發(fā)布時(shí)間：2018-03-23 02:30

  本文選題：故障診斷　切入點(diǎn)：小波包能量譜　出處：《浙江大學(xué)》2017年碩士論文　論文類(lèi)型：學(xué)位論文

【摘要】：滾動(dòng)軸承是機(jī)械設(shè)備中常見(jiàn)且十分重要的通用零件,軸承發(fā)生故障往往會(huì)造成嚴(yán)重的生命財(cái)產(chǎn)安全問(wèn)題。所以,對(duì)滾動(dòng)軸承進(jìn)行故障檢測(cè)研究是國(guó)內(nèi)和國(guó)外工業(yè)領(lǐng)域的熱門(mén)話題。實(shí)際工程中受實(shí)際數(shù)據(jù)采集環(huán)境影響,常出現(xiàn)祥本不平衡、樣本數(shù)量有限問(wèn)題,本文針對(duì)這兩大問(wèn)題提出異常檢測(cè)算法結(jié)合支持向量機(jī)的一種新的雙步診斷方法。首先對(duì)軸承的振動(dòng)特性進(jìn)行分析:在對(duì)軸承的基本結(jié)構(gòu)、振動(dòng)特性和軸承的動(dòng)力學(xué)特性進(jìn)行分析后,確定采用振動(dòng)信號(hào)的故障診斷方式。接著采用小波包能量譜特征提取方式提取軸承振動(dòng)信號(hào):采用小波包變換對(duì)軸承信號(hào)進(jìn)行分解重構(gòu),得到能量譜特征向量,小波包變換不僅可以去噪,而且可以對(duì)高頻段的故障振動(dòng)信號(hào)進(jìn)行分解細(xì)化,有效地提取故障軸承的信號(hào)特征。在軸承信號(hào)經(jīng)過(guò)小波包變換處理后,分析軸承樣本識(shí)別診斷問(wèn)題。首先研究基于多分類(lèi)支持向量機(jī)(SVM)的單步故障診斷方法:分析不同核函數(shù)下多分類(lèi)支持向量機(jī)的診斷效果,確定最佳核函數(shù)。針對(duì)樣本不平衡的情況,發(fā)現(xiàn)優(yōu)化后的SVM分類(lèi)器在該情況下診斷性能仍有不佳。在研究單步故障診斷方法后進(jìn)一步研究基于結(jié)合異常檢測(cè)算法的雙步故障診斷方法:基于異常檢測(cè)算法的特點(diǎn),對(duì)小波包能量譜的特征信息進(jìn)一步優(yōu)化,確定先采用異常檢測(cè)算法進(jìn)行故障檢測(cè)再采用SVM分類(lèi)器進(jìn)行故障分類(lèi)的雙步故障診斷模型。該方法可以將大量的正常軸承先一步檢測(cè)出來(lái),大大減少后續(xù)支持向量機(jī)的分類(lèi)負(fù)擔(dān)。最后建立單步與雙步故障診斷模型的評(píng)價(jià)準(zhǔn)則:根據(jù)實(shí)際工業(yè)中不同誤判帶來(lái)的損失是不相同的,創(chuàng)建一種結(jié)合工業(yè)實(shí)際損失與故障診斷誤判概率的誤判損失評(píng)價(jià)準(zhǔn)則。該準(zhǔn)則不僅更全面而且更加符合實(shí)際情況。基于該準(zhǔn)則,針對(duì)樣本不平衡、樣本數(shù)量有限的情況,驗(yàn)證了基于異常檢測(cè)算法結(jié)合支持向量機(jī)的雙步故障檢測(cè)模型的有效性和優(yōu)越性。
[Abstract]:Rolling bearing is a common and very important universal part in mechanical equipment. Bearing failure often causes serious life and property safety problems. The research of rolling bearing fault detection is a hot topic in domestic and foreign industrial field. Affected by the environment of actual data collection, the problem of unbalance and limited sample size often occurs in practical engineering. In this paper, a new two-step diagnosis method, which combines anomaly detection algorithm with support vector machine (SVM), is proposed to solve these two problems. Firstly, the vibration characteristics of bearings are analyzed. After analyzing the vibration characteristics and the dynamic characteristics of the bearing, The fault diagnosis method of vibration signal is determined, and then the bearing vibration signal is extracted by wavelet packet energy spectrum feature extraction. The energy spectrum characteristic vector is obtained by decomposing and reconstructing the bearing signal by wavelet packet transform. Wavelet packet transform can not only denoise, but also decompose and refine the fault vibration signal in high frequency band, and extract the signal characteristics of the fault bearing effectively. After the bearing signal is processed by wavelet packet transform, First, the single-step fault diagnosis method based on multi-classification support vector machine (SVM) is studied, and the diagnosis effect of multi-classification support vector machine under different kernel functions is analyzed. Determine the best kernel function. In case of sample imbalance, It is found that the performance of the optimized SVM classifier is still poor in this case. After studying the single-step fault diagnosis method, the two-step fault diagnosis method based on the anomaly detection algorithm is further studied: based on the characteristics of the anomaly detection algorithm. The characteristic information of wavelet packet energy spectrum is further optimized. A two-step fault diagnosis model using anomaly detection algorithm and SVM classifier is determined. This method can detect a large number of normal bearings in one step. The classification burden of subsequent support vector machines is greatly reduced. Finally, the evaluation criteria for single-step and two-step fault diagnosis models are established: according to the loss caused by different misjudgments in actual industry, An evaluation criterion of misjudgment loss combining industrial actual loss and fault diagnosis misjudgment probability is established. The criterion is not only more comprehensive but also more in line with the actual situation. Based on this criterion, the sample is unbalanced and the sample size is limited. The effectiveness and superiority of the two-step fault detection model based on anomaly detection algorithm and support vector machine are verified.
【學(xué)位授予單位】：浙江大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類(lèi)號(hào)】：TH133.33

