《電子技術(shù)應(yīng)用》
您所在的位置:首頁(yè) > 其他 > 設(shè)計(jì)應(yīng)用 > 基于EEMD分解與PCA-FCM聚類的 岸橋減速箱故障診斷方法
基于EEMD分解與PCA-FCM聚類的 岸橋減速箱故障診斷方法
2021年電子技術(shù)應(yīng)用第4期
顧能華1,,侯銀銀2,韓雪龍1
1.衢州學(xué)院 電氣與信息工程學(xué)院,,浙江 衢州324000,; 2.國(guó)網(wǎng)浙江省電力有限公司衢州供電公司,浙江 衢州324000
摘要: 針對(duì)岸邊橋式起重機(jī)(岸橋)減速箱特征提取以及故障診斷問(wèn)題,,提出了一種集合經(jīng)驗(yàn)?zāi)B(tài)分解(EEMD)與主成分分析(PCA)-模糊C均值(FCM)聚類的減速箱故障診斷組合方法,。首先,通過(guò)EEMD分解將減速箱非線性,、非平穩(wěn)振動(dòng)信號(hào)分解為若干個(gè)固有模態(tài)函數(shù)(IMF),,提取每個(gè)IMF分量的多維故障特征;然后,,使用主成分分析法對(duì)故障特征進(jìn)行降維,,并分析了減速箱振動(dòng)信號(hào)的特征值與故障模式之間的關(guān)系,通過(guò)模糊C均值聚類算法對(duì)減速箱的狀態(tài)進(jìn)行識(shí)別,。實(shí)驗(yàn)結(jié)果表明,,EEMD-PCA-FCM方法對(duì)減速箱的3種狀態(tài)有很高的識(shí)別準(zhǔn)確率,表明該方法是一種準(zhǔn)確有效的減速箱故障診斷方法,。
中圖分類號(hào): TN07
文獻(xiàn)標(biāo)識(shí)碼: A
DOI:10.16157/j.issn.0258-7998.200418
中文引用格式: 顧能華,侯銀銀,,韓雪龍. 基于EEMD分解與PCA-FCM聚類的岸橋減速箱故障診斷方法[J].電子技術(shù)應(yīng)用,,2021,47(4):101-106,,111.
英文引用格式: Gu Nenghua,,Hou Yinyin,Han Xuelong. Fault diagnosis for quayside container crane reducer based on EEMD decomposition and PCA-FCM clustering[J]. Application of Electronic Technique,,2021,,47(4):101-106,111.
Fault diagnosis for quayside container crane reducer based on EEMD decomposition and PCA-FCM clustering
Gu Nenghua1,Hou Yinyin2,,Han Xuelong1
1.College of Electrical and Information Engineering,,Quzhou University,Quzhou 324000,,China,; 2.Zhejiang Quzhou Power Supply Company State Grid,Quzhou 324000,,China
Abstract: Aiming at the feature extraction and fault diagnosis issue of quayside container crane(quayside crane) reducer, a combination method of reducer fault diagnosis based on ensemble empirical mode decomposition(EEMD) and principal component analysis(PCA)-fuzzy C-means(FCM) clustering is proposed. Firstly, the nonlinear and non-stationary vibration signals of the reducer are decomposed into several intrinsic mode functions(IMF) by EEMD decomposition, and the multi-dimensional fault characteristics of each IMF component are extracted. Then, the principal component analysis method is used to visually reduce the dimension of the fault feature, the relationship between the characteristic value of the vibration signal of the reducer and the fault mode is analyzed, and the state of the reducer is identified by the fuzzy C-means clustering algorithm. The experimental results show that EEMD-PCA-FCM method has high recognition accuracy for the three states of the reducer, which indicates that the method is an accurate and effective reducer fault diagnosis method.
Key words : fault diagnosis,;quayside container crane reducer;ensemble empirical mode decomposition(EEMD),;fuzzy C-means clustering,;principal component analysis(PCA)

0 引言

    岸橋常作業(yè)于高速、重載,、大沖擊的工作環(huán)境中,,其起升減速箱由于傳動(dòng)力矩大,且長(zhǎng)時(shí)間受到強(qiáng)烈動(dòng)載的振動(dòng)沖擊,,是岸橋中最容易出現(xiàn)故障的部件之一[1],。因此,診斷監(jiān)測(cè)岸橋減速箱的狀態(tài)變得尤其重要,。本質(zhì)上,對(duì)減速箱進(jìn)行故障診斷是一種模式識(shí)別問(wèn)題,,為了更準(zhǔn)確地識(shí)別減速箱的狀態(tài),需要對(duì)減速箱振動(dòng)信號(hào)進(jìn)行有效的特征提取和更準(zhǔn)確的分類,。

