自动编码器与典型相关分析方法联合驱动的工业过程质量监测

本文将自动编码器(AE)特征提取方法和典型相关分析方法(CCA)有机结合,提出了一种联合驱动的质量监测模型及其质量相关的故障检测方法.首先,利用AE算法对输入样本进行无监督自动学习和重构,实现数据的特征提取和降维;其次,利用CCA算法实现特征与质量变量关联最大化,建立质量变量与特征变量的关系模型;根据监测模型的潜结构投影,构建T2统计量和SPE统计量及其相应控制限.将提出的方法用于分析带钢热连轧过程现场实际数据,结果表明,基于自动编码器-典型相关分析方法(AE-CCA)的质量监测方法能够准确的检测出故障,并且检测效果优于传统的核典型相关分析(KCCA)算法.     

Parallel quality-related dynamic principal component regression method for chemical process monitoring

Traditional quality-related process monitoring mainly focuses on the magnitude change of the quality variables caused by additive faults. However, the abnormal fluctuations in the quality variables caused by multiplicative faults are often overlooked. In this paper, a novel parallel dynamic principal component regression (P-DPCR) algorithm is proposed to monitor the changes in the magnitude and fluctuation of the quality variables simultaneously. Firstly, in order to eliminate the interference of quality-unrelated variables, the quality-related process variables are selected on the basis of correlation analysis. Secondly, the dynamic extension and moving window are carried out for process variables and quality variables, in which the dynamic variables space (called X-space/Y-space) and the variance space (called VX-space/VY-space) are constructed. Afterwards, double quality-related statistics based on the regression model of these four spaces are given, and the comprehensive monitoring decision can be obtained. Finally, two numerical cases and the Tennessee Eastman process are used to show the effectiveness of the proposed method.     

Weighted part mutual information related component analysis for quality-related process monitoring

Industrial products have become the core of today’s highly competitive international society, but quality-related faults happened in practical industrial processes heavily affect product quality. In this paper, we will consider the problem of the detection of quality-related faults. Inspired by part mutual information (PMI), we develop a process monitoring method called weighted PMI based related component analysis (WPMI-RCA). Firstly, combining PMI and Bayesian weighted fusion, process variables strongly related to quality are selected with the supervision of multi-quality indicators. Then, the selected variables are modeled by related component analysis (RCA) and thus orthogonal related components (RCs) containing the main information of quality variations can be obtained. The process data space can be divided into two subspaces and the monitoring statistics are developed for the quality-related fault detection. Finally, the validity of WPMI-RCA is demonstrated by a numerical example and the benchmark Tennessee Eastman process (TEP). The proposed method can improve the detection rates of quality-related faults and significantly reduce the nuisance detections. It may be helpful to improve the management efficiency for practical industrial processes.     

Nonlinear process monitoring using kernel dictionary learning with application to aluminum electrolysis process

In practice, because of complex mechanism processes, such as heating process, volume heterogeneity, and various chemical reaction characteristics, there is a nonlinear relationship among variables in industrial systems. The nonlinearity brings some difficulties to process monitoring. In order to ensure that the process monitoring system can work normally in nonlinear production processes, the nonlinear relationship between variables ought to be considered. In this work, a new fault detection and isolation method based on kernel dictionary learning is presented. In detail, the linearly inseparable data is mapped to a high-dimensional space. Then, a new nonlinear dictionary learning method based on kernel method was proposed to learn the dictionary. After obtaining the dictionary, the control limit can be calculated from the training data according to the kernel density estimation (KDE) method. When new data arrive, they can be represented by the well-learned dictionary, and the kernel reconstruction error can be used as a classifier for process monitoring. As for the fault data, the iterative reconstruction based method is proposed for fault isolation. In order to evaluate the effectiveness of the proposed process monitoring method, some extensive experiments on a numerical simulation, the continuous stirred tank heater (CSTH) process, and a real industrial aluminum electrolysis process are conducted. The proposed method is compared with several state-of-the-art process monitoring methods and the experimental results show that the proposed method can provide satisfactory monitoring results, especially for some small faults, thus it is suitable for process monitoring of nonlinear industrial processes.     

Nonlinear process monitoring based on kernel global–local preserving projections

A new nonlinear dimensionality reduction method called kernel global–local preserving projections (KGLPP) is developed and applied for fault detection. KGLPP has the advantage of preserving global and local data structures simultaneously. The kernel principal component analysis (KPCA), which only preserves the global Euclidean structure of data, and the kernel locality preserving projections (KLPP), which only preserves the local neighborhood structure of data, are unified in the KGLPP framework. KPCA and KLPP can be easily derived from KGLPP by choosing some particular values of parameters. As a result, KGLPP is more powerful than KPCA and KLPP in capturing useful data characteristics. A KGLPP-based monitoring method is proposed for nonlinear processes. T² and SPE statistics are constructed in the feature space for fault detection. Case studies in a nonlinear system and in the Tennessee Eastman process demonstrate that the KGLPP-based method significantly outperforms KPCA, KLPP and GLPP-based methods, in terms of higher fault detection rates and better fault sensitivity.     

