《旋轉(zhuǎn)機(jī)械故障診斷技術(shù)探究》開(kāi)題報(bào)告（有提綱）3700字

PAGEPAGE2旋轉(zhuǎn)機(jī)械故障診斷技術(shù)研究開(kāi)題報(bào)告一、選題的理論意義與實(shí)際意義現(xiàn)代化工業(yè)生產(chǎn)越來(lái)越大型化、高速化、自動(dòng)化，特別是石化、冶金、電力等工程工業(yè)，設(shè)備投資大，連續(xù)生產(chǎn)流程長(zhǎng)，機(jī)械裝備故障停機(jī)可以造成重大經(jīng)濟(jì)損失，甚至導(dǎo)致機(jī)會(huì)人亡的重大事故?，F(xiàn)代生產(chǎn)和某些特種裝備對(duì)人的依賴(lài)程度越來(lái)越低，對(duì)設(shè)備的依賴(lài)程度越來(lái)越高，對(duì)故障的預(yù)示和診治越來(lái)越重要。近二三十年來(lái)國(guó)內(nèi)設(shè)備診斷技術(shù)的研究開(kāi)發(fā)異?；钴S，發(fā)展迅速，在工廠應(yīng)用經(jīng)常取得出人意料的失效，設(shè)備診斷技術(shù)在工廠企業(yè)得到了普及和應(yīng)用。設(shè)備診斷技術(shù)就是掌握設(shè)備的現(xiàn)在狀態(tài)與異?；蚬收现g的關(guān)系，以預(yù)測(cè)未來(lái)的技術(shù)。它包含兩方面的內(nèi)容：一是對(duì)設(shè)備的運(yùn)行進(jìn)行檢測(cè)；二是在發(fā)現(xiàn)異常情況后對(duì)設(shè)備的故障進(jìn)行分析和診斷。二、論文綜述（綜述國(guó)內(nèi)外有關(guān)選題的研究動(dòng)態(tài)）基于旋轉(zhuǎn)機(jī)械振動(dòng)信號(hào)的預(yù)警技術(shù)是當(dāng)前企業(yè)應(yīng)用最為廣泛的旋轉(zhuǎn)機(jī)械在線早期故障預(yù)警技術(shù)ADDINEN.CITE<EndNote><Cite><Author>Soleimani</Author><Year>2015</Year><RecNum>182</RecNum><DisplayText><styleface="superscript">[23]</style></DisplayText><record><rec-number>182</rec-number><foreign-keys><keyapp="EN"db-id="2wsstwe98t9dr3ewt5v5055otwdaw5etw90v"timestamp="1645793123">182</key></foreign-keys><ref-typename="JournalArticle">17</ref-type><contributors><authors><author>Soleimani,A.</author><author>Khadem,S.E.%JChaosSolitons</author><author>Fractals</author></authors></contributors><titles><title>Earlyfaultdetectionofrotatingmachinerythroughchaoticvibrationfeatureextractionofexperimentaldatasets</title></titles><pages>61-75</pages><volume>78</volume><dates><year>2015</year></dates><urls></urls></record></Cite></EndNote>[1]。學(xué)者們?cè)诜治鲂D(zhuǎn)機(jī)械振動(dòng)信號(hào)的基礎(chǔ)上研發(fā)出許多早期故障預(yù)警方法，這些方法大致可以分為機(jī)理模型驅(qū)動(dòng)法和數(shù)據(jù)驅(qū)動(dòng)法ADDINEN.CITE<EndNote><Cite><Author>Xu</Author><Year>2020</Year><RecNum>1238</RecNum><DisplayText><styleface="superscript">[24,25]</style></DisplayText><record><rec-number>1238</rec-number><foreign-keys><keyapp="EN"db-id="2wsstwe98t9dr3ewt5v5055otwdaw5etw90v"timestamp="1646631775">1238</key></foreign-keys><ref-typename="JournalArticle">17</ref-type><contributors><authors><author>Xu,X.</author><author>Tao,Z.</author><author>Ming,W.</author><author>An,Q.</author><author>Chen,M.%JMeasurement</author></authors></contributors><titles><title>Intelligentmonitoringanddiagnosticsusinganovelintegratedmodelbasedondeeplearningandmulti-sensorfeaturefusion</title></titles><pages>108086</pages><volume>165</volume><dates><year>2020</year></dates><urls></urls></record></Cite><Cite><Author>Shi</Author><Year>2020</Year><RecNum>184</RecNum><record><rec-number>184</rec-number><foreign-keys><keyapp="EN"db-id="2wsstwe98t9dr3ewt5v5055otwdaw5etw90v"timestamp="1645795233">184</key></foreign-keys><ref-typename="JournalArticle">17</ref-type><contributors><authors><author>Shi,Huaitao</author><author>Guo,Jin</author><author>Bai,Xiaotian</author><author>Guo,Lei</author><author>Liu,Zhenpeng</author><author>Sun,Jie</author></authors></contributors><titles><title>ResearchonaNonlinearDynamicIncipientFaultDetectionMethodforRollingBearings</title><secondary-title>AppliedSciences-Basel</secondary-title></titles><periodical><full-title>AppliedSciences-Basel</full-title></periodical><volume>10</volume><number>7</number><dates><year>2020</year><pub-dates><date>Apr</date></pub-dates></dates><accession-num>WOS:000533356200242</accession-num><urls><related-urls><url><GotoISI>://WOS:000533356200242</url></related-urls></urls><custom7>2443</custom7><electronic-resource-num>10.3390/app10072443</electronic-resource-num></record></Cite></EndNote>[2,3]?