盲分离与Hilbert-Huang结合应用于航空发动机振动信号分析

在航空发动机故障诊断中,首要任务是分析故障信号提取故障特征。针对航空发动机非平稳振动信号,提出了利用盲分离(BSS)获得发动机的振源信号,结合Hilbert-Huang变换(HHT)对振源信号进行时频分析提取故障特征的方法。首先利用仿真信号验证了此方法的有效性,然后分析了某航空涡扇发动机空中停车故障并与直接应用HHT分析的结果进行比较,证实了盲分离与HHT的结合能更准确地提取航空发动机非平稳故障特征。     

应用多分辨SVD和ICA的轴承复合故障诊断研究

针对滚动轴承复合故障信号中故障特征难以分离的问题,提出了基于多分辨奇异值分解(SVD)和独立分量分析(ICA)的复合故障诊断方法。首先利用多分辨SVD将复合故障振动信号分解为几个分量实现维数的增加;然后将分解得到的分量组合为混合信号,并利用ICA进行欠定盲分离;最后对分离后的独立分量进行Hilbert包络解调,由此实现对复合故障特征信息的分离和故障识别。通过对滚动轴承内外圈复合故障的试验信号分析表明,该方法可以有效地分离和提取轴承复合故障的特征信息。     

基于SK‑MOMEDA的风电机组轴承复合故障特征分

针对在实际工况中风电机组滚动轴承发生复合故障时,多个故障间相互作用,彼此干扰,造成复合故障特征难以分离问题,提出了基于谱峭度（spectral kurtosis,简称SK）与多点最优调整的最小熵解卷积（multipoint optimal minimum entropy deconvolution adjusted,简称MOMEDA）的风电机组滚动轴承复合故障特征分离提取方法。首先,对复合故障信号进行谱峭度分析,选出能量较大的共振频带,并通过构建带通滤波器对相应的共振频带进行滤波,对滤波信号进行包络谱分析,对单一故障特征进行分离提取;其次,对未能实现单一故障特征提取的滤波信号进行多点峭度谱分析并确定故障周期,应用MOMEDA完成后续分离提取过程。仿真信号和工程应用分析结果表明,该方法能够准确且有效地实现轴承复合故障特征的分离提取。     

机械故障信号的分离

针对机械故障信号经常是多种故障信号的混合，给正确的故障识别造成很大困难的实际情况，提出基于神经网络非线性主分量分析的机构故障信号分离方法。阐述了故障信息的分离与主分量分析的关系。并将二者统一起来，从理论上证明应用主分量分析方法进行故障分离的有效性；介绍神经网络非线性分离，取得令人满意的结果。     

基于ITD-KICA盲分离降噪的滚动轴承故障特征提取

滚动轴承在实际工况下的故障信号和故障信息常常淹没于噪声中,传统的故障特征提取方法很难有效提取出轴承故障特征信息。因此,采用时间固有尺度分解(ITD)和核独立分量分析(KICA)相结合的信噪盲分离分析法降噪。对轴承信号进行ITD分解,根据相关系数将分解得到的PRC分量重组以及构建虚拟噪声通道,利用KICA解混实现故障信号与噪声信号分离,对信噪分离后的有效分量信号做包络谱的分析。通过仿真及轴承故障实验分析和对比表明,该方法能有效提取轴承的故障特征。     

基于SK-MOMEDA的风电机组轴承复合故障特征分离提取

针对在实际工况中风电机组滚动轴承发生复合故障时,多个故障间相互作用,彼此干扰,造成复合故障特征难以分离问题,提出了基于谱峭度(spectral kurtosis,简称SK)与多点最优调整的最小熵解卷积(multipoint optimal minimum entropy deconvolution adjusted,简称MOMEDA)的风电机组滚动轴承复合故障特征分离提取方法。首先,对复合故障信号进行谱峭度分析,选出能量较大的共振频带,并通过构建带通滤波器对相应的共振频带进行滤波,对滤波信号进行包络谱分析,对单一故障特征进行分离提取;其次,对未能实现单一故障特征提取的滤波信号进行多点峭度谱分析并确定故障周期,应用MOMEDA完成后续分离提取过程。仿真信号和工程应用分析结果表明,该方法能够准确且有效地实现轴承复合故障特征的分离提取。     

