An improved PCA method with application to boiler leak detection

Principal component analysis (PCA) is a popular fault detection technique. It has been widely used in process industries, especially in the chemical industry. In industrial applications, achieving a sensitive system capable of detecting incipient faults, which maintains the false alarm rate to a minimum, is a crucial issue. Although a lot of research has been focused on these issues for PCA-based fault detection and diagnosis methods, sensitivity of the fault detection scheme versus false alarm rate continues to be an important issue. In this paper, an improved PCA method is proposed to address this problem. In this method, a new data preprocessing scheme and a new fault detection scheme designed for Hotelling's T2 as well as the squared prediction error are developed. A dynamic PCA model is also developed for boiler leak detection. This new method is applied to boiler water/steam leak detection with real data from Syncrude Canada's utility plant in Fort McMurray, Canada. Our results demonstrate that the proposed method can effectively reduce false alarm rate, provide effective and correct leak alarms, and give early warning to operators.     

基于支持向量分类器PCA方法及其在过程监控和故障检测中的应用

提出了基于支持向量分类器对过程进行性能监控和故障检测的改进 PCA方法 ,该方法避免了多元统计过程控制(MSPC)假设主元必须服从正态分布的前提。此外 ,通过对苯 -甲苯两组分精馏分离过程的仿真研究表明 ,该方法是有效的 ,并具有比传统多元统计过程控制更为优越的性能。     

冷轧过程断带故障的诊断研究

在冷轧过程中,断带故障是冷轧工序的主要生产故障之一。针对冷轧过程断带故障的特点,提出一种基于核主元分析(KPCA)非线性特征提取和最小二乘支持向量机(LSSVM)分类的故障诊断方法。此方法采用KPCA理论将冷轧过程原始空间数据映射到高维空间,并在高维空间进行主元分析,从而降维、去相关性,得到冷轧过程非线性特征向量。将降维后的特征主元作为LSSVM输入进行训练和识别,根据LSSVM的输出结果判断冷轧过程工作状态与故障类型。仿真结果表明:基于KPCA非线性特征提取和LSSVM分类的故障诊断方法计算速度快,能有效地提取冷轧过程断带故障特征,识别断带故障类型。     

基于PCA-PDBNs的故障检测与自学习辨识

如何提高工业过程故障识别的准确性及其算法训练的效率一直是故障检测与辨识研究领域的重点和热点。将深度学习方法引入该领域,结合粒子群优化(PSO)算法和深度信念网络(DBNs),提出了一种基于PSO的DBNs辨识方法(即PSODBNs,PDBNs),使用该方法对复杂函数的拟合进行了数值仿真。实验结果表明,相比于基本的DBNs模型,经PSO算法对网络参数优化后的DBNs模型获得了更好的函数逼近效果,具有更高的辨识精度。为验证该方法在实际工业过程故障检测中的可行性,结合主元分析(PCA),提出了一种PCA-PDBNs模型,并将此应用于田纳西-伊斯曼(TE)过程的故障检测中,结果表明,基于PCA-PDBNs方法降低了故障检测模型的复杂度,进一步提高了对未知故障类型的辨识精度,取得了较好效果。     

Anti-aliasing lifting scheme for mechanical vibration fault feature extraction

A troublesome problem in application of wavelet transform for mechanical vibration fault feature extraction is frequency aliasing. In this paper, an anti-aliasing lifting scheme is proposed to solve this problem. With this method, the input signal is firstly transformed by a redundant lifting scheme to avoid the aliasing caused by split and merge operations. Then the resultant coefficients and their single subband reconstructed signals are further processed to remove the aliasing caused by the unideal frequency property of lifting filters based on the fast Fourier transform (FFT) technique. Because the aliasing in each subband signal is eliminated, the ratio of signal to noise (SNR) is improved. The anti-aliasing lifting scheme is applied to analyze a practical vibration signal measured from a faulty ball bearing and testing results confirm that the proposed method is effective for extracting weak fault feature from a complex background. The proposed method is also applied to the fault diagnosis of valve trains in different working conditions on a gasoline engine. The experimental results show that using the features extracted from the anti-aliasing lifting scheme for classification can obtain a higher accuracy than using those extracted from the lifting scheme and the redundant lifting scheme.     

