Supervised process monitoring and fault diagnosis based on machine learning methods

This paper presents a novel Gabor-based kernel Principal Component Analysis (PCA) method by integrating the Gabor wavelet representation of face images and the kernel PCA method for face recognition. Gabor wavelets first derive desirable facial features characterized by spatial frequency, spatial locality, and orientation selectivity to cope with the variations due to illumination and facial expression changes. The kernel PCA method is then extended to include fractional power polynomial models for enhanced face recognition performance. A fractional power polynomial, however, does not necessarily define a kernel function, as it might not define a positive semidefinite Gram matrix. Note that the sigmoid kernels, one of the three classes of widely used kernel functions (polynomial kernels, Gaussian kernels, and sigmoid kernels), do not actually define a positive semidefinite Gram matrix either. Nevertheless, the sigmoid kernels have been successfully used in practice, such as in building support vector machines. In order to derive real kernel PCA features, we apply only those kernel PCA eigenvectors that are associated with positive eigenvalues. The feasibility of the Gabor-based kernel PCA method with fractional power polynomial models has been successfully tested on both frontal and pose-angled face recognition, using two data sets from the FERET database and the CMU PIE database, respectively. The FERET data set contains 600 frontal face images of 200 subjects, while the PIE data set consists of 680 images across five poses (left and right profiles, left and right half profiles, and frontal view) with two different facial expressions (neutral and smiling) of 68 subjects. The effectiveness of the Gabor-based kernel PCA method with fractional power polynomial models is shown in terms of both absolute performance indices and comparative performance against the PCA method, the kernel PCA method with polynomial kernels, the kernel PCA method with fractional power polynomial models, the Gabor wavelet-based PCA method, and the Gabor wavelet-based kernel PCA method with polynomial kernels.

In the recognition of digital modulation signals, extracting the characteristics that more easily distinguish the source signals, and have better performance of the signal, the source signals can be identified by using these characteristics. This paper optimized the original 20 features using orthogonal experiment method, and then identified the signals by the neural network, finally compared its results with that of PCA(principal component analysis) and KPCA(kernel principal component analysis) methods. Experimental results show that orthogonal experiment method in Gaussian and multipath environment can optimize the selection of the features to achieve higher recognition rate than the original features. Orthogonal experiment method has a better ability than PCA and KPCA methods in the optimal choice of these 20 characteristics.

Kernel principal component analysis (KPCA) is a method for performing a nonlinear form of principal component analysis (PCA). It involves nonlinear mapping of the input data onto a high dimensional feature space and performing PCA there. Through KPCA, nonlinear relations in the input space are detected as linear relations in the feature space. Furthermore, explicit computation of the mapping is not necessary in KPCA. Only dot products of the mapped data items in the feature space are required which are obtained directly from the input data using a kernel function. Moreover, control limits of traditional statistical process monitoring indices are usually determined under assumption of normal (Gaussian) distribution. Although, it is recognized that this is inappropriate in nonlinear processes, the impact of using kernel density estimation (KDE), which is a nonparametric method, to determine control limits in KPCA-based process monitoring has not been reported. This paper seeks to bridge this gap. The KPCA with KDE approach is applied to the Tennessee Eastman process for the detection of faults. The results confirm that associating KPCA with kernel density estimated control limits provides better monitoring performance than using control limits based on the normal probability density function.

Fault discriminant enhanced kernel principal component analysis incorporating prior fault information for monitoring nonlinear processes

Locality Preserving Projections (LPP) is a linear projective map that optimally preserves the neighborhood structure of the data set. Though LPP has been applied in many fields, it has limits to solve recognition problem. Thus, a new palmprint recognition method is proposed based on Kernel Locality Preserving Projections (KLPP). Different from LPP method, KLPP not only describes the nonlinear correlations between pixels, but also preserves the local structure of the palmprint image space. In this way, the unwanted variations resulting from in lighting may be eliminated or reduced. We compare our proposed approach with Principal Component Analysis (PCA), LPP and Kernel Principal Component Analysis (KPCA) methods on PolyU palmprint database. Experiment results demonstrate that KLPP achieves better recognition rate as the dimension of the palmprint subspace changes.

