【摘要】：齒輪箱是一種傳動(dòng)比恒定、傳動(dòng)效率高、傳動(dòng)比變化范圍大的傳動(dòng)部件，被廣泛應(yīng)用在各種機(jī)械設(shè)備中。但因其結(jié)構(gòu)復(fù)雜且工作環(huán)境惡劣因素很容易受到損害和出現(xiàn)故障，影響整個(gè)設(shè)備的正常運(yùn)轉(zhuǎn)。對齒輪箱的故障診斷對了解設(shè)備運(yùn)行狀態(tài)，及時(shí)發(fā)現(xiàn)設(shè)備異常具有著重要的現(xiàn)實(shí)意義。 齒輪箱的振動(dòng)信號中蘊(yùn)含著大量信息，能夠從嘈雜的振動(dòng)信號中提取到包含故障特征的信號是一個(gè)關(guān)鍵因素。使用獨(dú)立分量分析作為信號的預(yù)處理步驟，實(shí)現(xiàn)從眾多混雜的信號中將含源的機(jī)械狀態(tài)信號分離出來，基于“純凈”信號再對其進(jìn)行ARMA建模分析，提取狀態(tài)特征，提高故障診斷的有效性。 論文分析了齒輪箱的常見失效形式，，并對齒輪和軸承失效作了重點(diǎn)分析。在研究了獨(dú)立分量分析算法和量子粒子群算法的理論基礎(chǔ)上，將二者相結(jié)合，通過選取一定的目標(biāo)函數(shù)，將量子粒子群算法作為獨(dú)立分量分析的優(yōu)化算法，形成了量子粒子群優(yōu)化獨(dú)立分量分析算法。以JZQ250齒輪箱為實(shí)驗(yàn)對象，搭建了測試系統(tǒng)，采集了振動(dòng)信號。將獨(dú)立分量分析算法應(yīng)用于齒輪箱的故障診斷中，取得了一定的效果。利用MATLAB軟件平臺開發(fā)了基于ICA算法的齒輪箱故障特征值提取軟件，該軟件由數(shù)據(jù)采集模塊、ICA算法仿真模塊、齒輪箱故障特征提取診斷模塊組成，不僅模擬了信號的實(shí)時(shí)數(shù)據(jù)采集，而且成功的將ICA算法與ARMA模型用于齒輪箱的故障診斷中。
[Abstract]:Gearbox is a kind of transmission parts with constant transmission ratio, high transmission efficiency and wide range of transmission ratio. It is widely used in various mechanical equipment. However, because of its complex structure and poor working environment, it is easy to be damaged and malfunction, which affects the normal operation of the whole equipment. The fault diagnosis of gearbox is of great practical significance to understand the operation state of the equipment and find out the abnormal equipment in time. There is a lot of information in the vibration signal of the gearbox. It is a key factor to extract the signal containing the fault feature from the noisy vibration signal. Independent component analysis (ICA) is used as the signal preprocessing step to separate the mechanical state signal with source from many mixed signals. Based on the "pure" signal, it is modeled and analyzed by ARMA to extract the state feature. Improve the effectiveness of fault diagnosis. The common failure forms of gear box are analyzed, and the failure of gear and bearing is analyzed emphatically. Based on the theory of independent component analysis (ICA) and quantum particle swarm optimization (QPSO), quantum particle swarm optimization (QPSO) is used as the optimization algorithm of ICA by selecting a certain objective function. A quantum particle swarm optimization (QPSO) independent component analysis (ICA) algorithm is proposed. Taking the JZQ250 gearbox as the experimental object, the testing system was set up and the vibration signal was collected. The independent component analysis (ICA) algorithm is applied to the gearbox fault diagnosis. The software of gearbox fault eigenvalue extraction based on ICA algorithm is developed by using MATLAB software platform. The software consists of data acquisition module, ICA algorithm simulation module, gearbox fault feature extraction and diagnosis module. Not only the real-time data acquisition is simulated, but also the ICA algorithm and ARMA model are successfully used in the gearbox fault diagnosis.
【學(xué)位授予單位】：中北大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2014
【分類號】：TH165.3

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 黃晉英;潘宏俠;畢世華;楊喜旺;;基于盲信號處理的傳動(dòng)箱狀態(tài)監(jiān)測與故障診斷[J];彈箭與制導(dǎo)學(xué)報(bào);2006年02期

2 王春;彭東林;;機(jī)械噪聲信號的基于小波變換的快速盲源分離法[J];工具技術(shù);2008年03期

3 高永生;唐力偉;王國敏;;齒輪箱系統(tǒng)特性分析[J];中國設(shè)備工程;2006年09期

4 高永生;吳莎;孫博;;基于AR模型的齒輪箱系統(tǒng)特性分析[J];中國設(shè)備工程;2007年09期

5 杜艷琴;黃華;;基于獨(dú)立成分分析和量子粒子群算法的胎兒心電提取[J];生物醫(yī)學(xué)工程學(xué)雜志;2011年05期

6 錢潔;鄭建國;;采用群體統(tǒng)計(jì)學(xué)習(xí)的量子進(jìn)化算法[J];西安交通大學(xué)學(xué)報(bào);2012年02期

7 胥永剛,張發(fā)啟,何正嘉;獨(dú)立分量分析及其在故障診斷中的應(yīng)用[J];振動(dòng)與沖擊;2004年02期

8 張新廣;李志農(nóng);王心怡;袁振偉;;一種基于階比域的AR模型盲辨識算法[J];振動(dòng)與沖擊;2009年05期

9 余紅英,閆宏偉,潘宏俠;齒輪振動(dòng)信號分解及其在故障診斷中的應(yīng)用[J];振動(dòng)、測試與診斷;2005年02期

10 黃晉英;畢世華;潘宏俠;楊喜旺;;獨(dú)立分量分析在齒輪箱故障診斷中的應(yīng)用[J];振動(dòng)、測試與診斷;2008年02期

相關(guān)碩士學(xué)位論文 前1條

1 白天駒;基于小波與感應(yīng)電機(jī)電流的齒輪箱系統(tǒng)故障診斷[D];東北大學(xué);2009年

本文編號：2407748