本文關(guān)鍵詞：基于隱Markov模型的滾動(dòng)軸承故障診斷方法研究　出處：《西南交通大學(xué)》2013年碩士論文　論文類型：學(xué)位論文

  更多相關(guān)文章： 滾動(dòng)軸承 小波包分解 主成分分析 連續(xù)高斯混合密度HMM 離散HMM 故障診斷

【摘要】：滾動(dòng)軸承廣泛應(yīng)用于旋轉(zhuǎn)機(jī)械中,然而由于各種原因,滾動(dòng)軸承很容易發(fā)生各種形式的故障,因此對(duì)滾動(dòng)軸承開展故障診斷便成為保證設(shè)備正常運(yùn)行的關(guān)鍵,具有重大的現(xiàn)實(shí)意義。 針對(duì)傳統(tǒng)的模式識(shí)別方法(如神經(jīng)網(wǎng)絡(luò)識(shí)別法)一直停留在靜態(tài)模式識(shí)別上的不足,本文提出采用一種近年來在語音識(shí)別技術(shù)中發(fā)展較快的動(dòng)態(tài)模式識(shí)別技術(shù)——隱馬爾科夫模型(Hidden Markov Model, HMM)來對(duì)滾動(dòng)軸承進(jìn)行故障診斷。HMM建模時(shí)統(tǒng)計(jì)的是一個(gè)時(shí)間跨度上的動(dòng)態(tài)信息,特別適合對(duì)信息量大、非平穩(wěn)、特征重復(fù)性不佳的診斷信號(hào)進(jìn)行分類。通常情況下的動(dòng)態(tài)過程為序列行為改變的表現(xiàn),滾動(dòng)軸承亦如此。若一個(gè)短時(shí)信號(hào)定義為一個(gè)幀,則每一個(gè)故障類型特定幀之間的轉(zhuǎn)移是不同的,因此可以用HMM來對(duì)特定幀的存在和各幀之間的轉(zhuǎn)移做統(tǒng)計(jì)處理。此外,利用HMM進(jìn)行模型訓(xùn)練時(shí)所用的樣本較少、速度較快,并且診斷的精度較高、模式分類能力較強(qiáng),因此非常適合對(duì)滾動(dòng)軸承的振動(dòng)信號(hào)進(jìn)行故障建模和分類。 根據(jù)實(shí)際的滾動(dòng)軸承典型故障實(shí)驗(yàn)數(shù)據(jù)和HMM故障診斷原理,本文首先對(duì)從各個(gè)軸承狀態(tài)下采集的振動(dòng)信號(hào)進(jìn)行分幀處理,然后提取每幀信號(hào)的時(shí)域、頻域和小波包能量特征參數(shù),組成特征矢量,并運(yùn)用主成分分析(Principal Components Analysis,PCA)技術(shù)對(duì)特征矢量進(jìn)行降維處理,在損失狀態(tài)信息較少的情況下,將多個(gè)特征指標(biāo)轉(zhuǎn)化為幾個(gè)綜合的特征指標(biāo),用較少的特征參數(shù)來代表軸承狀態(tài)的絕大部分信息。將PCA降維后的特征參數(shù)組合在一起形成特征矢量,大大簡化了后續(xù)模型的輸入。 根據(jù)HMM按觀察值的分類,本文研究了基于連續(xù)高斯混合密度HMM(CGHMM)的滾動(dòng)軸承故障診斷和基于離散HMM (DHMM)的滾動(dòng)軸承故障診斷技術(shù),并通過實(shí)驗(yàn)將兩種方法進(jìn)行了對(duì)比分析。由于利用DHMM進(jìn)行故障診斷時(shí),需要對(duì)特征值進(jìn)行量化編碼,這必然會(huì)帶來一定的量化誤差,因此DHMM模型的識(shí)別率(90%以上)低于CGHMM(98%以上),但由于DHMM建模簡單,故其訓(xùn)練速度快于CGHMM�？梢葬槍�(duì)診斷系統(tǒng)側(cè)重點(diǎn)的不同,選擇兩種方法。 為了驗(yàn)證HMM模型(包括CGHMM和DHMM)用于故障診斷的有效性和優(yōu)勢(shì),本文又將其識(shí)別結(jié)果與模式識(shí)別方面應(yīng)用最廣泛的BP神經(jīng)網(wǎng)絡(luò)的識(shí)別結(jié)果進(jìn)行了對(duì)比分析。實(shí)驗(yàn)結(jié)果表明,BP神經(jīng)網(wǎng)絡(luò)的訓(xùn)練速度慢于HMM,而且識(shí)別率低于HMM。由此可見,動(dòng)態(tài)模式識(shí)別方法HMM比神經(jīng)網(wǎng)絡(luò)識(shí)別法在故障診斷方面更具優(yōu)勢(shì),具有更廣泛的應(yīng)用和發(fā)展前景。
[Abstract]:Rolling bearings are widely used in rotating machinery. However, due to various reasons, rolling bearings are prone to various faults. Therefore, fault diagnosis of rolling bearings is the key to ensure the normal operation of equipment, which is of great practical significance.
In view of the traditional pattern recognition methods (such as neural network recognition method) has been stuck in the lack of static pattern recognition, this paper adopts a speech recognition technology in recent years in the development of dynamic pattern recognition technology quickly, the hidden Markoff model (Hidden Markov Model, HMM) to statistics for fault diagnosis of rolling bearing when modeling.HMM the dynamic information is a time span, particularly suitable for a large amount of information, the non-stationary signal characteristics, diagnosis of repetitive poor classification. The dynamic process is usually the case for sequences of behavior change, rolling bearing is also true. If a short-time signal is defined as a frame, the transfer between each fault type specific frame is different, so you can use HMM to transfer between the specific frame and the presence of each frame to do statistical processing. In addition, the model is trained by HMM It has fewer samples, faster speed, higher diagnosis accuracy and stronger ability of pattern classification, so it is very suitable for fault modeling and classification of rolling bearing vibration signals.
According to the data of typical faults of rolling bearing fault diagnosis experiment and HMM principle of practice, this article first frame processing of vibration signals collected from each bearing condition, then each frame signal extraction in time domain, frequency domain and wavelet packet energy feature, form feature vector, and using principal component analysis (Principal Components, Analysis, PCA) the technology of feature vector dimension, the less loss of state information under the condition of a plurality of characteristic indexes into several comprehensive indexes, most with less feature parameters to represent bearing state information. PCA will reduce the dimension of feature parameters are combined together to form a feature vector, which greatly simplifies the follow-up the input of the model.
According to the classification of HMM according to the observations, we study the continuous mixture density based on Gauss HMM (CGHMM) and the fault diagnosis of rolling bearings based on discrete HMM (DHMM) of the rolling bearing fault diagnosis technology, and through the experiment of the two methods are compared and analyzed. Due to the use of DHMM for fault diagnosis, the need for characteristic value encoding, which will bring a certain quantization error, so the recognition rate of the DHMM model (more than 90%) than CGHMM (more than 98%), but because the DHMM modeling simple, so the training speed is faster than CGHMM. for the diagnosis system of the different focus, two kinds of methods.
In order to verify the HMM model (including CGHMM and DHMM) is effective for fault diagnosis and the advantages of combining BP neural network to identify the identification results and the pattern recognition of the most widely used results were compared and analyzed. Experimental results show that the BP neural network training speed is slower than that of HMM, and thus the recognition rate is lower than HMM. dynamic, HMM pattern recognition method than neural network recognition method in fault diagnosis has more advantages, applications and broader development prospects.

