摘要
人工神经网络已越来越多地被用于故障诊断领域。论文以感应电机为研究对象,研究将人工神经网络应用于感应电机故障诊断中的一系列问题。
在电机正常、转子断条、匝间短路、复合故障(转子断条和匝间短路情况下)四种不同工作状态下,对电机定子电流和振动信号进行采样,从采集的信号中,提取振动幅值信号的均方根值作为特征参数之一;应用“小波包-能量”法提取电流信号中的有效特征参数,将时域参数和频域参数相结合作为人工神经网络的输入矢量,对网络进行训练和测试。
传统的神经网络算法应用于故障诊断时,具有易陷入局部极小值,收敛速度较慢等缺点。而生物免疫系统的一些诱人特性为人们开发新的智能算法提供了新的途径。本文将免疫进化学习算法和人工神经网络相结合设计了一个免疫神经网络。利用免疫进化学习算法对神经网络的隐层中心进行聚类,得到隐层中心值后,利用四种故障模式下的特征样本对网络进行训练和测试。
利用小波网络精度高,学习速度快的特点,本文将小波理论与神经网络技术相结合。用小波函数代替传统神经网络中的激活函数,推导了小波神经网络算法,并对小波神经网络进行了加动量项的改进,改进后的算法能够有效地加快神经网络的收敛速度,避免陷入局部极小值。
对以上算法进行了大量的仿真研究,仿真结果表明:利用时域方法和小波包-能量法提取特征参数是有效的;使用免疫进化学习方法与RBF神经网络相结合,训练后的网络可以对感应电机故障作出较为准确的诊断。仿真结果还表明:改进的小波网络算法用于电机故障诊断是有效的。将小波网络与BP网络仿真结果进行比较,结果表明该算法提高了处理速度,精度较高。
Artificial neural networks are increasingly used in the areas of fault diagnosis.In this thesis, the research comes from the neural network,regarding induction motor as a target,and the research will use artificial neural network to solve series of induction motor fault diagnosis problems.
Under the four models of composite fault (broken bar and short-circuit fault)、short-circuit fault、broken bar、normal operation.the vibration and current signals are sampled, From the collection of signals,pick up the amplitude of vibration from the motor,use rms value as one of the characteristic parameters and use"wavelet packet - energy" extract useful parameters of current signal. Combine the time domain parameters and the frequency parameters to be the input vectors of the artificial neural network.then train and test the network.
Traditional neural network algorithms are easy to fall into the local minimum、slow convergence when in fault diagnosis.And the biological immune system has some attractive properties,they provide an important inspiration for people to develop new intelligent algorithm. Therefore,based on the advantages of immune algorithm,the paper combined immune evolutionary learning algorithms and artificial neural networks.use arithmetic immune neural network algorithm to cluster the hidden layer, after get the center of hidden layer, the further use of immune neural network to train four motor fault modes under teachers.
Because wavelet neural network have the following traits:high precision,learning rate fast etc.so in this thesis, we combined wavelet theory and neural network,using wavelet function instead of the activation function of traditional neural networks,put forward wavelet neural network,and improve it.the improved algorithm can effectively accelerate the convergence rate of neural networks, avoid the local minimum.
Series of simulating tests are conducted by algorithm in this theis,the results indicate that the effective of extracting characteristic parameters by using time domain wavelet packet - energy.combined immune evolutionary learning algorithm and RBF,after the train.the network have accurate fault diagnosis of induction motor breakdown. In addition,the results indicate that the effective of using improved wavelet network algorithm for fault motor diagnosis,compared wavelet network with BP Network,the results show that the algorithm increased the processing speed and have high accuracy.
