【摘要】：在實際工程應(yīng)用中,獲得的信號一般為非平穩(wěn)信號,對于非平穩(wěn)信號的分析與處理十分重要。在處理這些數(shù)據(jù)序列時,以往經(jīng)常用到的傳統(tǒng)時頻分析方法的根本都是傅里葉變換,因此在處理非平穩(wěn)信號時會存在一定的局限性。經(jīng)驗?zāi)B(tài)分解(EMD)是美國國家宇航局的華裔科學(xué)家Norden E.Huang等人于1998年在分析非平穩(wěn)、非線性信號時提出的一種新型的、具有自適應(yīng)性的時頻分析方法,在傳統(tǒng)的信號處理方法上取得了很大的改進(jìn),是一種現(xiàn)代化的信號處理方法,并且它不需要任何先驗知識,僅根據(jù)信號自身的特點自適應(yīng)的將任一復(fù)雜非平穩(wěn)信號分解為若干個內(nèi)稟模態(tài)分量(IMF)和一個余量之和,所有的內(nèi)稟模態(tài)分量經(jīng)傅里葉變換之后都能夠得到原信號的具有物理意義的瞬時頻率。EMD方法相較于傳統(tǒng)信號處理方法具有更多的優(yōu)點,被廣泛應(yīng)用到圖形處理、信號處理、振動測試和機械故障診斷等多個領(lǐng)域,都取得了良好的效果。本文在深入學(xué)習(xí)、研究EMD算法的基礎(chǔ)上,對其存在的模態(tài)混疊現(xiàn)象進(jìn)行了改進(jìn),提出了一種基于能量的掩膜信號法。根據(jù)能量守恒定律,當(dāng)內(nèi)稟模態(tài)分量中不存在虛假模態(tài)分量時,分解過程能量守恒,所有分量的能量之和等于原信號的能量,但是當(dāng)有虛假模態(tài)分量存在時,能量是不守恒的,原信號的能量低于各分量的能量之和,任意兩個分量之和的能量也是小于其能量之和的,以此確定了能量泄露的主要去向,降低了能量泄露對計算掩膜信號頻率的影響,彌補了掩膜信號法的不足并將改進(jìn)后的EMD算法應(yīng)用在實際工程中的非平穩(wěn)信號處理上。連續(xù)采煤機是大型的地下采掘設(shè)備,主要振動信號頻率為低頻,以連采機截割臂振動信號這一非平穩(wěn)信號為例來進(jìn)行研究。首先通過計算在連采機截割臂上不同點模態(tài)運動能的大小,從而對傳感器的安裝位置進(jìn)行優(yōu)化,得到了連采機截割臂在工作過程中的振動信號,在原信號中會存在噪聲,通過EMD方法進(jìn)行降噪,去掉信號中的高頻噪聲;然后通過改進(jìn)的掩膜信號法對降噪后的信號進(jìn)行分析研究,結(jié)果中成功消除了EMD中存在的模態(tài)混疊現(xiàn)象,說明了能量掩膜信號法在實際工程應(yīng)用中也能達(dá)到消除模態(tài)混疊現(xiàn)象上的效果。
[Abstract]:In practical engineering applications, the obtained signals are generally non-stationary signals, which is very important for the analysis and processing of non-stationary signals. In the processing of these data sequences, the traditional time-frequency analysis methods often used in the past are based on Fourier transform, so there are some limitations in the processing of non-stationary signals. Empirical mode decomposition (EMD) is a new adaptive time-frequency analysis method proposed by Norden E.Huang et al., a Chinese scientist from NASA, in 1998 when analyzing nonstationary and nonlinear signals. Great improvement has been made in the traditional signal processing method, which is a modern signal processing method, and it does not require any prior knowledge. According to the characteristics of the signal itself, any complex non-stationary signal is decomposed into the sum of several intrinsic modal components (IMF) and a residue. After Fourier transform, all intrinsic modal components can obtain the physical instantaneous frequency of the original signal. Compared with the traditional signal processing method, the EMD method has more advantages and is widely used in graphic processing and signal processing. Many fields such as vibration test and mechanical fault diagnosis have achieved good results. In this paper, based on the study of the EMD algorithm, the existing mode aliasing is improved, and an energy-based mask signal method is proposed. According to the conservation law of energy, when there is no false mode component in intrinsic mode component, the energy conservation of decomposition process, the sum of energy of all components is equal to the energy of the original signal, but when there is false mode component, the energy is not conserved. The energy of the original signal is lower than the sum of the energy of each component, and the energy of the sum of any two components is also smaller than the sum of its energy. The main direction of the energy leakage is determined, and the influence of the energy leakage on the calculation of the frequency of the mask signal is reduced. It makes up for the deficiency of the mask signal method and applies the improved EMD algorithm to the non-stationary signal processing in practical engineering. Continuous shearer is a large underground mining equipment, the main frequency of vibration signal is low frequency, taking the non-stationary signal of cutting arm of continuous mining machine as an example to study. First of all, the vibration signal of the cutting arm of the continuous mining machine is obtained by calculating the magnitude of the different mode motion energy at different points on the cutting arm of the continuous mining machine, and the installation position of the sensor is optimized, and the noise in the original signal is obtained. The high frequency noise in the signal is removed by the EMD method, and then the noise reduction signal is analyzed by the improved mask signal method, and the mode aliasing phenomenon in the EMD is successfully eliminated. It is shown that the energy mask signal method can also be used in practical engineering to eliminate the phenomenon of mode aliasing.
【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】：2017
【分類號】：TN911.7

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 段若晨;王豐華;周荔丹;;基于優(yōu)化HHT算法與洛侖茲信息量度的換流變用有載分接開關(guān)機械特征提取[J];中國電機工程學(xué)報;2016年11期

2 楊陽;吉曉冬;楊子賢;黃向玉;吳淼;;掘進(jìn)機回轉(zhuǎn)臺振動信號狀態(tài)監(jiān)測和分析[J];煤炭工程;2015年08期

3 凌立鵬;毛毳;李長躍;;一種改進(jìn)掩膜信號法及其在車橋振動響應(yīng)信號中的應(yīng)用[J];噪聲與振動控制;2015年03期

4 徐卓飛;劉凱;;基于極值符號序列分析的EMD端點效應(yīng)處理方法[J];振動.測試與診斷;2015年02期

5 肖瑛;殷福亮;;解相關(guān)EMD:消除模態(tài)混疊的新方法[J];振動與沖擊;2015年04期

6 石瑞敏;楊兆建;;基于改進(jìn)EMD的多繩摩擦提升機載荷信息特征提取[J];煤炭學(xué)報;2014年04期

7 路亮;龍源;謝全民;劉好全;趙長嘯;李興華;;提升小波包最優(yōu)基分解算法在爆破振動信號分析中的應(yīng)用研究[J];振動與沖擊;2014年05期

8 朱文龍;周建中;肖劍;肖漢;李超順;;獨立分量分析 經(jīng)驗?zāi)B(tài)分解特征提取在水電機組振動信號中的應(yīng)用[J];中國電機工程學(xué)報;2013年29期

9 李舜酩;郭海東;李殿榮;;振動信號處理方法綜述[J];儀器儀表學(xué)報;2013年08期

10 苑宇;李聰明;趙玉龍;;BEMD方法的若干研究及改進(jìn)[J];振動.測試與診斷;2013年S1期

，

本文編號：2282068