；跈C(jī)理模型驅(qū)動(dòng)的早期故障檢測(cè)或故障模式識(shí)別的相關(guān)成果有：唐貴基等人通過(guò)最大相關(guān)峭度解卷積提取滾動(dòng)軸承早期微弱故障特征ADDINEN.CITEADDINEN.CITE.DATA[4,5]；王宏超等人通過(guò)最小熵解卷積對(duì)軸承的振動(dòng)原始時(shí)域波形數(shù)據(jù)進(jìn)行預(yù)處理后，再利用稀疏分解等方法識(shí)別軸承早期故障特征頻率ADDINEN.CITE<EndNote><Cite><Author>王宏超</Author><Year>2013</Year><RecNum>189</RecNum><DisplayText><styleface="superscript">[28]</style></DisplayText><record><rec-number>189</rec-number><foreign-keys><keyapp="EN"db-id="2wsstwe98t9dr3ewt5v5055otwdaw5etw90v"timestamp="1645796832">189</key></foreign-keys><ref-typename="JournalArticle">17</ref-type><contributors><authors><author>王宏超</author><author>陳進(jìn)</author><author>董廣明</author></authors></contributors><auth-address>上海交通大學(xué)機(jī)械系統(tǒng)與振動(dòng)國(guó)家重點(diǎn)實(shí)驗(yàn)室;</auth-address><titles><title>基于最小熵解卷積與稀疏分解的滾動(dòng)軸承微弱故障特征提取%J機(jī)械工程學(xué)報(bào)</title></titles><pages>88-94</pages><volume>49</volume><number>01</number><keywords><keyword>最小熵解卷積</keyword><keyword>稀疏分解</keyword><keyword>滾動(dòng)軸承</keyword><keyword>微弱故障</keyword><keyword>特征提取</keyword></keywords><dates><year>2013</year></dates><isbn>0577-6686</isbn><call-num>11-2187/TH</call-num><urls></urls><remote-database-provider>Cnki</remote-database-provider></record></Cite></EndNote>[6]；楊蕊等人通過(guò)最優(yōu)小波尺度循環(huán)譜提取振動(dòng)信號(hào)的早期故障微弱故障特征頻率ADDINEN.CITE<EndNote><Cite><Author>楊蕊</Author><Year>2018</Year><RecNum>188</RecNum><DisplayText><styleface="superscript">[29]</style></DisplayText><record><rec-number>188</rec-number><foreign-keys><keyapp="EN"db-id="2wsstwe98t9dr3ewt5v5055otwdaw5etw90v"timestamp="1645796773">188</key></foreign-keys><ref-typename="JournalArticle">17</ref-type><contributors><authors><author>楊蕊</author><author>李宏坤</author><author>賀長(zhǎng)波</author><author>王奉濤</author></authors></contributors><auth-address>大連理工大學(xué)機(jī)械工程學(xué)院;</auth-address><titles><title>利用最優(yōu)小波尺度循環(huán)譜的滾動(dòng)軸承早期故障特征提取%J機(jī)械工程學(xué)報(bào)</title></titles><pages>208-217</pages><volume>54</volume><number>17</number><keywords><keyword>循環(huán)周期譜</keyword><keyword>連續(xù)小波變換</keyword><keyword>相關(guān)峭度</keyword><keyword>最優(yōu)小波尺度循環(huán)譜</keyword><keyword>滾動(dòng)軸承</keyword></keywords><dates><year>2018</year></dates><isbn>0577-6686</isbn><call-num>11-2187/TH</call-num><urls><related-urls><url>/kcms/detail/11.2187.TH.20170822.1646.002.html</url></related-urls></urls><remote-database-provider>Cnki</remote-database-provider></record></Cite></EndNote>[7]；Fan等人通過(guò)變分模態(tài)分解（VariationalModeDecomposition，VMD）、經(jīng)驗(yàn)?zāi)B(tài)分解（EmpiricalModeDecomposition，EMD）及其改進(jìn)型處理旋轉(zhuǎn)機(jī)械振動(dòng)信號(hào)后，得到故障特征頻率更加清晰的旋轉(zhuǎn)機(jī)械頻譜，從而實(shí)現(xiàn)在線早期故障檢測(cè)ADDINEN.CITEADDINEN.CITE.DATA[8,8]；李宏坤等人提出對(duì)振動(dòng)信號(hào)進(jìn)行連續(xù)小波分解后再進(jìn)行重構(gòu)的方法對(duì)旋轉(zhuǎn)機(jī)械早期微弱的故障特征頻率進(jìn)行識(shí)別，進(jìn)而辨識(shí)所發(fā)生早期故障的模式ADDINEN.CITE<EndNote><Cite><Author>李