Machine learning algorithms play an important role in fault detection and fault diagnosis of gas sensor arrays. Because the gas sensor array will see stability degradation and a shift in output signal amplitude under long-term operation, it is very important to detect the abnormal output signal of the gas sensor array in time and achieve accurate fault location. In order to solve the problem of low detection accuracy of micro-faults in gas sensor arrays, this paper adopts the serial principal component analysis (SPCA) method, which combines the advantages of principal component analysis (PCA) in the linear part and the advantages of kernel principal component analysis (KPCA) in the nonlinear part. The experimental results show that this method is more sensitive to micro-faults and has better fault detection accuracy than the fault detection methods of PCA and KPCA. In addition, in order to solve the current problem of low accuracy of multiple-fault isolation, a SPCA-based reconstruction contribution fault isolation method is proposed in this paper. The experimental results show that this method has higher fault isolation accuracy than the method based on contribution graph.

In this paper, we develop an improved fault detection (FD) technique in order to enhance the monitoring abilities of nonlinear biological processes. Generalized likelihood ratio test (GLRT)-based kernel principal component analysis (KPCA) (called also kernel GLRT) is an effective data-driven technique for monitoring nonlinear processes. However, it is well known that the data collected from complex and multivariate processes are multiscale due to the variety of changes that could occur in process with different localization in time and frequency. Thus, to enhance the process monitoring abilities, we propose to combine the advantages of kernel GLRT and multiscale representation using wavelets by developing a multiscale kernel GLRT (MS-KGLRT) detection chart. The proposed fault detection approach is addressed so that the KPCA is used to compute the model in the feature space and the MS-KGLRT chart is applied to detect the faults. The detection performance of the new chart is studied using two examples, one using synthetic data and the other using biological process representing a Cad System in E. Coli (CSEC) model for detecting small and moderate shifts (offset or bias and drift). The MS-KGLRT chart is used to enhance fault detection of the CSEC model through monitoring some of the key variables involved in this model such as enzymes, lysine, and cadaverine.

The practical application of traditional data‐driven techniques for process monitoring encounters significant challenges due to the inherent nonlinear and dynamic nature of most industrial processes. Aiming at the problem of nonlinear dynamic process monitoring, a novel fault detection method based on dynamic kernel principal component analysis combined with weighted structural difference (DKPCA‐WSD) is proposed in this paper. Initially, the proposed method leverages a sophisticated nonlinear transformation to project the augmented matrix of the original input data into a high‐dimensional feature space, thereby facilitating the establishment of a DKPCA model. Subsequently, the WSD statistic is computed, utilizing a widely known sliding window technique, to quantify the mean and standard deviation differences across data structures. Ultimately, the WSD statistic is utilized for fault detection, completing the process monitoring task. By integrating the capability of DKPCA to capture nonlinear dynamic characteristics with the effectiveness of the WSD statistic in mitigating the impact of non‐Gaussian data distributions, DKPCA‐WSD significantly enhances the monitoring performance of traditional DKPCA in nonlinear dynamic processes. The proposed method is evaluated through a numerical case exhibiting nonlinear dynamic behaviors and a simulation model of a continuous stirred tank reactor. A comparative analysis with conventional methods, including principal component analysis (PCA), dynamic principal component analysis, KPCA, PCA similarity factor (SPCA), DKPCA, and moving window KPCA (MWKPCA), demonstrates that DKPCA‐WSD outperforms traditional fault detection techniques in nonlinear dynamic processes, offering a substantial improvement in monitoring performance.