Fault diagnosis of rotating machine by isometric feature mapping

Principal component analysis (PCA) and linear discriminate analysis (LDA) are well-known linear dimensionality reductions for fault classification. However, since they are linear methods, they perform not well for high-dimensional data that has the nonlinear geometric structure. As kernel extension of PCA, Kernel PCA is used for nonlinear fault classification. However, the performance of Kernel PCA largely depends on its kernel function which can only be empirically selected from finite candidates. Thus, a novel rotating machine fault diagnosis approach based on geometrically motivated nonlinear dimensionality reduction named isometric feature mapping (Isomap) is proposed. The approach can effectively extract the intrinsic nonlinear manifold features embedded in high-dimensional fault data sets. Experimental results with rotor and rolling bearing data show that the proposed approach overcomes the flaw of conventional fault pattern recognition approaches and obviously improves the fault classification performance.     

基于动态特征矩阵的k近邻风电机组故障检测方法

受风的间歇性和随机性影响风电机组运行状态频繁切换,导致设备状态异常检测误报和漏报情况严重,风电企业运维成本居高不下。为此,提出了基于动态特征矩阵的k近邻故障检测方法,该方法采用基于互信息的动态特征矩阵描述风电机组的动态特性,通过加权k近邻同时考虑动态特征矩阵中的特征贡献率与累计互信息的影响,利用动态阈值计算降低运行状态突变造成的误报。分别以美国可再生能源实验室5 MW海上风机基准模型的常见传感器和执行器故障以及SCADA数据中记录的变桨系统故障为例,将所提方法的故障检测结果分别与PCA、KPCA、FD-kNN以及PC-kNN故障检测方法进行对比,结果表明所提方法能够准确进行故障信息的检测,所提方法优于其他对比故障检测方法。     

主分量分析和因子隐Markov模型在机械故障诊断中的应用

主分量分析(principal component analysis,PCA)是统计学中分析数据的一种有效方法,可以将高维数据空间变换到低维特征空间,因而可用于多通道冗余消除和特征提取.因子隐Markov模型是隐Markov模型的扩展,它比隐Markov模型更有优势,适用于动态过程时间序列的建模,并具有强大的时序模型分类能力,特别适合非平稳、信号特征重复再现性不佳的信号分析.文中结合主分量分析与因子隐Markov模型,提出一种新的故障识别方法,即以主分量分析方法进行冗余消除和故障特征提取,因子隐Markov模型作为分类器.并应用到机械故障诊断中,同时与基于主分量分析的隐Markov模型的识别方法相比较,实验结果表明基于PCA的因子隐Markov模型识别法和基于PCA的隐Markov模型识别法在故障识别上都是有效的,但对于相同的状态空间,前者的训练速度快于后者,尤其是状态空间越大,这种优势越明显.     

Fault detection of feed water treatment process using PCA-WD with parameter optimization

Feed water treatment process (FWTP) is an essential part of utility boilers; and fault detection is expected for its reliability improvement. Classical principal component analysis (PCA) has been applied to FWTPs in our previous work; however, the noises of T² and SPE statistics result in false detections and missed detections. In this paper, Wavelet denoise (WD) is combined with PCA to form a new algorithm, (PCA-WD), where WD is intentionally employed to deal with the noises. The parameter selection of PCA-WD is further formulated as an optimization problem; and PSO is employed for optimization solution. A FWTP, sustaining two 1000 MW generation units in a coal-fired power plant, is taken as a study case. Its operation data is collected for following verification study. The results show that the optimized WD is effective to restrain the noises of T² and SPE statistics, so as to improve the performance of PCA-WD algorithm. And, the parameter optimization enables PCA-WD to get its optimal parameters in an automatic way rather than on individual experience. The optimized PCA-WD is further compared with classical PCA and sliding window PCA (SWPCA), in terms of four cases as bias fault, drift fault, broken line fault and normal condition, respectively. The advantages of the optimized PCA-WD, against classical PCA and SWPCA, is finally convinced with the results.     

Variance sensitive adaptive threshold-based PCA method for fault detection with experimental application

Principal Component Analysis (PCA) is a statistical process monitoring technique that has been widely used in industrial applications. PCA methods for Fault Detection (FD) use data collected from a steady-state process to monitor T² and Q statistics with a fixed threshold. For the systems where transient values of the processes must be taken into account, the usage of a fixed threshold in PCA method causes false alarms and missing data that significantly compromise the reliability of the monitoring systems. In the present article, a new PCA method based on variance sensitive adaptive threshold (T_vsa) is proposed to overcome false alarms which occur in the transient states according to changing process conditions and the missing data problem. The proposed method is implemented and validated experimentally on an electromechanical system. The method is compared with the conventional monitoring methods. Experimental tests and tabulated results confirm the fact that the proposed method is applicable and effective for both the steady-state and transient operations and gives early warning to operators.     