CEEMDAN和盲源分离在轴承复合故障诊断中的应用

滚动轴承的复合故障信号中往往含有多个特征信息及背景噪声,为更高效实现故障信息的提取,提出一种基于具有自适应白噪声的完备集成经验模态分解（CEEMDAN）和盲源分离的滚动轴承复合故障特征提取方法。对实验所获取的故障数据进行CEEMDAN分解,得出一组固有模态函数（IMF）,利用加权峭度因子选取其中有效IMF重构信号,再将重构的信号进行BSS分离。对分离出的信号做解调包络分析,从其解调谱中提取故障信号的特征频率。结果证明了此方法可以有效地分离轴承的内外圈故障,使故障特征更易被提取。     

基于包络解调的离心式压缩机弱故障信号增强方法研究

在离心式压缩机使用要求不断提高下,为了增强故障诊断精确性,提出基于包络解调的非平稳工况下离心式压缩机弱故障信号增强方法。将小波包分析和独立分量分析结合,通过小波包分析法对含有噪声的混合信号进行降噪,根据 FastICA 算法分离降噪后的混合信号,对分离出的信号采用收缩函数实行频段内的去噪操作,完成多源故障信号分离去噪。在故障信号分离的基础上,考虑到被分离出的信号伴随着微弱噪声,进一步通过包络解调随机共振实现弱故障信号增强。对多源信号分离结果进行包络解调操作,并对包络信号实行变尺度随机共振输出处理,实现故障特征信号增强,达到故障诊断的目的。通过实验分别对此方法的信号去噪增强效果和故障诊断精确性进行验证,实验结果表明,该方法不仅弱故障信号增强效果显著,且故障诊断鲁棒性强,精度高,具有可实践性。     

发动机转子振动信号分析

针对航空发动机故障类型难以识别和转子振动信号复杂、难以分离的问题,提出运用盲源分离中的Fast ICA算法建立振动信号的分离模型,从采集信号中准确分离出独立的故障信号,快速识别转子中的故障类型。通过搭建发动机转子振动平台采集转子的振动信号,同时计算出不同故障状态下的故障频率。对比分析得出振动信号经过Fast ICA算法处理后具有更高的辨识性,由分离后的信号可以判断出转子振动的故障类型为转子通过外环。分析结果表明:基于Fast ICA算法的分离模型可以快速、有效地分离出此类发动机转子振动信号。     

双树复小波包和ICA用于滚动轴承复合故障诊断

针对滚动轴承复合故障信号特征难以分离的问题,提出将双树复小波包变换和独立分量分析(independent component analysis,简称ICA)结合的方法应用到滚动轴承复合故障诊断中。首先,利用双树复小波包变换将复杂的、非平稳的复合故障信号分解为若干不同频带的分量;其次,引入ICA对各个分量所组成的混合信号进行盲源分离,从而尽可能消除频率混叠;最后,对从混合信号中分离出来的独立信号分量进行希尔伯特解调,即可实现对复合故障特征信息的分离和故障识别。试验结果表明,该方法可以有效地分离和提取轴承复合故障的特征频率,验证了方法的可行性和有效性。     

主元分析方法的故障可检测性研究

主元分析（PCA）是一种有效的多元统计过程监测方法，PCA监测方法不依赖于过程的精确数学模型，这使得其难以对故障的可检测性问题进行系统的研究，基于故障子空间的描述方式，本文在主要元子空间的残差子空间中分别讨论了PCA故障可检测性的充分和必要条件，并提出了临界故障值的概念，通过对双效蒸发过程的仿真故障检测，表明所获得的结果能较好地刻画PCA的故障检测行为。     

提高大型复杂机电系统故障诊断质量的几种新方法

分析了大型复杂机电系统故障的可诊断性问题,探讨了影响故障可诊断性的主要因素及其评价标准,研究了HHT(Hilbert/Huang transform)时频分析方法在提高大型复杂机电系统诊断信息质量中的应用。为了进一步提高诊断系统对未知故障的诊断质量,分析讨论了自组织特征映射、生成拓扑映射,以及曲元分析等无监督机器学习算法在大型复杂机电系统故障诊断中的应用。针对故障信号的非线性特征以及多类复杂故障的线性不可分问题,结合机器学习领域的最新研究成果,探讨了基于核的机械故障特征提取方法与基于核的故障模式分类方法,并对采用核方法分析设备运行状态的趋势变化作了初步探讨。     

A method for the compound fault diagnosis of gearboxes based on morphological component analysis

An improved morphological component analysis (MCA) method is proposed for the compound fault diagnosis of gearboxes. When gear fault and bearing fault occur simultaneously, the compound fault signal of the gearbox contains meshing components (related to the gear fault) and periodic impulse components (related to the bearing fault). The corresponding fault characteristics can be separated by MCA according to the morphological differences of the components. In the proposed method, the optimal dictionary, which can represent the characteristics of bearing faults, is first selected based on the principle of minimum information entropy. Then, the compound fault signal is decomposed into the meshing component and the periodic impulse component using MCA. Finally, the separated components are subjected to the Hilbert envelope spectrum analysis. The faults of the gear and the bearing can be diagnosed according to the envelope spectra of the separated fault signal components. Simulation and experimental studies validate the effectiveness of the proposed method for the compound fault diagnosis of gearboxes.     

