基于EMD和小波包的軸承故障特征提取方法研究.pdf

1、沈陽(yáng)航空工業(yè)學(xué)院碩士學(xué)位論文基于EMD和小波包的軸承故障特征提取方法研究姓名：張毅申請(qǐng)學(xué)位級(jí)別：碩士專業(yè)：信息與信息處理指導(dǎo)教師：林峰20100103沈陽(yáng)航空工業(yè)學(xué)院碩士學(xué)位論文AbstractFirst，afteranalyzingthecharacteristicsandlimitationsofclassicsignalprocessing，thepaperintroducestodaysresearchfocus_——Hilbe

2、rtHuangtransform(HHT)andWavelettransform(WT)，whicharethemainmethodofnonstationarysignalsprocessing，anddeeplystudiedthebasictheoryandalgorithmsSecond，thispaperisfocusedoncombiningempiricalmodedecompositionandwaveletandpoi

3、ntoutthatmostofthiscombinationisusedagoodreductionnoiseofwavelettodopre—processingSincetheactualmotorvibrationsignalsareverycomplexandmixsomenoise，itisnecessarytofindanoisereductionmethodHoweveritisdifficulttogetunifiedn

4、oisereductionmodel，andSOitisverydifficulttoextractfaultfeatureAfterstudyingthetheoreticalofEMDfoundthatthedecompositionprocessisnotonlyselfadaptivebutalsoadaptivefilteringTherefore，thepapercombinesadaptivefilteringofEMDa

5、ndwaveletpacketdecomposition(thepaperisknownasEMD_WP)todealwithnonstationarysignalEMDWPisfurtheranalyzingtheIMFsdecomposedbywaveletpacketSimulationresultsshowthatinacaseofnoise，themethodcanstillextractmoreaccuratecharact

6、eristicfrequencyDuringthetrialalsodemonstratesthatwhenthenoiseintensityislarge，withthesecondgenerationwaveletnoisereduction，EMD～WP‘SeffectwillbeimprovedFinallythispaperwithnonstationarybearingfaultsignalsastheresearchobj

7、ectanddescribesthebearingdefectmechanism，byusingsignalprocessingmethodstoextractthefeatureparametersIntheend，theEMD—WPappliedintobearingsfaultfeatureextractionAndtheresultprovesthaitnotonlyhasthehigherfeasibilitybutalsoe