宏坤</Author><Year>2014</Year><RecNum>192</RecNum><DisplayText><styleface="superscript">[32]</style></DisplayText><record><rec-number>192</rec-number><foreign-keys><keyapp="EN"db-id="2wsstwe98t9dr3ewt5v5055otwdaw5etw90v"timestamp="1645797115">192</key></foreign-keys><ref-typename="JournalArticle">17</ref-type><contributors><authors><author>李宏坤</author><author>劉洪軼</author><author>徐福健</author><author>張曉雯</author><author>張學(xué)峰</author></authors></contributors><auth-address>大連理工大學(xué)機(jī)械工程學(xué)院;</auth-address><titles><title>連續(xù)小波最優(yōu)重構(gòu)尺度確定方法與故障早期識(shí)別%J機(jī)械工程學(xué)報(bào)</title></titles><pages>69-76</pages><volume>50</volume><number>17</number><keywords><keyword>連續(xù)小波變換</keyword><keyword>信號(hào)重構(gòu)</keyword><keyword>小波熵</keyword><keyword>早期故障識(shí)別</keyword></keywords><dates><year>2014</year></dates><isbn>0577-6686</isbn><call-num>11-2187/TH</call-num><urls><related-urls><url>/kcms/detail/11.2187.th.20140528.1023.085.html</url></related-urls></urls><remote-database-provider>Cnki</remote-database-provider></record></Cite></EndNote>[10]；張?jiān)茝?qiáng)等人基于廣義S變換和傅里葉逆變換的基礎(chǔ)上，提出一種雙時(shí)域變換，并用于增強(qiáng)時(shí)域振動(dòng)信號(hào)，提高軸承早期故障的識(shí)別準(zhǔn)確度ADDINEN.CITE<EndNote><Cite><Author>張?jiān)茝?qiáng)</Author><Year>2016</Year><RecNum>197</RecNum><DisplayText><styleface="superscript">[33]</style></DisplayText><record><rec-number>197</rec-number><foreign-keys><keyapp="EN"db-id="2wsstwe98t9dr3ewt5v5055otwdaw5etw90v"timestamp="1645797360">197</key></foreign-keys><ref-typename="JournalArticle">17</ref-type><contributors><authors><author>張?jiān)茝?qiáng)</author><author>張培林</author><author>王懷光</author><author>吳定海</author></authors></contributors><auth-address>軍械工程學(xué)院車(chē)輛與電氣工程系;</auth-address><titles><title>基于雙時(shí)域微弱故障特征增強(qiáng)的軸承早期故障智能識(shí)別%J機(jī)械工程學(xué)報(bào)</title></titles><pages>96-103</pages><volume>52</volume><number>21</number><keywords><keyword>滾動(dòng)軸承</keyword><keyword>早期故障診斷</keyword><keyword>雙時(shí)域變換</keyword><keyword>脈沖耦合神經(jīng)網(wǎng)路</keyword></keywords><dates><year>2016</year></dates><isbn>0577-6686</isbn><call-num>11-2187/TH</call-num><urls><related-urls><url>/kcms/detail/11.2187.th.20160420.1508.060.html</url></related-urls></urls><remote-database-provider>Cnki</remote-database-provider></record></Cite></EndNote>[11]；Jose等人引入了機(jī)器的自由體旋轉(zhuǎn)自由度，提出一個(gè)綜合考慮滾動(dòng)阻力現(xiàn)象的深溝球軸承模型，實(shí)現(xiàn)了非靜止調(diào)價(jià)下的旋轉(zhuǎn)機(jī)械角域建模ADDINEN.CITE<EndNote><Cite><Author>Gomez</Author><Year>2019</Year><RecNum>108</RecNum><DisplayText><styleface="superscript">[34]</style></DisplayText><record><rec-number>108</rec-number><foreign-keys><keyapp="EN"db-id="2wsstwe98t9dr3ewt5v5055otwdaw5etw90v"timestamp="1603348157">108</key><keyapp="ENWeb"db-id="">0</key></foreign-keys><ref-typename="JournalArticle">17</ref-type><contributors><authors><author>Gomez,JoseL.