambardella L M.Ant algorithms for discrete optimization[J].Artificial Life,1999,5:137-172
[45]张纪会,高齐圣,徐心和.自适应蚁群算法[J].控制理论与应用,2000,17(1):1-3
[46]Gelle G,Colas M.Blind source separation:a tool for rotating machine monitoring by vibrations analysis[J].Journal of Sound and Vibration,2001,248(5):865-885
[47]王海清.PCA过程监测方法的故障检测行为分析[J].化工学报,2002,53(3):297-301
[48]A Sierra.High-order Fisher's discriminant analysis[J].Pattern Recongnition,2002,35(6):1291-1302
[49]盛兆顺,尹畸岭.设备状态监测与故障诊断技术及应用[M].北京:化学工业出版社,2003
[50]谷玉梅.基于虚拟仪器的大型旋转机械机组故障分析及趋势预测技术研究[D].北京:北京机械工业学院,2004
[51]高金吉.机泵群实时监测网络和故障诊断专家系统[J].中国工程科学,2001,3(9):41-47
[52]高金吉.高速涡轮机械振动故障机理及诊断方法的研究[D].北京:清华大学力学工程系,1993
[53]萨师煊,王珊.数据库系统概论[M].北京:高等教育出版社,2000
[54]Z Pawlak.Rough set theory and its applications to data analysis[J].Cybemetics and Systems:An International Journal,1998,29:661-688
[55]Z Pawlak,S K M Wong,W Ziarko.Rough sets:probabilistic versus deterministic approach[J].International Journal of Man-Machine Studies,1988,29:81-95
[56]陈杰,蒋祖华,赵云松.基于扩展的信息熵的决策表属性约简算法[J].计算机工程与应用,2007,43(7):167-169
[57]许宝杰,张建民,徐小力,等.粗糙集合理论在旋转机械故障诊断技术中的应用[J].机械设计与制造,2006,12:95-97
[58]阮跃.振动故障诊断系统中的征兆获取[J].热能动力工程,1999,14(83):382-384
[59]杨善林,倪志伟.机器学习与智能决策支持系统[M].北京:科学出版社,2004
[60]何勇,方红芳.模糊诊断技术在旋转机械设备故障诊断中的应用[J].中国纺织大学学报,1998,24(3):16-18
[61]吴振生,杨安国,贾军峰.基于模糊多属性决策的旋转机械故障自动诊断系统[J].石油矿场机械,2006,35(5):4-8