Viewing to monitoring the continuous casting process, a nonlinear fault detection method based on kernel principal component analysis (KPCA) was introduced. KPCA can efficiently compute principal components in high-dimensional feature spaces by means of integral operators and nonlinear kernel functions, which is to first map the input space into a feature space via nonlinear mapping and then to compute the principal components in that feature space. Based on T2 and SPE charts in feature space, principal component analysis(PCA)can be used to detect faults.. The simulation results show that the proposed approach effectively captured the nonlinear relationship in the process variables and showed superior process monitoring performance compared to linear PCA.

Mode changes in today's sophisticated industrial processes result in unanticipated shutdowns, which could reduce the lifespan of critical machinery and result in high maintenance costs. A technique that can identify the fault of equipment operating in various modes is needed to prevent this issue. In light of this, we suggest a novel fault detection technique that makes use of the k-nearest neighbor normalization-based weight local outlier factor (WLOF). The suggested approach carries out local normalization using neighbors to take into account potential mode changes in the normal data, and WLOF is used for fault detection. In contrast to statistical methods, such as principal component analysis (PCA) and independent component analysis (ICA), the local outlier factor (LOF) uses the density of neighbors. However, conventional LOF is greatly affected by the distance between its neighbors, performance may deteriorate when faulty data is adjacent to normal data. To improve the defect detection performance of LOF, the proposed method multiplies proportionally by the distance between each neighbor. A circulating fluidized bed boiler and a multimode numerical case were used to assess the effectiveness of the suggested method. The results of the experiments demonstrate that the suggested method performs better than conventional PCA, kernel PCA (KPCA), k-nearest neighbor (kNN), and LOF. In particular, compared to conventional techniques, the suggested method increased detection accuracy by 20%. Therefore, the proposed technique is applicable to a real process operating in multiple modes.

This paper presents a novel algorithm - robust kernel principal component analysis (robust KPCA), on the basis of the research of kernel principal component analysis (KPCA) and robust principal component analysis (RPCA). First, this algorithm sets the radius of the images of the training samples in the feature space using kernel tricks, then determines whether the samples are outliers or not, and finally analyzes the training samples which have eliminated the outliers using KPCA algorithm. The improved KPCA algorithm not only retains the non-linearity property of KPCA algorithm but also gets better robustness. Because the effects of outliers are eliminated, robust KPCA algorithm gets higher accuracy than KPCA algorithm for data analysis. The simulation experiments show that the robust KPCA algorithm developed is better than the KPCA algorithm.

We review a neural network implementation of the statistical technique of Principal Component Analysis (PCA) and Factor Analysis. We now derive a new method based on Kernel Principal Components Analysis (KPCA) and extend the Kernel PCA method to sparsified Kernel PCA. We then apply two methods to the data set which is composed of 10 faces in a mixture of poses. We wish to identify only the most significant poses on a data set. We found the better result from the sparsified Kernel PCA method.

In this paper, fault detection methods based on principal component analysis (PCA) have been investigated. The methods have been evaluated using the data acquired from a practical nuclear power plant simulation platform. PCA, dynamic principal component analysis (DPCA), kernel principal component analysis (KPCA), two-step principal component analysis (TS-PCA) have been considered. The performance of these methods in fault detection has been compared.

It is supposed that there exist means to generate material microstructure representations by varying processing parameters numerically [1] or experimentally [2], the goal of this work is to establish a parameterized geometrical description model from learning the set of given snapshot instances. In this work, we propose a parameterization model for the representation of the Representative Volume Elements (RVE) of composite materials based on Kernel Principal Component Analysis (KPCA) method. The set of synthetic RVE snapshots governed by a series of control parameters is firstly mapped into a high-dimensional feature space. Then, Principal Component Analysis (PCA) is performed, together with the establishment of approximated response surfaces of the retained PCA projection coefficients. We showcase the performance of KPCA-based surrogate by applying it for an optimal design of the effective mechanical properties of a two-phase composite material.

A deep kernel method for lithofacies identification using conventional well logs