引文
[1] Subhasis Nandi and Humid A.Tuliyat.Fault Diagnosis of Electrical Machines-a Review.Proceedings From Electric Machines and Drives,1999:219～221
[2]周东华,叶银忠.现代故障诊断与容错控制.北京:清华大学出版社,2000
[3] William T,Thomson, Fenger M. Current Signature Analysis to Detect Induction Motor Faults. IEEE Trans. on Industry Applications.2001.12(8): 26~34
[4]沈标正.电机故障诊断技术.北京:机械工业出版社,1996
[5]何仰赞,温增根,汪馥瑛等.电力系统分析.武汉:华中理工大学出版社,1995
[6]宁玉泉.笼型感应电动机转子故障时的参数计算.电工技术学报,2002.17(4):84~88
[7] Olaf Moseler and Rolf Isermann.Model-based Fault Detection to Brushless DC Motor using Parameter Estimation.Proceedings From IEEE Industrial Electronics Society,1998.4:1956 ~1960
[8] Olaf Moseler and Rolf Isermann.Application of model-based Fault Detection to a Brushless DC Moter.IEEE Trans.on Industrial Electronics,2000.47(5):1015~1020
[9]郭玉英,朱正为,靳玉红.基于等价空间算法的飞控系统故障诊断.控制与决策,2007.Vol.22(2):51
[10]邱阿瑞.用起动电流的时变频谱诊断鼠笼异步电动机转子故障.中国电机工程学报,1995.Vol.15(4): 267~273
[11]安永红,夏昌浩.笼型异步电动机转子断条故障诊断技术.电力学报,2006.vol.21 (3):312
[12] Deleroi W.Broken bar in squirrel cage rotor of an induction motor,Partl,Descriptionby superimposed fault currents.Arch.fur Elektrotechnik,1984.67(2):89~98
[13]朱惠莲.神经网络在发动机燃油系统故障诊断中的研究.计算机技术应用,2006(3):20
[14]赵月明,陈在平.基于遗传-神经网络的电机故障诊断.天津理工大学学报,2006.Vol.22 (5):42~44
[15]宇缨,李清华.统计学习理论和支持向量机.沈阳大学学报,2005.17(4):42~46
[16] VapnikV.The Nature of Statistical Learning Theory .N.Y.Springer,1998
[17]时文刚,刘树林,张嘉钟等.基于支持向量机的往复泵泵阀故障诊断.机械强度,2002.24(3):362~364
[18] Tavner P. J., Penman J. Condition monitoring of electrical machines.England:Research Studies Press Ltd.Letch worth Hertfordshire,1987
[19]鄢玉.电机故障诊断及小波基函数的选择:[硕士学位论文].太原:太原理工大学,2006.5
[20]邱阿瑞.提取感应电动机转子故障特征的新方法.清华大学学报(自然科学版),1997.37(1):35~37
[21]沈标正.电机故障诊断技术.北京:机械工业出版社,1996
[22]田慕玲.基于小波分析的异步电机故障诊断:[硕士学位论文].太原:太原理工大学,2006
[23]蔚兰.基于小波分析的异步电动机故障诊断的研究:[硕士学位论文].太原:太原理工大学,2005
[24]袁科新.基于神经网络的发动机故障诊断研究:[硕士学位论文].山东:山东大学,2006
[25]叶世伟,史忠植译.神经网络原理(Simon Haykin:Neural Networks).机械工业出版社,2004
[26] Mcculloch W S,Pittsw.A Logical Calculus of The Ideas Immanent in Nervous Activity.Bulletin of Mathematical Biophysics,1943.(5):115~133
[27]廖晓峰,李传东.神经网络研究的发展趋势.发展方向,2005.43~44
[28]王磊,潘进,焦李成.免疫算法.电子学报,2002.28(7):74-78
[29] Chun J S,Kim M K,Jung H k.Shape Optimization of Electromagnetic Devices using Immune Algorithm.IEEE Transactions on Magnetics,1997.33(2):1876~1879
[30] Forrest S,Perelson A,Sllen L et al.Self-Nonself discrimination in a compute.proceedings of IEEE symposium on research Security and Privacy,Oakland:Public by IEEE.1994.202~212
[31]刘树林,王日新等.基于反面选择算法的压缩机振动在线监测方法研究.压缩机技术,2001.Vol.5:9~11
[32]杨江云,李蓓智等.人工免疫机理在故障诊断中的应用.东华大学学报,2004.vol.30(1):41~44
[33]王磊.免疫进化计算理论及应用:[博士研究生学位论文].西安:西安电子科技大学,2001
[34]张春喜,刘海丽,吕志香等.感应电机转子断条故障诊断系统的研究.黑龙江水利学报,2005.Vol. 32(4):23
[35] Luis Alberto Pereira,Daniel da Silva Gazzana.Rotor Broken Bar Detection and Diagnosis in Induction Motors Using Stator Current Signature Analysis and Fuzzy Logic.The 30th Annum Conference of the IEEE Industrial Electronics Society,Busan,Korea.2004:3019~3023