基于高效核偏最小二乘的质量相关故障检测

核偏最小二乘(KPLS)是一种多元统计方法, 广泛应用于过程监控, 然而, KPLS采用斜交分解, 导致质量相关空间存在冗余信息易引发误报警. 因此, 本文提出了高效核偏最小二乘(EKPLS)模型, 所提方法通过奇异值分解(SVD)将核矩阵正交分解为质量相关空间和质量无关空间, 有效降低质量相关空间中的冗余信息, 并采用主成分分析(PCA)按方差大小将质量相关空间分解为质量主空间和质量次空间. 此外, 为进一步降低由质量无关故障引发的误报警, 提出基于质量估计的正交信号修正(OSC)预处理方法, 并结合EKPLS模型提出了OSC-EKPLS算法. OSCEKPLS通过质量估计值对被测数据进行OSC预处理, 降低了计算复杂度和误报率. 最后, 通过数值仿真和田纳西–伊斯曼过程验证了OSC-EKPLS具有良好的故障检测性和更低的误报率.     

Nonlinear process monitoring and fault isolation using extended maximum variance unfolding

Kernel principal component analysis (KPCA) has become a popular technique for process monitoring, owing to its capability of handling nonlinearity. Nevertheless, KPCA suffers from two major disadvantages. First, the underlying manifold structure of data is not considered in process modeling. Second, the selection of kernel parameters is problematic. To avoid such deficiencies, a manifold learning technique named maximum variance unfolding (MVU) is considered as an alternative. However, such method is merely able to deal with the training data, but has no means to handle new samples. Therefore, MVU cannot be applied to process monitoring directly. In this paper, an extended MVU (EMVU) method is proposed, extending the utilization of MVU to new samples by approximating the nonlinear mapping between the input space and the output space with a Gaussian process model. Consequently, EMVU is suitable to nonlinear process monitoring. A cross-validation algorithm is designed to determine the dimensionality of the EMVU output space. For online monitoring, three different types of monitoring indices are developed, including squared prediction error (SPE), Hotelling-T², and the prediction variance of the outputs. In addition, a fault isolation algorithm based on missing data analysis is designed for EMVU to identify the variables contributing most to the faults. The effectiveness of the proposed methods is verified by the case studies on a numerical simulation and the benchmark Tennessee Eastman (TE) process.     

基于特征样本的KPCA在故障诊断中的应用

核函数主元分析（KPCA）可用于非线性过程监控.建立KPCA模型首先要计算核矩阵K,K的维数等于训练样本的数量,对于大样本集,计算K很困难.对此提出一种基于特征样本的KPCA（SKPCA）,其基本思想是,首先利用非线性映射函数将输入空间映射到特征子空间,然后在特征子空间中计算主元.将SKPCA应用于监控Tennessee Eastman过程,并与基于全体样本的KPCA作比较,仿真结果显示,二者诊断结果基本相同,然而特征样本只是训练样本中的一小部分,因此减少了K的维数,解决了K的计算问题.     

基于核典型相关性-熵成分分析的工业过程质量监测方法

工业过程多变量、数据高维度和非线性的特点使得对其质量监测及质量相关的故障诊断变得复杂.融合核熵成分分析(KECA)及典型相关分析(CCA)方法的思想,进行特征提取降维的同时确保所提取特征与质量变量的最大相关性,提出一种新的质量相关的工业过程故障检测方法.首先,采用KECA对输入数据进行核空间的映射及特征提取,同时融合CCA算法思想使得所提取特征与质量变量间关联最大化;然后,构建监测统计量并用Parzen窗估计其控制限,用于过程的故障检测;最后,运用所提方法对带钢热连轧工业过程实际生产数据进行分析,并与其他4种传统非线性算法对比分析,实验结果验证了所提方法的准确性、有效性及先进性.     