【學(xué)位授予單位】：西南交通大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2013
【分類號(hào)】：TH133.33;TH165.3

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 陳錫先;王立健;;多講話者漢語單字識(shí)別[J];北京郵電學(xué)院學(xué)報(bào);1989年02期

2 馮長建;康晶;吳斌;胡紅英;;DHMM的動(dòng)態(tài)模式識(shí)別理論在旋轉(zhuǎn)機(jī)械故障診斷中的應(yīng)用[J];大連民族學(xué)院學(xué)報(bào);2006年03期

3 馬曉旭,張傳利,夏明超,黃益莊;基于離散隱馬爾科夫模型的變壓器勵(lì)磁涌流鑒別新算法[J];電力系統(tǒng)自動(dòng)化;2000年03期

4 何強(qiáng),毛士藝,張有為;多觀察序列連續(xù)隱含馬爾柯夫模型的無溢出參數(shù)重估[J];電子學(xué)報(bào);2000年10期

5 謝賢亞,王家鼎,李煥潤;EM-HMM多頻率線跟蹤算法[J];電子學(xué)報(bào);2002年01期

6 姚鴻勛,呂雅娟,高文;基于色度分析的唇動(dòng)特征提取與識(shí)別[J];電子學(xué)報(bào);2002年02期

7 陳友仁,趙正校;基于隱馬爾可夫模型的車牌自動(dòng)識(shí)別技術(shù)[J];紅外與激光工程;2001年02期

8 王春立,高文;具有不同數(shù)目狀態(tài)結(jié)點(diǎn)的HMMs在中國手語識(shí)別中的應(yīng)用[J];計(jì)算機(jī)研究與發(fā)展;2001年01期

9 陸汝華;楊勝躍;朱穎;樊曉平;;基于DHMM的軸承故障音頻診斷方法[J];計(jì)算機(jī)工程與應(yīng)用;2007年17期

10 何曉霞,沈玉娣,張西寧;連續(xù)小波變換在滾動(dòng)軸承故障診斷中的應(yīng)用[J];機(jī)械科學(xué)與技術(shù);2001年04期

相關(guān)博士學(xué)位論文 前2條

1 馮長建;HMM動(dòng)態(tài)模式識(shí)別理論、方法以及在旋轉(zhuǎn)機(jī)械故障診斷中的應(yīng)用[D];浙江大學(xué);2002年

2 蘇文勝;滾動(dòng)軸承振動(dòng)信號(hào)處理及特征提取方法研究[D];大連理工大學(xué);2010年

相關(guān)碩士學(xué)位論文 前5條

1 孔亞林;基于振動(dòng)信號(hào)的滾動(dòng)軸承故障診斷方法研究[D];大連理工大學(xué);2006年

2 張新明;聲發(fā)射技術(shù)在滾動(dòng)軸承故障診斷中的應(yīng)用[D];清華大學(xué);2006年

3 姜紹俊;小波變換在軸承故障診斷中的研究與應(yīng)用[D];大連交通大學(xué);2010年

4 查偉;基于HHT的滾動(dòng)軸承故障診斷的理論與方法研究[D];國防科學(xué)技術(shù)大學(xué);2009年

5 陸利威;基于聲發(fā)射技術(shù)的滑動(dòng)軸承故障診斷[D];湖南科技大學(xué);2010年

，

本文編號(hào)：1363145