基于动态PCA与改进SVM的航空发动机故障诊断

为了对航空发动机进行高效地故障诊断,确保飞机的飞行安全,提出了一种基于动态主元分析和改进支持向量机的航空发动机智能故障诊断方法。该方法结合了动态主元分析(principal component analysis,简称PCA)在特征提取方面和改进支持向量机（support vector machine,简称SVM）在故障诊断方面的优势。动态PCA方法对所涉及的过程变量进行去噪、降维、消除相关性等预处理和特征提取,采用改进SVM方法将所得的特征向量进行故障诊断诊断。所提出的方法可解决航空发动机模型精度和传感器测量参数有限情况下的滑油系统故障诊断精度差、效率低和易误诊、漏诊等问题。以某型真实航空发动机滑油系统为例,对提出方法的有效性进行试验验证。结果表明,采用的动态PCA和改进SVM故障诊断方法能有效提高故障诊断正确率,实现航空发动机滑油系统故障诊断的效能,具有较好的应用价值与前景。     

基于混合方法的智能终端故障检测

提出了一种基于混合方法的智能终端故障检测方法。首先介绍了所设计的智能终端故障检测方法，然后提出了一种基于连续小波变换(CWT)和BP神经网络的混合方案，该方案通过连续小波变换得到故障特征值，通过BP神经网络的学习机制对故障特征值进行自动识别和分类，最后将其应用于山火监测装置和微气象装置。结果表明，该方法的智能终端故障检测的准确性虽然会随着天气环境的恶劣情况有所降低，但仍能够始终保持在98%以上，有较高的准确性。     

Process monitoring using kernel density estimation and Bayesian networking with an industrial case study

Principal component analysis has been widely used in the process industries for the purpose of monitoring abnormal behaviour. The process of reducing dimension is obtained through PCA, while T-tests are used to test for abnormality. Some of the main contributions to the success of PCA is its ability to not only detect problems, but to also give some indication as to where these problems are located. However, PCA and the T-test make use of Gaussian assumptions which may not be suitable in process fault detection. A previous modification of this method is the use of independent component analysis (ICA) for dimension reduction combined with kernel density estimation for detecting abnormality; like PCA, this method points out location of the problems based on linear data-driven methods, but without the Gaussian assumptions. Both ICA and PCA, however, suffer from challenges in interpreting results, which can make it difficult to quickly act once a fault has been detected online. This paper proposes the use of Bayesian networks for dimension reduction which allows the use of process knowledge enabling more intelligent dimension reduction and easier interpretation of results. The dimension reduction technique is combined with multivariate kernel density estimation, making this technique effective for non-linear relationships with non-Gaussian variables. The performance of PCA, ICA and Bayesian networks are compared on data from an industrial scale plant.     

Design and application of exponential chart for monitoring time-between-events data under random process shift

Exponential charts based on time-between-events (TBE) data were developed for monitoring high-yield process like the process which has achieved six-sigma quality level and has recently shown to be very useful in manufacturing systems, in reliability and maintenance monitoring, and also in service-related applications in general. This article develops an economic model of the exponential chart (known as TBE_random chart) for monitoring time-between-events data; the design algorithm considers the random characteristic of the process shifts and therefore better reflects the real process conditions. The probability distribution of the random process shift is modeled by a Rayleigh distribution based on the sample data acquired during the operation of the control chart. The design of the proposed control chart scheme is demonstrated, and the properties are compared with that of other exponential charts. The results of the numerical studies show that the consideration of the random process shift in designing an exponential chart provides an excellent in-control stability of the charting scheme, which in turn helps in saving time and cost for searching the false alarms. The proposed control chart is easy to understand and operate, and thus the floor operators can utilize and understand it as easily as with a traditional charting scheme.     

