欢迎来到天天文库
浏览记录
ID:57974900
大小:332.16 KB
页数:5页
时间:2020-04-18
《基于多维振动特征的滚动轴承故障诊断方法-论文.pdf》由会员上传分享,免费在线阅读,更多相关内容在行业资料-天天文库。
1、第34卷第3期噪声与振动控制V_0l34No.32014年6月N0ISEANDVIBRAT10NC0NTROLJn.2014文章编号:1006—1355(2014)03—0165—05基于多维振动特征的滚动轴承故障诊断方法付云骁,贾利民,季常煦,姚德臣,李文球(1.北京交通大学轨道交通控制与安全国家重点实验室,北京100044;2.北京交通大学电气工程学院,北京100044;3.广州地下铁道总公司,广州510030)摘要:单独提取滚动轴承振动信号的时域或频域特征进行故障诊断,是目前常用的轴承诊断方法,诊断精度有待提高。以时域和频域的多维振动特征参量为指标,以历史诊断正确率作为特
2、征参量权值,分别对滚动轴承的无故障和经常出现的滚珠故障、内环故障和外环故障工况进行特征提取和故障识别。多维时频域振动特征是单维特征依据诊断精度权重的集合。运用BP神经网络分别对信号的时域特征(TDF)、IMF能量矩(IEM)、小波包能量矩(WPEM),以及多维时频域特征进行智能故障判别。实验验证用多维时频域振动特征参量综合诊断的方法进行滚动轴承故障诊断,比单维特征的诊断结果精确且效率较高,该方法可以在滚动轴承故障诊断领域展开应用。关键词:振动与波;多维特征;BP神经网络;故障诊断:滚动轴承中图分类号:TB53文献标识码:ADOI编码:10.3969~.issn.1006.133
3、5.2014.03.035FaultDiagnosisMethodofRollingBearingsBasedonMulti.dimensional6rationFeatures,Yun-xiao1,2,JIALi-min,刀Chang-xu,YAODe—chen,Wen—qiu(1.StateKeyLaboratoryofRailTramcControlandSafety,BeijingJiaotongUniversity,Beijing100044,China;2.SchoolofElectricalEngineering,BeijingJiaotongUniversity
4、,Beijing100044,China;3.GuangzhouUndergroundHeadOfice,Guangzhou510030,China)AbaUact:Extractingthetime-domainorthefrequency-domainfeaturesofvibrationsignalsforanalysisisaconventionalmethodforrollingbeatingsfaultdiagnosis.Buttheefectsofthisdiagnosismethodneedtobeimproved.Inthispaper,takingthemu
5、lti-·dimensionalvibrationcharacteristicparametersintime-domainandfrequency·-domainastheindexesandtheCOlTectnessrateofhistoricaldiagnosisastheparame~cweight,thefeaturesoffault-freerollingbearingsandthefeaturesofrollingbearingswithballfault,innerandouterracefaultsareextractedandthefaultsareide
6、ntified.Itshowsthatthemulti-dimensionalvibrationcharacteristicintime-frequencydomainsistheassemblageofsinglefeatures.BPneuralnetworkisusedforintelligentfaultclassificationofsignalsaccordingtothetime-domainfeature(TDF)parameters,IMFenergymoment(IEM),waveletpackageenergymoment(WPEM)andmulti—di
7、mensionalfeaturesrespectively.Resultsofthediagnosesarecomparedoneanother.Theexperimentresultsverifythatusingthemulti-dimensionalfeatureintimeandfrequencydomainstoevaluatetherollingbearingfaultsisaccurateandeficient.ThismethodCanbeappliedinthefieldo
此文档下载收益归作者所有