A COMPARISON OF ARTIFICIAL NEURAL NETWORKS AND OTHER STATISTICAL METHODS FOR ROTATING MACHI

A COMPARISON OF ARTIFICIAL NEURAL NETWORKS AND OTHER STATISTICAL METHODS FOR ROTATING MACHI

ID:38212550

大小:80.23 KB

页数:6页

时间:2019-05-27

A COMPARISON OF ARTIFICIAL NEURAL NETWORKS AND OTHER STATISTICAL METHODS FOR ROTATING MACHI_第1页
A COMPARISON OF ARTIFICIAL NEURAL NETWORKS AND OTHER STATISTICAL METHODS FOR ROTATING MACHI_第2页
A COMPARISON OF ARTIFICIAL NEURAL NETWORKS AND OTHER STATISTICAL METHODS FOR ROTATING MACHI_第3页
A COMPARISON OF ARTIFICIAL NEURAL NETWORKS AND OTHER STATISTICAL METHODS FOR ROTATING MACHI_第4页
A COMPARISON OF ARTIFICIAL NEURAL NETWORKS AND OTHER STATISTICAL METHODS FOR ROTATING MACHI_第5页
资源描述:

《A COMPARISON OF ARTIFICIAL NEURAL NETWORKS AND OTHER STATISTICAL METHODS FOR ROTATING MACHI》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、ACOMPARISONOFARTIFICIALNEURALNETWORKSANDOTHERSTATISTICALMETHODSFORROTATINGMACHINECONDITIONCLASSIFICATIONA.C.McCormickandA.K.NandiAbstractStatisticalestimatesofvibrationsignalssuchasthemeanandvariancecanprovideindicationoffaultsinrotatingmachinery.Usingtheseestimatesjoi

2、ntlycangiveamorerobustclassi cationthanusingeachindividually.Arti cialneuralnetworkarchitecturesandsomestatisticalalgorithmsarecomparedwithemphasisontrainingrequirementsandreal-timeimplementationaswellasoverallperformance.IntroductionAnalysisofvibrationscanindicatefaul

3、tconditionsinrotatingmachinery[1]suchasshaftunbalanceorrubbing.Onecommonapproachistoestimatetime-invariantfeaturesfromthevibrationtimeserieswhichchangewhenafaultoccursinthemachine.Thesefeaturescanthenbeinputintosomeformofclassi cationsystemtodecidethemachine'scondition

4、.Arti cialneuralnetworkssuchasmulti-layerperceptrons(MLPs)provideasystemwhichcantheo-reticallyprovideBayesoptimalclassi cationoftheconditionbaseduponmanyfeatures[2].Trainingnetworkscanhowevertakeasubstantiallengthoftimeandisnotguaranteedto ndtheoptimalso-lution.Radialb

5、asisfunction(RBF)neuralnetworksprovideanalternativearchitecturewhichcanbetrainedinamuchshorterperiodoftime.Traditionalstatisticaldiscriminantanalysis[3]algorithmscanbeverysimpletoimplementanddonotrequireatimeconsumingtrainingalgorithm.Butinmanycases,theyrequirecertaina

6、ssumptionstobemadeabouttheinputdata.Iftheseassumptionsarenotvalid,theymaynotprovideasgoodasolutionasaneuralnetwork.ExperimentalSet-UpThevibrationsweremeasuredfromasmallexperimentalmachineset.Thisconsistedofanelectricmotorwhichdroveashaftwitha ywheel.Smallweightscouldbe

7、attachedtothe ywheelunbalancingtheshaftandrubbingcouldbeappliedusingascrewattachedtoaframe.Thevibrationsweremeasuredhorizontallyandverticallyusingaccelerometersattachedtoabearingblock.Thisset-upallowedthecreationoffourmachineconditions:NN-nofaultsapplied;NR-onlytherubf

8、aultisapplied;WN-onlytheunbalancefaultisapplied;WR-bothrubandunbalancefaultsareapplied.TheshaftDepartmentofElectronic

当前文档最多预览五页,下载文档查看全文

此文档下载收益归作者所有

当前文档最多预览五页,下载文档查看全文
温馨提示:
1. 部分包含数学公式或PPT动画的文件,查看预览时可能会显示错乱或异常,文件下载后无此问题,请放心下载。
2. 本文档由用户上传,版权归属用户,天天文库负责整理代发布。如果您对本文档版权有争议请及时联系客服。
3. 下载前请仔细阅读文档内容,确认文档内容符合您的需求后进行下载,若出现内容与标题不符可向本站投诉处理。
4. 下载文档时可能由于网络波动等原因无法下载或下载错误,付费完成后未能成功下载的用户请联系客服处理。