[36] Thomson W T.On-line MCSA to diagnose shorted turns in low voltage stator windings of 3-phase induction motors prior to failure.IEEE IMDCO1,2001:891~898
[37]侯新国,吴正国,夏立等.基于融合分析的感应电机轴承故障检测方法.数据采集与处理,2006.Vol. 21(1)
[38] Schoen R R,Habetler T G,Kamran F,et a1.Motor bearing damage detection using stator current monitoring.IEEE Trans on Industry Applications,1995.31(6):1274～1279
[39]孙立军,吕志象,刘海丽等.基于DSO-2100的异步电机转子偏心故障检测与诊断.电机与控制学报,2005.Vol.9(6):538~539
[40]王凤昌,李面换.异步电机故障的电流监测和诊断.哈尔滨工业大学学报,1994.26(6):88~93
[41]杨福生.小波变换的工程分析与应用.北京,科学出版社,1999
[42]胡昌华,张军波,夏军等.基于MATLAB的系统分析与设计-小波分析.西安:西安电子科技大学出版社,1999
[43]崔锦泰,程正兴译.小波分析导论.西安:西安交通大学出版社,1995
[44]程止兴.小波分析算法与应用.西安:西安交通大学出版社,1998
[45]李建平等.小波分析与信号处理理论、应用及软件实现.重庆:重庆出版社,1997.46
[46]魏海坤.神经网络结构设计的理论与方法.国防工业出版社.2005.7
[47]陈鹏.基于神经网络的滚动轴承故障诊断方法研究与应用:[硕士学位论文].成都:四川大学,2006
[48]王承.基于神经网络的模拟电路故障诊断方法研究:[博士学位论文].成都:电子科技大学,2005
[49]刘树林,黄文虎,夏松波,丛蕊.活塞压缩机气阀故障检测的免疫神经网络方法研究.中国机械工程,2005.vol.16(15):1384-1387
[50]唐斌,胡光锐.基于免疫神经网络的雷达天线扫描方式的识别.应用科学学,2003.Vo1.21(1):3738
[51]唐斌,胡光锐.免疫神经网络用于侦察测向的技术.上海交通大学学报,2002.Vo1.36(12):1851~ 1853
[52]郑建国,刘芳,焦李成.用免疫神经网络做数据挖掘.计算机学报,2002.Vol.29(10):51~55
[53]侯家利,朱梅阶,彭宏.模块化免疫神经网络的模型研究.电子学报,2005.Vol.33(8):1502~1505
[54]闻新,周露等.MATLAB神经网络仿真与应用.科学出版社,2003.7:285~287
[55]吴月伟,杨慧.基于RBF神经网络的飞机发动机故障诊断研究.2005.vol.31:266
[56]徐章遂,房立清,王希武等.故障信息诊断原理及应用.国防工业出版社, 2000.7:11~32,104~184
[57] Kohonen T.Self-organization and Assoricative Memory.Springer-Verlag,Berlin.1984
[58] Timmis.J.Neal M.A resouse limited artificial immune system for data analysis.Knowledge-Base- Systems,2001.(14):121~132
[59] arakanov S,Dasgupta D.A formal model of an artificial immune system . Biosystems,2000.55
[60]莫宏伟.人工免疫系统原理及应用.哈尔滨工业大学出版社,2002
[61]刘树林,黄文虎,夏松波,丛蕊.活塞压缩机气阀故障检测的免疫神经网络方法研究.中国机械工程, 2005.vol.16(15):1384~1387
[62]陈强,郑德玲.一种基于人工免疫的数据模式进化学习模型及其应用研究.计算机工程与应用,2005.20:40~43
[63]邹鲤,袁俊泉,龚享铱.MATLAB6.x信号处理.北京:清华大学出版社,2002
[64]飞思科技产品研发中心.MATLAB6.5辅助小波分析与应用.北京:电子工业出版社,2003
[65] Szu H.Neural network adaptive wavelets neural for signal representation and classification.Optical Engineering,1992.31(9):1907~1916
[66]张贤达.现代信号处理.北京:清华大学出版社,1995,1~300
[67] TangTH,LiuYJ,LiJR et aI.ANN-based fault diagnosis method with a combined BP algorithm.In:Proc.of UKACC Int.Conf.On Control,1996.2:861~866
[68]吕伯权,李天择等.一种用于函数学习的小波神经网络.自动化学报,1998.24(2):548~551
[69] Dexian Huang,Yihui Jin.The Application of Wavelet Neural Networks to Nonlinear Predictive Control. Proceedings of the 1997 IEEE International Conference on Neural Networks,Part 2(of 4).1997.723~728
[70] Zhang Qinghua,Benvenist A.Wavelt networks.IEEE transon NN,1992.3(6):889~898
[71]郑建斌,张君等.小波变换与神经网络相结合用于示波计时电位测定.北京大学学报(自然科学出版社),2001.Vo1.3 (7).No.2:167~172
[72]张增芳等.基于小波神经网络的心电数据压缩研究.广西工学院学报,2002.12~15
[73]仲京臣.基于小波神经网络的故障诊断研究:[硕士学位论文].青岛:中国海洋大学,2004
[74] Zhang Jun,etal.Wavelet neural networks for function Iearning.IEEE transom SP,1995.43(6):1485~1497
[75] Nielsen R H.Theory of the back propagation neural network.In:Proc.of IJCNN,Vol.I.1989
[76] Kolmogorov.On the of representation of continuous functions of many variables by continuous functions of one variable and condition.Dokl.Akad.Nauk.USSR,1957.114:953~956