Fault Detection for MIMO Systems Integrating Multivariate Statistical Analysis and Identification Methods

FAULT DETECTION FOR LARGE SCALE SYSTEMS USING DYNAMIC PRINCIPAL COMPONENTS ANALYSIS WITH ADAPTATION

Fault detection and diagnosis in process data using one-class support vector machines

Comparison Between Statistical and Observer-Based Approaches for Fault Detection and Isolation in a Chemical Process

Fault detection in continuous industrial processes is crucial to ensure production, quality, safety, environmental and economic aspects. Reliable fault detection systems have become even more necessary due to increasingly stringent regulations. The complexity of processes, on the one hand, and the availability of data, on the other, have spread the adoption of the data-driven approach for industrial fault detection. Deep Reinforcement Learning (RL) has recently been applied for this purpose, mainly in rotating machinery and rolling bearings. In chemical processes, most RL applications are in process control. The main contribution of this work lies in proposing a reinforcement learning (RL)-based framework for fault detection in continuous chemical processes. The challenges are related to the large number of variables, nonlinear relationships and changing operational dynamics, to name a few. The classic Tennessee Eastman Process (TEP) benchmark was used as a case study. The results were compared with Principal Component Analysis (PCA), commonly used for chemical process monitoring, dynamic PCA (DPCA), and neural networks. Compared to PCA, the RL approach showed considerably lower missed detection rate (MDR) and time to detection (TTD) while maintaining reasonable false alarm rate (FAR) for all hard-to-detect faults, contributing to early fault detectability. Similar results were obtained with regard to DPCA. Compared to the supervised approach, RL proved to be more flexible regarding model parameterization, which can be very useful for practical aspects of industrial fault detection. The results show that RL-based approaches can be very useful for fault detection in continuous, non-linear, and time-varying systems, such as chemical processes. For example, with regard to adaptive alarm thresholds and the detection of known and unknown faults. The latter represents a major challenge in industrial practice.

Fast and selective DC fault detection and location schemes are needed for fault protection of DC Grids. In this paper, the frequency domain characteristics of DC voltage besides the current limiting reactors are analyzed and used to put forward a new fault detection method for DC grids. Hilbert-Huang transform is applied on the DC voltage waves to extract the transient frequency and transient amplitude of the DC voltage. The influence of fault type, fault distance and fault resistance of DC faults and the noise on the frequency domain transient characteristics on voltage waves are then studied. Support vector machine learns the mapping between the transient frequency and transient amplitude of the DC voltage and desired outputs, fault type and fault location through training process. Then, a fast and accurate fault detection scheme based on frequency-domain characteristics of DC voltage is proposed by SVM for DC fault detection and location in MTDC systems. This fault detection scheme requires no telecommunication, has anti-interference ability and good sensitivity and selectivity for detection of long-distance fault and high resistance grounding fault. Based on the four-terminal HVDC system built by PSCAD/EMTDC, simulations under different fault conditions are carried out to verify the effectiveness of the proposed DC fault detection scheme.

In traditional dynamic principal component analysis (DPCA) for fault detection, there are some drawbacks such as an excess of the number of principal components (PCs), low computational efficiency, etc. For dealing with the problem, this paper develops a hybrid dynamic principal component analysis (HDPCA) technique, this method can remove spacial and serial correlation by divide-and-conquer algorithm instead of parallel processing strategy, which can detect individual fault accurately and efficiently. The strip breaking fault in steel rolling process is used to demonstrate the improved performance of developed technique in comparison with traditional DPCA fault detection methods. It can be perceived that HDPCA algorithm has the better performance of fault detection and computational efficiency.

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.

This paper examines a robust fault detection scheme that can be used to detect faulty sensors of jet engines. The fault detection scheme has to be insensitive to disturbances while being highly sensitive to sensor faults (robust). The paper presents a complete description of a robust fault detection approach based on eigenstructure assignment, both in continuousand discrete-time domains. By assigning the left (or right) eigenvectors of the observer orthogonal (or parallel) to the disturbance directions, the robust (disturbance decoupling) fault detection is achieved. The approach has been applied to a realistic jet engine simulation system. The system is a 17th-order system, and a reduced-order model is used to approximate the system. Modeling errors are considered as disturbances acting on the fault detection scheme. A particularly novel feature of the work is the development and use of a new method (new in this context) for estimating disturbance direction. The robust fault detection scheme design uses this estimated direction as that of the direction of unknown inputs (disturbances). Simulation results show that the scheme can detect soft or incipient faults efficiently.