    起升減速箱振動(dòng)信號(hào)為非平穩(wěn),、非線性、非周期信號(hào),,傳統(tǒng)的時(shí)域,、頻域以及時(shí)頻域方法缺乏對(duì)非平穩(wěn)和非線性信號(hào)的多分辨率分析和自適應(yīng)處理能力[2]。HUANG N E等[3]提出的經(jīng)驗(yàn)?zāi)B(tài)分解(Empirical Mode Decomposition,,EMD)由于其良好的自適應(yīng)分解特性在處理非線性和非平穩(wěn)信號(hào)時(shí)具有很大的優(yōu)勢(shì),。然而,EMD分解會(huì)產(chǎn)生模態(tài)混疊現(xiàn)象,。WU Z等[4]通過(guò)改進(jìn)EMD分解方法得到一種新的EEMD分解法,,能夠有效解決該問(wèn)題。王玉靜等[5]通過(guò)EEMD分解得到滾動(dòng)軸承振動(dòng)信號(hào)的固有模態(tài)函數(shù),,并結(jié)合峭度值,、相關(guān)系數(shù)提取信號(hào)的初始特征,能夠很好地提取故障特征信息,;魏文軍等[6]采用EEMD多尺度樣本熵提取特征對(duì)S700K轉(zhuǎn)轍機(jī)進(jìn)行故障診斷,,通過(guò)EEMD分解提取轉(zhuǎn)轍機(jī)不同狀態(tài)下的特征參數(shù)并進(jìn)行聚類分析,,驗(yàn)證了該方法對(duì)故障診斷的精度和效率有明顯的提高。

    通過(guò)EEMD分解提取到的減速箱故障特征維數(shù)較高,,會(huì)導(dǎo)致故障診斷效率降低,,故選取PCA法對(duì)故障特征降維。PCA能較好融合減速箱的多個(gè)典型故障特征,,將高維故障特征集映射到低維空間中,,減少特征參數(shù)信息冗余[7]。故障特征參數(shù)的提取和選擇是機(jī)器診斷的關(guān)鍵,,而狀態(tài)識(shí)別則是診斷的核心,。FCM算法作為一種無(wú)監(jiān)督學(xué)習(xí)算法,可以根據(jù)特征參數(shù)樣本的相似性進(jìn)行分類,,使處于同一類的相似度最大,,并保證不同類間的差異性較大[8]。王印松等[9]將FCM應(yīng)用在控制系統(tǒng)的故障診斷中,,不僅可以較好地識(shí)別不同部件的故障,,還可以對(duì)同一部件不同類型的故障進(jìn)行診斷。樊紅衛(wèi)等[10]針對(duì)電主軸轉(zhuǎn)子不平衡故障,,提出一種對(duì)稱極坐標(biāo)圖像和FCM相結(jié)合的失衡故障診斷方法,,結(jié)果顯示具有較高的分類準(zhǔn)確率。

    本文結(jié)合EEMD分解和PCA-FCM聚類算法對(duì)岸橋減速箱進(jìn)行故障診斷,。首先,,將減速箱振動(dòng)信號(hào)進(jìn)行EEMD分解并提取故障特征,然后利用PCA對(duì)高維特征參數(shù)進(jìn)行約簡(jiǎn),,最后使用FCM算法對(duì)減速箱的狀態(tài)進(jìn)行聚類,,并通過(guò)實(shí)驗(yàn)分析驗(yàn)證了該方法的有效性。




本文詳細(xì)內(nèi)容請(qǐng)下載:http://forexkbc.com/resource/share/2000003485




作者信息:

顧能華1,,侯銀銀2,,韓雪龍1

(1.衢州學(xué)院 電氣與信息工程學(xué)院,浙江 衢州324000,;

2.國(guó)網(wǎng)浙江省電力有限公司衢州供電公司,,浙江 衢州324000)

此內(nèi)容為AET網(wǎng)站原創(chuàng),未經(jīng)授權(quán)禁止轉(zhuǎn)載,。