机电系统BIT的非永久故障分析建模与诊断

非永久故障是导致机电系统BIT（built-in test）虚警的一个主要原因,诊断非永久故障既可保证BIT的高故障检测率,同时又可有效地抑制虚警。但是目前缺乏对非永久故障的机理分析与建模,相应的诊断研究也很少。在分析研究非永久故障的表现形式、产生原因与机理的基础上,对被测系统的状态进行马尔可夫建模,再根据被测系统和隐马尔可夫模（hidden Markov model,HMM）的状态都是通过表现来感知的特点,利用HMM对BIT被测系统建模,并提出基于HMM的BIT非永久故障诊断方法,最后通过实验验证表明,此方法能有效地诊断非永久故障,降低BIT虚警。     

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。     

NONLINEAR DATA RECONCILIATION METHOD BASED ON KERNEL PRINCIPAL COMPONENT ANALYSIS

In the industrial process situation, principal component analysis (PCA) is a general method in data reconciliation. However, PCA sometime is unfeasible to nonlinear feature analysis and limited in application to nonlinear industrial process. Kernel PCA (KPCA) is extension of PCA and can be used for nonlinear feature analysis. A nonlinear data reconciliation method based on KPCA is proposed. The basic idea of this method is that firstly original data are mapped to high dimensional feature space by nonlinear function, and PCA is implemented in the feature space. Then nonlinear feature analysis is implemented and data are reconstructed by using the kernel. The data reconciliation method based on KPCA is applied to ternary distillation column. Simulation results show that this method can filter the noise in measurements of nonlinear process and reconciliated data can represent the true information of nonlinear process.     

基于相对变换PLS的故障检测方法

针对偏最小二乘方法(partial least squares,PLS)在无量纲标准化处理后导致的特征值大小近似相等,难以获得代表性的潜变量等问题,提出了一种基于相对变换PLS( relative-transformation PLS,RTPLS)的故障检测方法.该方法引入马氏距离相对变换理论,通过计算采样数据之间的马氏距离,将原始空间数据变换到相对空间.然后在相对空间进行PLS分解,提取有代表性的潜变量,建立故障检测模型,实现采样数据的在线检测.通过对TE (Tennessee Eastman)过程故障和轧钢机系统力传感器故障的仿真实验验证了所提出方法的有效性和实用性.理论分析和仿真实验均表明,基于RTPLS的故障检测方法能有效地消除量纲的影响,提取具有更大的变化度和代表性的隐变量,增加故障检测的精度和实时性.     

基于三层信息融合的提升机制动系统故障诊断

为了充分利用矿井提升机运行过程中的监测历史数据,判定故障原因并进行准确定位,将信息融合技术引入提升机的故障诊断中,提出了一种基于三层多源信息融合的故障诊断方法。该方法依据主成分分析法(principal component analysis,简称PCA)建立主元模型,对原始完备数据集进行降维去噪,实现特征提取,完成数据层的融合;特征层采用具有记忆功能的Elman神经网络作为融合算法,不断调整权值对数据层各信息源提取出来的特征进行训练,通过压缩融合信息量完成时间上的融合;决策层使用DS(dempster-shafer,简称DS)证据理论对特征层训练输出的信息进行融合,判定故障原因,实现了空间上的融合;最后依据PCA故障诊断原理,定位故障发生的部位,完成诊断过程。该融合方法通过对监测系统所测信息的合理选择、综合与利用,对其进行空间和时间上的融合互补。试验结果表明,该方法能够充分利用大量历史数据对系统进行诊断,可以显著提高系统的可靠性。     

正交试验下基于FMMESA和GRA的环境应力-故障关联分析

环境应力是导致机电系统故障的主要原因,环境应力因素与机电系统故障间存在复杂的关联关系。首先,在故障模式、机理及影响分析方法的基础上,提出故障模式、机理、环境应力分析方法,用于定性分析导致机电系统故障的环境应力因素。其次,以定性分析结论为指导,应用加权评判法和正交试验设计法优化设计环境应力试验方案。然后,以试验数据为基础,应用灰色系统理论,提出环境应力特征与故障灰色关联分析方法,用于定量分析环境应力及其特征与故障的关联关系。最后,进行案例应用与分析。实验结果表明,所提出的方法能在优化后的试验次数情况下定性并定量的分析出影响系统故障的主要环境应力及其特征,可用于指导机电系统测试性虚拟试验中的环境建模、环境应力等级评定等,还可以作为故障检测、诊断与预测的信息基础。