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.

Discriminative common vector in sufficient data Case: A fault detection and classification application on photovoltaic arrays

The authors describe a robust scheme for use in the role of fault detection in sensors and actuators of process systems. The fault detection scheme is designed to be insensitive to disturbances, while being sensitive to faults. The authors present the robust fault detection approach based on eigenstructure assignment in the discrete-time domain. By assigning the left eigenvectors of the observer orthogonal to the disturbance, robust (disturbance decoupling) fault detection is achievable. The approach has been applied to an 11th-order nuclear reactor system. A reduced-order model is used to approximate this system and modeling errors are considered as disturbances acting on the reduced-order model contained in the observer. The work involves the development and use of a method for estimating the disturbance direction. The robust fault detection scheme design uses this direction as that of the direction of unknown input acting on a fourth-order model. Simulation results show that the scheme can detect soft or incipient faults efficiently. >

Monitoring large-scale processes is a crucial task to ensure the safety and reliability of the plants. This paper proposes an approach for decentralized fault detection in large-scale processes. The measured variables of the plant are divided into multiple and possibly overlapping blocks using different techniques based on data. Local monitoring methods are applied in each block using DPCA (Dynamic Principal Component Analysis) model. The local results are then fused by the Bayesian inference strategy. This paper also compares different techniques to decompose the plant looking for the best strategy from the point of view of the fault detection results. The proposed method was applied to the widely used benchmark Tennessee Eastman Process, showing its effectiveness when compared with a centralized method and another decentralized technique.

This paper presents and compares model-based and data-driven fault detection approaches for coal mill systems. The first approach detects faults with an optimal unknown input observer developed from a simplified energy balance model. Due to the time-consuming effort in developing a first principles model with motor power as the controlled variable, data-driven methods for fault detection are also investigated. Regression models that represent normal operating conditions (NOCs) are developed with both static and dynamic principal component analysis and partial least squares methods. The residual between process measurement and the NOC model prediction is used for fault detection. A hybrid approach, where a data-driven model is employed to derive an optimal unknown input observer, is also implemented. The three methods are evaluated with case studies on coal mill data, which includes a fault caused by a blocked inlet pipe. All three approaches detect the fault as it emerges. The optimal unknown input observer approach is most robust, in that, it has no false positives. On the other hand, the data-driven approaches are more straightforward to implement, since they just require the selection of appropriate confidence limit to avoid false detection. The proposed hybrid approach is promising for systems where a first principles model is cumbersome to obtain.

In this paper, fault detection scheme is introduced to improve the security and reliability of the NTRU in realistic environments, such as operating over a network. As the NTRU will be working in real networks, which have their own set of transient errors, handling such errors in analyzing the NTRU becomes a must. A single transient error occurring during the NTRU encryption (or decryption) process will likely result in a large number of errors in the encrypted/decrypted data. Such faults must be detected before sending data to avoid the transmission and use of erroneous data. Concurrent fault detection is important not only to protect the encryption/decryption process from random faults, but it will also protect the decryption circuitry from an attacker who may maliciously inject faults in order to find the secret key. We first describe the effects that faults may have on NTRU security while operating over a network by analyzing the propagation of such faults to the outputs. We then present two fault detection schemes: The first is a redundancy-based scheme while the second uses an error detection code. The latter present fault detection schemes by using an error detecting code, using one parity bit. We will add the parity bits to the polynomials and disable the device output when any of these parity checks are violated. This fault detection scheme has more than 99% coverage fault detection in the encryption process and in the first part in the decryption process. In the second part in the decryptionprocess, it has more than 67 % coverage fault detection. We can increase this ratio to high fault detection by adjusting the choices of the parameters of the NTRU network security system.

This article addresses fault detection, estimation, and compensation problem in a class of disturbance driven time delay nonlinear systems. The proposed approach relies on an iterative learning observer (ILO) for fault detection, estimation, and compensation. When there are no faults in the system, the ILO supplies accurate disturbance estimation to the control system where the effect of disturbances on estimation error dynamics is attenuated. At the same time, the proposed ILO can detect sudden changes in the nonlinear system due to faults. As a result upon the detection of a fault, the same ILO is used to excite an adaptive control law in order to offset the effect of faults on the system. Further, the proposed ILO-based adaptive fault compensation strategy can handle multiple faults. The overall fault detection and compensation strategy proposed in the paper is finally demonstrated in simulation on an automotive engine example to illustrate the effectiveness of this approach. Copyright © 2005 John Wiley & Sons, Ltd.