</author><author>Khelf,Ilyes</author><author>Bourdon,Adeline</author><author>André,Hugo</author><author>Rémond,Didier</author></authors></contributors><titles><title>Angularmodelingofarotatingmachineinnon-stationaryconditions:Applicationtomonitoringbearingdefectsofwindturbineswithinstantaneousangularspeed</title><secondary-title>MechanismandMachineTheory</secondary-title></titles><periodical><full-title>MechanismandMachineTheory</full-title></periodical><pages>27-51</pages><volume>136</volume><section>27</section><dates><year>2019</year></dates><isbn>0094114X</isbn><urls></urls><electronic-resource-num>10.1016/j.mechmachtheory.2019.01.028</electronic-resource-num></record></Cite></EndNote>[12]；Shi等人提出一種基于動(dòng)態(tài)模型的非均勻載荷滾動(dòng)軸承在線狀態(tài)監(jiān)測(cè)方法，該方法通過(guò)信號(hào)采集、軸承動(dòng)態(tài)模型建立以及狀態(tài)識(shí)別來(lái)實(shí)現(xiàn)對(duì)非均勻載荷狀態(tài)軸承的在線監(jiān)測(cè)ADDINEN.CITE<EndNote><Cite><Author>Shi</Author><Year>2020</Year><RecNum>109</RecNum><DisplayText><styleface="superscript">[35]</style></DisplayText><record><rec-number>109</rec-number><foreign-keys><keyapp="EN"db-id="2wsstwe98t9dr3ewt5v5055otwdaw5etw90v"timestamp="1603348231">109</key><keyapp="ENWeb"db-id="">0</key></foreign-keys><ref-typename="JournalArticle">17</ref-type><contributors><authors><author>Shi,H.T.</author><author>Bai,X.T.</author></authors></contributors><titles><title>Model-basedunevenloadingconditionmonitoringoffullceramicballbearingsinstarvedlubrication</title><secondary-title>MechanicalSystemsandSignalProcessing</secondary-title></titles><periodical><full-title>MechanicalSystemsandSignalProcessing</full-title></periodical><volume>139</volume><section>106583</section><dates><year>2020</year></dates><isbn>08883270</isbn><urls></urls><electronic-resource-num>10.1016/j.ymssp.2019.106583</electronic-resource-num></record></Cite></EndNote>[13]；Li等人提出一種基于改進(jìn)小波尺度圖方法來(lái)識(shí)別軸承的早期微弱沖擊特征，該方法揭示了時(shí)頻分布與特征信息之間的關(guān)系，且實(shí)驗(yàn)驗(yàn)證結(jié)果表明此方法能夠有效的識(shí)別齒輪箱軸承的早期微弱故障ADDINEN.CITE<EndNote><Cite><Author>Li</Author><Year>2018</Year><RecNum>110</RecNum><DisplayText><styleface="superscript">[36]</style></DisplayText><record><rec-number>110</rec-number><foreign-keys><keyapp="EN"db-id="2wsstwe98t9dr3ewt5v5055otwdaw5etw90v"timestamp="1603349876">110</key><keyapp="ENWeb"db-id="">0</key></foreign-keys><ref-typename="JournalArticle">17</ref-type><contributors><authors><author>Li,Hongkun</author><author>Yang,Rui</author><author>Wang,Chaoge</author><author>He,Changbo</author></authors></contributors><titles><title>InvestigationonPlanetaryBearingWeakFaultDiagnosisBasedonaFaultModelandImprovedWaveletRidge</title><secondary-title>Energies</secondary-title></titles><periodical><full-title>Energies</full-title><abbr-1>Energies</abbr-1></periodical><volume>11</volume><number>5</number><section>1286</section><dates><year>2018</year></dates><isbn>1996-1073</isbn><urls></urls><electronic-resource-num>10.3390/en11051286</electronic-resource-num></record></Cite></EndNote>[14]。上述早期故障檢測(cè)、早期故障模式識(shí)別方法依賴(lài)于旋轉(zhuǎn)機(jī)械復(fù)雜的故障機(jī)理、專(zhuān)家經(jīng)驗(yàn)知識(shí)診斷以及信號(hào)濾波降噪等預(yù)處理方法。文獻(xiàn)ADDINEN.CITEADDINEN.CITE.DATA[15-17]基于軸承的先驗(yàn)物理知識(shí)構(gòu)建軸承狀態(tài)監(jiān)測(cè)模型以實(shí)現(xiàn)早期故障檢測(cè)，該方法需要建立一個(gè)精確的滾動(dòng)軸承物理模型，但滾動(dòng)軸承的內(nèi)應(yīng)力等物理量在運(yùn)動(dòng)過(guò)程中復(fù)雜多變且難以表征ADDINEN.CITE<EndNote><Cite><Author>Shi</Author><Year>2020</Year><RecNum>20</RecNum><DisplayText><styleface="superscript">[37]</style></DisplayText><record><rec-number>20</rec-number><foreign-keys><keyapp="EN"db-id="2wsstwe98t9dr3ewt5v5055otwdaw5etw90v"timestamp="1600515921">20</key><keyapp="ENWeb"db-id