A fault identification and classification scheme for an automobile door assembly process

A process fault identification and classification scheme for an automobile door assembly process is presented based on multivariate in-line dimensional measurements and principal component factor analysis. First, the door assembly process and the dimensional measurement system are briefly introduced. Second, the technique of principal component factor analysis is presented for process fault identification. Process faults are summarized based on off-line identified case studies. Finally a machine classification scheme based on principal components and principal factors is presented and evaluated, using the pattern knowledge obtained off-line. This scheme is shown to be effective in classifying process faults using production data.     

Adaptive PCA based fault diagnosis scheme in imperial smelting process

In this paper, an adaptive fault detection scheme based on a recursive principal component analysis (PCA) is proposed to deal with the problem of false alarm due to normal process changes in real process. Our further study is also dedicated to develop a fault isolation approach based on Generalized Likelihood Ratio (GLR) test and Singular Value Decomposition (SVD) which is one of general techniques of PCA, on which the off-set and scaling fault can be easily isolated with explicit off-set fault direction and scaling fault classification. The identification of off-set and scaling fault is also applied. The complete scheme of PCA-based fault diagnosis procedure is proposed. The proposed scheme is first applied to Imperial Smelting Process, and the results show that the proposed strategies can be able to mitigate false alarms and isolate faults efficiently.     

Multiple manifolds analysis and its application to fault diagnosis

A novel approach to fault diagnosis is proposed using multiple manifolds analysis (MMA) to extract manifold information from the vibration signals collected from a mechanical system. The basic idea of MMA is to reconstruct a manifold by embedding time series into a high-dimensional phase space. The tangent direction of the neighborhood for each point is then used to approximate its local geometry. The variation of the multiple manifolds representing different states of the mechanical system can be revealed by performing multi-way principal component analysis. The vibration signals acquired from roller bearings are employed to validate the proposed algorithms. Test results show that the proposed MMA-based approach can interpret different machine conditions and is effective to the fault diagnosis, and the MMA-based fault clustering and trend analysis algorithms have outperformed the conventional fault diagnosis methods.     

Fault diagnosis of star-connected auto-transformer based 24-pulse rectifier

This paper proposes a fault diagnosis method for star-connected auto-transformer based 24-pulse rectifier unit (ATRU) by integrating artificial neural networks (ANN) with wavelet packet decomposition (WPD) and principal component analysis (PCA). The WPD is employed to extract the features of different fault waveforms of the output voltage of the rectifier. PCA is adopted to reduce the dimensionality of the extracted feature vectors, which leads to fast computation of the algorithm. Back Propagation (BP) neural network is adopted to classify the fault types and determine the fault location according to the extracted features. These faults are simulated in real-time simulation platform and the obtained data are then analyzed with MATLAB toolbox, and finally verified with digital signal processor. Compared with other diagnosis methods, the proposed method shows better performance and faster computing speed.     

指数加权动态核主元分析法及其在故障诊断中应用

核主元分析法能充分利用核函数来解决非线性问题,具有很好的非线性逼近能力,但传统的核主元分析不能处理动态问题。在分析核主元分析法的基础上,提出一种新的指数加权核主元分析算法,建立一个多变量加权自回归统计核主元模型,选择Q统计量来判断系统是否发生故障,给出指数加权核主元分析法诊断故障的具体计算步骤。对液压泵进行了试验,利用小波包对液压泵端盖的振动信号进行处理,提取由13个时域和时频域特征量构成的故障特征矢量。试验结果表明,与传统的核主元分析法相比,新方法能实时更新主元模型和控制限Qa,合理地利用实时动态信息,能较好地处理动态问题,通过计算比较选择合适的加权因子,能获得良好的故障诊断效果,该方法是可行而有效的。     

Fault detection and classification in chemical processes based on neural networks with feature extraction

Feed forward neural networks are investigated here for fault diagnosis in chemical processes, especially batch processes. The use of the neural model prediction error as the residual for fault diagnosis of sensor and component is analyzed. To reduce the training time required for the neural process model, an input feature extraction process for the neural model is implemented. An additional radial basis function neural classifier is developed to isolate faults from the residual generated, and results are presented to demonstrate the satisfactory detection and isolation of faults using this approach.