基于核主元分析与核密度估计的非线性过程故障监测与识别

在针对将核主元分析(kernel principal components analysis, KPCA)与基于高斯分布的控制限(control limits,CLS)相结合会降低其性能的问题,提出了一种基于核主元分析与核密度估计(kernel principal components analysiskernel density estimation, KPCA-KDE)相结合的非线性过程故障监测与识别方法.该方法采用核密度估计(kernel density estimation, KDE)技术来估计基于KPCA的非线性过程监控的CLS.通过研究KPCA和KPCA-KDE所有20个故障的检出率发现,与相应的基于高斯分布的方法进行比较, KDE具有较高的故障检出率;此外,基于KDE的检测延迟等于或低于其他方法.通过改变带宽和保留的主元数量进行故障检测, KPCA记录的FAR值较高,相反, KPCA-KDE方法仍然没有记录任何假报警.在田纳西伊斯曼过程(Tennessee Eastman, TE)上的应用表明,KPCA-KDE比基于高斯假设的CLS的KPCA在灵敏度和检测时间上都具有更好...     

Information concentrated variational auto-encoder for quality-related nonlinear process monitoring

As the deep learning technology develops, many process monitoring methods based on auto-encoder (AE) are designed for the nonlinear industrial processes. However, these methods mainly focus on process variables and ignore the quality indicator which is crucial for the final production. To extract the latent variables which represent both process information and quality information, this paper proposes a novel algorithm named information concentrated variational auto-encoder (IFCVAE). To concentrate the quality-related information, a loading matrix regularization based on mutual information is designed, so that the strongly quality-related variables tend to have larger weights in the loading matrix. In addition, to monitor processes from the quality-related and unrelated aspects, IFCVAE decomposes the original space into two subspaces that are mutually orthogonal based on variational auto-encoder (VAE). With the help of an additional regression network, the two subspaces can correspond to the quality-related and unrelated spaces. For process monitoring, two statistics are designed for the subspaces according to Kullback–Leibler divergence. Finally, the effectiveness of IFCVAE is demonstrated by a numerical case and an industrial case.     

Nonlinear process monitoring based on maximum variance unfolding projections

Kernel principal component analysis (KPCA) has recently proven to be a powerful dimensionality reduction tool for monitoring nonlinear processes with numerous mutually correlated measured variables. However, the performance of KPCA-based monitoring method largely depends on its kernel function which can only be empirically selected from finite candidates assuming that some faulty process samples are available in the off-line modeling phase. Moreover, KPCA works at high computational cost in the on-line monitoring phase due to its dense expansions in terms of kernel functions. To overcome these deficiencies, this paper proposes a new process monitoring technique comprising fault detection and identification based on a novel dimensionality reduction method named maximum variance unfolding projections (MVUP). MVUP firstly applies the recently proposed manifold learning method maximum variance unfolding (MVU) on training samples, which can be seen as a special variation of KPCA whose kernel matrix is automatically learned such that the underlying manifold structure of training samples is “unfolded” in the reduced space and hence the boundary of distribution region of training samples is preserved. Then MVUP uses linear regression to find the projection that best approximates the implicit mapping from training samples to their lower dimensional embedding learned by MVU. Simulation results on the benchmark Tennessee Eastman process show that MVUP-based process monitoring method is a good alternative to KPCA-based monitoring method.     

基于自适应混合核典型变量分析的工业过程质量相关故障检测

质量相关故障检测技术是保障工业过程安全顺行和质量稳定的重要手段,是当前流程工业过程控制领域的研究热点.针对工业过程的非线性与动态特性及其质量相关故障的时变特性,提出一种基于自适应混合核典型变量分析(AMKCVA)的质量相关故障检测方法.该方法通过设计合理的混合核函数和自适应监测统计量,提升了工业过程质量相关故障的检测性...     

Quality-relevant fault detection based on adversarial learning and distinguished contribution of latent variables to quality

Quality-relevant fault detection is a primary task to reveal the changes of quality variables in process monitoring. Current works mainly focus on learning quality-relevant features, however, how to distinguish quality-relevant and irrelevant information is responsible for the excellent monitoring performance. In this study, a novel quality-relevant fault detection method is proposed on the basis of adversarial learning and distinguished contribution of latent features to quality is originally introduced. First of all, we map the input variables into a gaussian manifold in adversarial and unsupervised manner. Then a fully connected neural network is trained to learn the relationship between latent and quality variables. To distinguish necessary information in such manifold, the Jacobi operator at the corresponding point is calculated to project the latent variables into quality-relevant and quality-irrelevant subspaces. Third, fault detection is implemented in these dynamic subspaces using the probabilities of latent variables. Finally, the proposed method is evaluated by numerical example, the Tennessee-Eastman process and wind turbine blade icing process.     

Latent Variable Regression for Supervised Modeling and Monitoring

A latent variable regression algorithm with a regularization term(r LVR) is proposed in this paper to extract latent relations between process data X and quality data Y. In rLVR,the prediction error between X and Y is minimized, which is proved to be equivalent to maximizing the projection of quality variables in the latent space. The geometric properties and model relations of rLVR are analyzed, and the geometric and theoretical relations among r LVR, partial least squares, and canonical correlation analysis are also presented. The rLVR-based monitoring framework is developed to monitor process-relevant and quality-relevant variations simultaneously. The prediction and monitoring effectiveness of rLVR algorithm is demonstrated through both numerical simulations and the Tennessee Eastman(TE) process.     