="">0</key></foreign-keys><ref-typename="JournalArticle">17</ref-type><contributors><authors><author>Shi,Huaitao</author><author>Guo,Jin</author><author>Bai,Xiaotian</author><author>Guo,Lei</author><author>Liu,Zhenpeng</author><author>Sun,Jie</author></authors></contributors><titles><title>ResearchonaNonlinearDynamicIncipientFaultDetectionMethodforRollingBearings</title><secondary-title>AppliedSciences</secondary-title></titles><periodical><full-title>AppliedSciences</full-title></periodical><volume>10</volume><number>7</number><section>2443</section><dates><year>2020</year></dates><isbn>2076-3417</isbn><urls></urls><electronic-resource-num>10.3390/app10072443</electronic-resource-num></record></Cite></EndNote>[18]，從而導(dǎo)致模型的泛化能力低。數(shù)據(jù)驅(qū)動(dòng)的旋轉(zhuǎn)機(jī)械早期故障預(yù)警技術(shù)的相關(guān)科學(xué)研究成果有：姚鵬川通過(guò)單分類(lèi)支持向量機(jī)（SupportVectorMachine，SVM）、孤立森林、局部異常因子三種算法及時(shí)識(shí)別出核動(dòng)力裝置的早期故障異常狀態(tài)ADDINEN.CITE<EndNote><Cite><Author>姚鵬川</Author><Year>2020</Year><RecNum>221</RecNum><DisplayText><styleface="superscript">[41]</style></DisplayText><record><rec-number>221</rec-number><foreign-keys><keyapp="EN"db-id="2wsstwe98t9dr3ewt5v5055otwdaw5etw90v"timestamp="1646274134">221</key></foreign-keys><ref-typename="JournalArticle">17</ref-type><contributors><authors><author>姚鵬川</author></authors></contributors><auth-address>西安交通大學(xué);</auth-address><titles><title>基于數(shù)據(jù)驅(qū)動(dòng)的核動(dòng)力裝置狀態(tài)監(jiān)測(cè)方法研究%J核動(dòng)力工程</title></titles><pages>135-139</pages><volume>41</volume><number>S1</number><keywords><keyword>異常監(jiān)測(cè)</keyword><keyword>核動(dòng)力裝置</keyword><keyword>支持向量機(jī)</keyword><keyword>局部異常因子</keyword><keyword>孤立森林算法</keyword></keywords><dates><year>2020</year></dates><isbn>0258-0926</isbn><call-num>51-1158/TL</call-num><urls></urls><electronic-resource-num>10.13832/j.jnpe.2020.S1.0135</electronic-resource-num><remote-database-provider>Cnki</remote-database-provider></record></Cite></EndNote>[20]；賈峰等人通過(guò)提取振動(dòng)信號(hào)的多維度排列熵輸入SVM中進(jìn)行旋轉(zhuǎn)機(jī)械早期故障模式識(shí)別ADDINEN.CITE<EndNote><Cite><Author>賈峰</Author><Year>2014</Year><RecNum>190</RecNum><DisplayText><styleface="superscript">[42]</style></DisplayText><record><rec-number>190</rec-number><foreign-keys><keyapp="EN"db-id="2wsstwe98t9dr3ewt5v5055otwdaw5etw90v"timestamp="1645796937">190</key></foreign-keys><ref-typename="JournalArticle">17</ref-type><contributors><authors><author>賈峰</author><author>武兵</author><author>熊曉燕</author><author>熊詩(shī)波</author></authors></contributors><auth-address>太原理工大學(xué)機(jī)械電子工程研究所;太原理工大學(xué)新型傳感器與智能控制教育部重點(diǎn)實(shí)驗(yàn)室;</auth-address><titles><title>基于多維度排列熵與支持向量機(jī)的軸承早期故障診斷方法%J計(jì)算機(jī)集成制造系統(tǒng)</title></titles><pages>2275-2282</pages><volume>20</volume><number>09</number><keywords><keyword>多維度排列熵</keyword><keyword>支持向量機(jī)</keyword><keyword>早期故障診斷</keyword><keyword>滾動(dòng)軸承</keyword></keywords><dates><year>2014</year></dates><isbn>1006-5911</isbn><call-num>11-5946/TP</call-num><urls></urls><electronic-resource-num>10.13196/j.cims.2014.09.024</electronic-resource-num><remote-database-provider>Cnki</remote-database-provider></record></Cite></EndNote>[21]；顧煜炯等人通過(guò)階比重采樣方法獲得風(fēng)機(jī)齒輪箱平穩(wěn)的角域信號(hào)，再將構(gòu)建的無(wú)量綱特征值輸入