Sensor Fault Detection and Identification Method with KPCA in the Process of Aero-Engine Ground Testing

In this paper, a novel sensor fault detection and identification technique based on kernel principal component analysis (KPCA) and its fast data reconstruction is presented. Although it has been proved that KPCA shows a better performance for sensor fault detection, the fault identification method has rarely been developed. Using the fast data reconstruction based on distance constraint, we employ the residuals of variables to identify the faulty sensor. Since the proposed method does not include iterative calculation, it has a lower calculation burden and is more suitable for online application. The simulation results show that the proposed method effectively identifies the source of typical sensor faults.

This paper investigates the fault detection problem of instruments in process industry. Considering the difficulty of fault identification and the problems of multivariable and large computation complexity based on traditional kernel principal component analysis (KPCA), this paper presents a new method for fault detection and identification, which combines the coupling analysis with kernel principal component for multivariable fault detection and employed the local outlier factor (LOF) for multivariable fault identification. The new method consists of three parts. Firstly, according to nonlinear correlation of multivariable, coupling analysis and module division of variables based on detrended cross-correlation analysis (DCCA) are considered to reduce false alarm rate (FAR) and missed detection rate (MDR) in fault detection and identification. Secondly, KPCA is employed to detect fault in each sub-module of variables. Finally, for the sub-module which has the fault detected in second step, the LOF is adopted to calculate abnormal contribution of each variable in sub-modules to realize fault identification. To prove that the new method has the better capability of processing multivariable fault detection and the more accuracy rate on fault detection and identification than the conventional methods of KPCA, a case study on Tennessee process is carried out at the end.

Fault detection, identification and reconstruction of sensors in nuclear power plant with optimized PCA method

Detection, isolation and identification of multiple actuator and sensor faults in nonlinear dynamic systems: Application to a waste water treatment process

Sensor fault detection and isolation of an industrial gas turbine using partial adaptive KPCA

Principal Components Analysis (PCA) is used to classify the running condition of a machine by means of projecting the original data to the Principal Components space. However, if the data are concentrated in a nonlinear subspace, PCA will fail to work well. Kernel Principal Components Analysis (KPCA) transforms the input data from the original input space into a higher dimensional feature space with the nonlinear mapping, and then uses the nonlinear principal components to realize the classification. In this paper a case of gear fault diagnosis was studied with KPCA. The characteristic values of frequent domain from vibration signals of the gearbox under the running condition were extracted, and the KPCA method was used to classify gear crack fault. The result shows that KPCA is more effective to distinguish the state of the gear and more suitable to diagnose the gear faults in early stage.

Summary The principal component analysis (PCA) is exists the problem of lack of probability model and the absence of higher-order statistics information. Therefore，in order to overcome its shortcomings，this paper studies the method that can overcome two disadvantages of the principal component analysis (PCA) — the probability kernel principal component analysis (PKPCA) which is based on Bayesian theory and kernel principal component analysis (KPCA). First，the sample data are mapped to the high dimensional feature space，then define probability model of the data in high-dimensional space，and finally，expectation maximization (EM) estimated is used to get the best results. This method has both the advantage of probability analysis and kernel principal component analysis (KPCA). It is able to effectively adapt to more complex reservoir conditions and can realize the non-linear probability analysis. The probability of kernel principal component analysis (PKPCA) is applied to reservoir prediction of the oilfield. The predicted results show that the method can improve the precision attribute optimization，while improving the accuracy of the forecasts of reservoir.

Fault diagnosis (FD), comprising fault detection, isolation, identification and accommodation, enables structural health monitoring (SHM) systems to operate reliably by allowing timely rectification of sensor faults that may cause data corruption or loss. Although sensor fault identification is scarce in FD of SHM systems, recent FD methods have included fault identification assuming one sensor fault at a time. However, real-world SHM systems may include combined faults that simultaneously affect individual sensors. This paper presents a methodology for identifying combined sensor faults occurring simultaneously in individual sensors. To improve the quality of FD and comprehend the causes leading to sensor faults, the identification of combined sensor faults (ICSF) methodology is based on a formal classification of the types of combined sensor faults. Specifically, the ICSF methodology builds upon long short-term memory (LSTM) networks, i.e. a type of recurrent neural networks, used for classifying ‘sequences’, such as sets of acceleration measurements. The ICSF methodology is validated using real-world acceleration measurements from an SHM system installed on a bridge, demonstrating the capability of the LSTM networks in identifying combined sensor faults, thus improving the quality of FD in SHM systems. Future research aims to decentralize the ICSF methodology and to reformulate the classification models in a mathematical form with an explanation interface, using explainable artificial intelligence, for increased transparency.

Based on ensemble empirical mode decomposition (EEMD) and the support vector machine (SVM), an algorithm used in the sensor fault detection and classification is put forward in this paper. Using this method and through EEMD, the sensor signal is decomposed into several segments, including the original signals, several intrinsic mode functions (IMFs) and the residual signals. Moreover, as features of the sensor fault, their variance, mean, entropy and the slope of the original signal are calculated in accordance with the characteristics of different fault types and the inherent physical meanings of each IMF. Subsequently, the feature vectors are inputted into the SVM, which is used to classify the detection and identification of sensor faults. Finally, the simulation results of the fault diagnosis of a carbon dioxide sensor indicate that this method may not only be effectively applied to fault diagnosis of carbon dioxide sensors but also provides a reference for that of other sensors.

A novel sensor fault diagnosis method based on Modified Ensemble Empirical Mode Decomposition and Probabilistic Neural Network

Reliable fault diagnosis and identification are essential for permanent magnet synchronous motor (PMSM) drive systems with high-reliability requirements. Since open-switch fault and output current sensor fault both affect the residual of output phase current, it is difficult to identify them apart from each other. Thus, this article proposes an open-switch fault and output current sensor fault diagnosis and identification method for the matrix converter (MC)-based PMSM drive system. The finite control set-model predictive control (FCS-MPC) method is applied in the MC-based PMSM drive system. The fault identification method is proposed based on the extracted different faulty features of the open-switch fault and the output current sensor fault. After the open-switch fault is separated from the current sensor fault, an error-voltage-based open-switch fault diagnosis strategy is applied to locate the faulty switch. Hardware-in-the-loop tests are carried out to verify the effectiveness of the proposed method.

Sensor fault identification based on time-lagged PCA in dynamic processes

Enhanced chiller sensor fault detection, diagnosis and estimation using wavelet analysis and principal component analysis methods

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.

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.