Design teacher and supervised dual stacked auto-encoders for quality-relevant fault detection in industrial process

Current fault detection methods based on deep neural networks only consider process information and ignore quality indicators. In order to obtain features representing both process variables and quality indicators efficiently, this paper designs teacher and supervise dual stacked auto-encoder (TSSAE) for quality-relevant fault detection in industrial process which separates the feature extraction and model construction. To separate the feature extraction and model construction, a mixing stacked auto-encoder which consists of a nonlinear encoder and a linear decoder is designed to extract features of process variables and quality indicators. Another encoder is supervised by the extracted features and further predict the process variables and quality indicators only from process variables. Then quality-relevant, quality-irrelevant and residual subspaces are constructed in a linear way and fault detection is implemented in these subspaces based on Euclidean distance and kernel density estimation. Finally, the effectiveness of TSSAE is evaluated by a numerical example and the Tennessee-Eastman process.     

Quality relevant nonlinear batch process performance monitoring using a kernel based multiway non-Gaussian latent subspace projection approach

Multiway kernel partial least squares method (MKPLS) has recently been developed for monitoring the operational performance of nonlinear batch or semi-batch processes. It has strong capability to handle batch trajectories and nonlinear process dynamics, which cannot be effectively dealt with by traditional multiway partial least squares (MPLS) technique. However, MKPLS method may not be effective in capturing significant non-Gaussian features of batch processes because only the second-order statistics instead of higher-order statistics are taken into account in the underlying model. On the other hand, multiway kernel independent component analysis (MKICA) has been proposed for nonlinear batch process monitoring and fault detection. Different from MKPLS, MKICA can extract not only nonlinear but also non-Gaussian features through maximizing the higher-order statistic of negentropy instead of second-order statistic of covariance within the high-dimensional kernel space. Nevertheless, MKICA based process monitoring approaches may not be well suited in many batch processes because only process measurement variables are utilized while quality variables are not considered in the multivariate models. In this paper, a novel multiway kernel based quality relevant non-Gaussian latent subspace projection (MKQNGLSP) approach is proposed in order to monitor the operational performance of batch processes with nonlinear and non-Gaussian dynamics by combining measurement and quality variables. First, both process measurement and quality variables are projected onto high-dimensional nonlinear kernel feature spaces, respectively. Then, the multidimensional latent directions within kernel feature subspaces corresponding to measurement and quality variables are concurrently searched for so that the maximized mutual information between the measurement and quality spaces is obtained. The I² and SPE monitoring indices within the extracted latent subspaces are further defined to capture batch process faults resulting in abnormal product quality. The proposed MKQNGLSP method is applied to a fed-batch penicillin fermentation process and the operational performance monitoring results demonstrate the superiority of the developed method as apposed to the MKPLS based process monitoring approach.     

An efficient KPCA algorithm based on feature correlation evaluation

Classic kernel principal component analysis (KPCA) is less computationally efficient when extracting features from large data sets. In this paper, we propose an algorithm, that is, efficient KPCA (EKPCA), that enhances the computational efficiency of KPCA by using a linear combination of a small portion of training samples, referred to as basic patterns, to approximately express the KPCA feature extractor, that is, the eigenvector of the covariance matrix in the feature extraction. We show that the feature correlation (i.e., the correlation between different feature components) can be evaluated by the cosine distance between the kernel vectors, which are the column vectors in the kernel matrix. The proposed algorithm can be easily implemented. It first uses feature correlation evaluation to determine the basic patterns and then uses these to reconstruct the KPCA model, perform feature extraction, and classify the test samples. Since there are usually many fewer basic patterns than training samples, EKPCA feature extraction is much more computationally efficient than that of KPCA. Experimental results on several benchmark data sets show that EKPCA is much faster than KPCA while achieving similar classification performance.     

基于KPCA和CPSO的故障检测方法

提出一种基于核主元分析(KPCA)和混沌粒子优化群(CPSO)算法的非线性故障检测方法。通过核函数完成非线性变换,将变量由非线性的输入空间转换到线性的特征空间来计算主元,构造平方预测误差统计量检测故障是否发生。为避免粒子群算法的早熟现象,利用混沌优化的搜索特性,将CPSO算法应用到KPCA核参数的优化中。变压器故障检测结果表明,与基于PCA、KPCA和 PSO-KPCA的故障检测方法相比,该方法的检测正确率较高。