分類(lèi)器中進(jìn)行早期故障模式識(shí)別ADDINEN.CITEADDINEN.CITE.DATA[22]；孟玲霞等人通過(guò)Gabor重排對(duì)數(shù)時(shí)頻譜的脊線構(gòu)建高維空間特征向量，再經(jīng)過(guò)降維與模式識(shí)別的方法實(shí)現(xiàn)變工況風(fēng)機(jī)關(guān)鍵部件的早期故障模式識(shí)別ADDINEN.CITE<EndNote><Cite><Author>孟玲霞</Author><Year>2017</Year><RecNum>196</RecNum><DisplayText><styleface="superscript">[45]</style></DisplayText><record><rec-number>196</rec-number><foreign-keys><keyapp="EN"db-id="2wsstwe98t9dr3ewt5v5055otwdaw5etw90v"timestamp="1645797309">196</key></foreign-keys><ref-typename="JournalArticle">17</ref-type><contributors><authors><author>孟玲霞</author><author>徐小力</author><author>徐楊梅</author><author>王紅軍</author></authors></contributors><auth-address>北京理工大學(xué)機(jī)械與車(chē)輛學(xué)院;北京信息科技大學(xué)現(xiàn)代測(cè)控技術(shù)教育部重點(diǎn)實(shí)驗(yàn)室;</auth-address><titles><title>變工況時(shí)頻脊流形早期故障預(yù)警方法研究%J北京理工大學(xué)學(xué)報(bào)</title></titles><pages>942-947</pages><volume>37</volume><number>09</number><keywords><keyword>變工況</keyword><keyword>時(shí)頻脊</keyword><keyword>流形學(xué)習(xí)</keyword><keyword>早期故障預(yù)警</keyword></keywords><dates><year>2017</year></dates><isbn>1001-0645</isbn><call-num>11-2596/T</call-num><urls></urls><electronic-resource-num>10.15918/j.tbit1001-0645.2017.09.011</electronic-resource-num><remote-database-provider>Cnki</remote-database-provider></record></Cite></EndNote>[23]；彭成等人提取軸承聲發(fā)射早期微弱故障信號(hào)的多維特征值，再經(jīng)過(guò)降維后輸入增強(qiáng)K近鄰分類(lèi)器中，實(shí)現(xiàn)對(duì)未知數(shù)據(jù)的分類(lèi)檢測(cè)ADDINEN.CITE<EndNote><Cite><Author>彭成</Author><Year>2021</Year><RecNum>198</RecNum><DisplayText><styleface="superscript">[46]</style></DisplayText><record><rec-number>198</rec-number><foreign-keys><keyapp="EN"db-id="2wsstwe98t9dr3ewt5v5055otwdaw5etw90v"timestamp="1645797406">198</key></foreign-keys><ref-typename="JournalArticle">17</ref-type><contributors><authors><author>彭成</author><author>賀婧</author><author>唐朝暉</author><author>陳青</author><author>桂衛(wèi)華</author></authors></contributors><auth-address>湖南工業(yè)大學(xué)計(jì)算機(jī)學(xué)院;中南大學(xué)自動(dòng)化學(xué)院;</auth-address><titles><title>基于雙維度EKNN的滾動(dòng)軸承早期故障分類(lèi)算法%J計(jì)算機(jī)集成制造系統(tǒng)</title></titles><pages>90-101</pages><volume>27</volume><number>01</number><keywords><keyword>聲發(fā)射信號(hào)</keyword><keyword>增強(qiáng)K近鄰分類(lèi)器</keyword><keyword>滾動(dòng)軸承</keyword><keyword>早期故障分類(lèi)</keyword><keyword>故障診斷</keyword></keywords><dates><year>2021</year></dates><isbn>1006-5911</isbn><call-num>11-5946/TP</call-num><urls><related-urls><url>/kcms/detail/11.5946.TP.20200323.1053.002.html</url></related-urls></urls><electronic-resource-num>10.13196/j.cims.2021.01.007</electronic-resource-num><remote-database-provider>Cnki</remote-database-provider></record></Cite></EndNote>[24]；鄭小霞等人使用改進(jìn)VMD提取風(fēng)機(jī)振動(dòng)信號(hào)各分量的高維特征值向量，將其輸入深度置信網(wǎng)絡(luò)中，實(shí)現(xiàn)風(fēng)機(jī)核心部件的故障早期檢測(cè)ADDINEN.CITE<EndNote><Cite><Author>鄭小霞</Author><Year>2019</Year><RecNum>199</RecNum><DisplayText><styleface="superscript">[47]</style></DisplayText><record><rec-number>199</rec-number><foreign-keys><keyapp="EN"db-id="2wsstwe98t9dr3ewt5v5055otwdaw5etw90v"timestamp="1645797709">199</key></foreign-keys><ref-typename="JournalArticle">17</ref-type><contributors><authors><author>鄭小霞</author><author>陳廣寧</