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 肖順根;宋萌萌;;基于小波包能量神經(jīng)網(wǎng)絡(luò)的滾動(dòng)軸承故障診斷方法[J];機(jī)械強(qiáng)度;2014年03期

2 蔡改改;劉海洋;黃偉國(guó);朱忠奎;;自適應(yīng)譜峭度濾波方法及其振動(dòng)信號(hào)檢測(cè)應(yīng)用[J];振動(dòng).測(cè)試與診斷;2014年02期

3 胡耀斌;厲善元;胡良斌;;基于神經(jīng)網(wǎng)絡(luò)的滾動(dòng)軸承故障診斷方法的研究[J];機(jī)械設(shè)計(jì)與制造;2012年02期

4 劉永斌;葉平;潘成亮;;輻射電場(chǎng)型壓電薄膜實(shí)驗(yàn)研究[J];壓電與聲光;2010年06期

5 許麗;張進(jìn)明;張廣明;徒偉;;基于PCA的滾動(dòng)軸承故障檢測(cè)方法[J];計(jì)算機(jī)仿真;2010年06期

6 萬(wàn)書(shū)亭;佟海俠;董炳輝;;基于最小二乘支持向量機(jī)的滾動(dòng)軸承故障診斷[J];振動(dòng).測(cè)試與診斷;2010年02期

7 畢錦煙;李巍華;;基于半監(jiān)督模糊核聚類(lèi)的齒輪箱離群檢測(cè)方法[J];機(jī)械工程學(xué)報(bào);2009年10期

8 張重雄;周敏;;基于虛擬儀器技術(shù)的汽車(chē)制動(dòng)性能檢測(cè)系統(tǒng)[J];電子測(cè)量技術(shù);2008年10期

9 萬(wàn)書(shū)亭;呂路勇;;基于小波包和細(xì)化包絡(luò)分析的滾動(dòng)軸承故障診斷方法研究[J];中國(guó)工程機(jī)械學(xué)報(bào);2008年03期

10 張玉恒;吳啟暉;王金龍;;基于時(shí)頻加窗短時(shí)傅里葉變換的LFM干擾抑制[J];電子與信息學(xué)報(bào);2007年06期

，

本文編號(hào)：1651528