author><author>任浩翰</author><author>李東東</author></authors></contributors><auth-address>上海電力大學(xué)自動(dòng)化工程學(xué)院;上海東海風(fēng)力發(fā)電有限公司;</auth-address><titles><title>基于改進(jìn)VMD和深度置信網(wǎng)絡(luò)的風(fēng)機(jī)易損部件故障預(yù)警%J振動(dòng)與沖擊</title></titles><pages>153-160+179</pages><volume>38</volume><number>08</number><keywords><keyword>變分模態(tài)分解</keyword><keyword>多特征提取(VMD)</keyword><keyword>深度置信網(wǎng)絡(luò)(DBN)</keyword><keyword>故障診斷</keyword></keywords><dates><year>2019</year></dates><isbn>1000-3835</isbn><call-num>31-1316/TU</call-num><urls></urls><electronic-resource-num>10.13465/ki.jvs.2019.08.023</electronic-resource-num><remote-database-provider>Cnki</remote-database-provider></record></Cite></EndNote>[25]；雷亞國(guó)等人通過(guò)降噪自動(dòng)編碼器訓(xùn)練深度神經(jīng)網(wǎng)絡(luò)，實(shí)現(xiàn)自適應(yīng)故障特征提取，實(shí)驗(yàn)結(jié)果表明：該方法可以實(shí)現(xiàn)不同工況、不同故障類(lèi)型以及不同故障位置的齒輪早期故障預(yù)警ADDINEN.CITE<EndNote><Cite><Author>雷亞國(guó)</Author><Year>2015</Year><RecNum>200</RecNum><DisplayText><styleface="superscript">[48]</style></DisplayText><record><rec-number>200</rec-number><foreign-keys><keyapp="EN"db-id="2wsstwe98t9dr3ewt5v5055otwdaw5etw90v"timestamp="1645797753">200</key></foreign-keys><ref-typename="JournalArticle">17</ref-type><contributors><authors><author>雷亞國(guó)</author><author>賈峰</author><author>周昕</author><author>林京</author></authors></contributors><auth-address>西安交通大學(xué)機(jī)械制造系統(tǒng)工程國(guó)家重點(diǎn)實(shí)驗(yàn)室;</auth-address><titles><title>基于深度學(xué)習(xí)理論的機(jī)械裝備大數(shù)據(jù)健康監(jiān)測(cè)方法%J機(jī)械工程學(xué)報(bào)</title></titles><pages>49-56</pages><volume>51</volume><number>21</number><keywords><keyword>機(jī)械健康監(jiān)測(cè)</keyword><keyword>深度學(xué)習(xí)理論</keyword><keyword>大數(shù)據(jù)分析</keyword></keywords><dates><year>2015</year></dates><isbn>0577-6686</isbn><call-num>11-2187/TH</call-num><urls><related-urls><url>/kcms/detail/11.2187.th.20150907.1643.066.html</url></related-urls></urls><remote-database-provider>Cnki</remote-database-provider></record></Cite></EndNote>[26]；Li等人從振動(dòng)信號(hào)中提取一種基于規(guī)范變量分析的指標(biāo)，該指標(biāo)綜合考慮了過(guò)去和未來(lái)典型變量之間的差異，通過(guò)引入指數(shù)加權(quán)移動(dòng)平均技術(shù)，提出一種基于皮爾遜相關(guān)分析的旋轉(zhuǎn)機(jī)械早期故障模式識(shí)別新方法ADDINEN.CITE<EndNote><Cite><Author>Li</Author><Year>2019</Year><RecNum>201</RecNum><DisplayText><styleface="superscript">[49]</style></DisplayText><record><rec-number>201</rec-number><foreign-keys><keyapp="EN"db-id="2wsstwe98t9dr3ewt5v5055otwdaw5etw90v"timestamp="1645797793">201</key></foreign-keys><ref-typename="JournalArticle">17</ref-type><contributors><authors><author>Li,X.</author><author>Yang,X.</author><author>Yang,Y.</author><author>Bennett,I.</author><author>Mba,D.%JAppliedSoftComputing</author></authors></contributors><titles><title>Anoveldiagnosticandprognosticframeworkforincipientfaultdetectionandremainingservicelifepredictionwithapplicationtoindustrialrotatingmachines</title></titles><pages>105564</pages><volume>82</volume><dates><year>2019</year></dates><urls></urls></record></Cite></EndNote>[27]；張明等人使用L1TF獲取振動(dòng)數(shù)據(jù)的真實(shí)劣化趨勢(shì)，通過(guò)beta自學(xué)習(xí)構(gòu)建報(bào)警閾值線，實(shí)現(xiàn)了旋轉(zhuǎn)機(jī)械在線早期故障告警ADDINEN.CITE<EndNote><Cite><Author>張明</Author><Year>2014</Year><RecNum>202</RecNum><DisplayText><styleface="superscript">[50]</style></DisplayText><record><rec-number>202</rec-number><foreign-keys><keyapp="EN"db-id="2wsstwe98t9dr3ewt5v5055otwdaw5etw90v"timestamp="1645797833">202</key></foreign-keys><ref-typename="JournalArticle">17</ref-type><contributors><authors><author>張明</author><author>馮坤</author><author>江志農(nóng)%J振動(dòng)與沖擊</author></authors></contributors><titles><title>基于動(dòng)態(tài)自學(xué)習(xí)閾值和趨勢(shì)濾波的機(jī)械故障智能預(yù)警方法</title></titles><pages>7</pages><volume>33</volume><number>24</number><dates><year>2014</year></dates><urls></urls></record></Cite></EndNote>[28]。上述方法通過(guò)構(gòu)建數(shù)據(jù)驅(qū)動(dòng)的旋轉(zhuǎn)機(jī)械性能退化模型來(lái)實(shí)現(xiàn)早期故障檢測(cè)。綜上所述，大部分?jǐn)?shù)據(jù)驅(qū)動(dòng)的旋轉(zhuǎn)機(jī)械早期故障預(yù)警方法依靠先進(jìn)的機(jī)器學(xué)習(xí)技術(shù)，無(wú)需知道旋轉(zhuǎn)機(jī)械性能劣化機(jī)理、無(wú)需依靠專(zhuān)家知識(shí)識(shí)別頻譜故障特征頻率，但需要本設(shè)備或相似設(shè)備的全生命周期數(shù)據(jù)或部分故障數(shù)據(jù)訓(xùn)練數(shù)據(jù)驅(qū)動(dòng)的早期故障預(yù)警模型，部分方法還依賴(lài)于專(zhuān)家經(jīng)驗(yàn)知識(shí)設(shè)置超參數(shù)。此外，部分?jǐn)?shù)據(jù)驅(qū)動(dòng)的早期故障預(yù)警模型將設(shè)備性能退化的趨勢(shì)追蹤功能與故障早期預(yù)警功能混淆，導(dǎo)致模型對(duì)早期故障點(diǎn)不夠敏感，或出現(xiàn)故障發(fā)生后健康指標(biāo)波動(dòng)較大，甚至回到正常狀態(tài)的不符合旋轉(zhuǎn)機(jī)械實(shí)際性能劣化過(guò)程的現(xiàn)象，即早期故障預(yù)警模型的異常檢測(cè)能力和性能退化全局跟蹤能力混淆。三、論文提綱前言1緒論1.1旋轉(zhuǎn)機(jī)械故障診斷的意義1.2旋轉(zhuǎn)機(jī)械故障診斷的研究現(xiàn)狀1.3旋轉(zhuǎn)機(jī)械故障診斷技術(shù)2齒輪故障診斷基礎(chǔ)2.1齒輪故障常見(jiàn)形式2.1.1齒面磨損2.1.2齒面膠合和擦傷2.1.3齒面接觸疲勞（點(diǎn)蝕、削落）2.1.4彎曲疲勞和斷齒2.2齒輪常見(jiàn)故障征兆2.2.1設(shè)備在外觀方面的故障征兆2.2.2齒輪在性能方面的故障征兆2.3齒輪振動(dòng)信號(hào)的特征分析2.3.1齒輪軸的轉(zhuǎn)動(dòng)頻率及其各次諧波2.3.2齒輪的嚙合頻率2.3.3齒輪振動(dòng)的特征頻率2.3.4幾種特殊狀態(tài)齒輪的頻域特征2.4齒輪故障診斷試驗(yàn)臺(tái)及齒輪振動(dòng)信號(hào)簡(jiǎn)介3齒輪故障診斷時(shí)域方法分析3.1時(shí)域分析的基本理論3.1.1時(shí)頻域分析3.2小波變換（Wavelettransform）3.2.1連續(xù)小波變換3.2.2離散小波變換3.3一維離散小波MATLAB中實(shí)現(xiàn)方法3.4基于一維離散小波對(duì)齒輪故障診斷的研究4基于EMD的齒輪故障診斷4.1基于EMD（經(jīng)驗(yàn)?zāi)Ｊ椒纸猓┑恼駝?dòng)信號(hào)特征提取4.1.1EMD的研究背景4.1.2經(jīng)驗(yàn)?zāi)Ｊ椒纸?EMD)方法原理4.1.3基于EMD的振動(dòng)信號(hào)特征提取分析4.2利用EMD研究齒輪故障診斷4.3針對(duì)仿真出來(lái)的波形進(jìn)行分析結(jié)論四、與選題相關(guān)的主要參考文獻(xiàn)（列出：作者、著作名稱(chēng)/論文名稱(chēng)、出版社/期刊名稱(chēng)、出版年月/刊期），以下為參考文獻(xiàn)格式舉例。[1] SOLEIMANIA,KHADEMSEJCS,FRACTALS.Earlyfaultdetectionofrotatingmachinerythroughchaoticvibrationfeatureextractionofexperimentaldatasets[J].2015,78:61-75.[2] XUX,TAOZ,MINGW,etal.Intelligentmonitoringanddiagnosticsusinganovelintegratedmodelbasedondeeplearningandmulti-sensorfeaturefusion[J].2020,165:108086.[3] SHIH,GUOJ,BAIX,etal.ResearchonaNonlinearDynamicIncipientFaultDetectionMethodforRollingBearings[J].AppliedSciences-Basel,2020,10(7).[4] 唐貴基,王曉龍.最大相關(guān)峭度解卷積結(jié)合1.5維譜的滾動(dòng)軸承早期故障特征提取方法%J振動(dòng)與沖擊[J].2015,34(12):79-84.[5] 唐貴基,王曉龍.自適應(yīng)最大相關(guān)峭度解卷積方法及其在軸承早期故障診斷中的應(yīng)用%J中國(guó)電機(jī)工程學(xué)報(bào)[J].2015,35(06):1436-44.[6] 王宏超,陳進(jìn),董廣明.基于最小熵解卷積與稀疏分解的滾動(dòng)軸承微弱故障特征提取%J機(jī)械工程學(xué)報(bào)[J].2013,49(01):88-94.[7] 楊蕊,李宏坤,賀長(zhǎng)波,etal.利用最優(yōu)小波尺度循環(huán)譜的滾動(dòng)軸承早期故障特征提取%J機(jī)械工程學(xué)報(bào)[J].2018,54(17):208-17.[8] 唐貴基,王曉龍.參數(shù)優(yōu)化變分模態(tài)分解方法在滾動(dòng)軸承早期故障診斷中的應(yīng)用%J西安交通大學(xué)學(xué)報(bào)[J].2015,49(05):73-81.[9] JIANGF,ZHUZ,LIW.AnImprovedVMDWithEmpiricalModeDecompositionandItsApplicationinIncipientFaultDetectionofRollingBearing[J].IEEEAccess,2018,6