Reconstruction-based Contribution Research Articles

An improved industrial fault diagnosis model by integrating enhanced variational mode decomposition with sparse process monitoring method

With the continuous development of intelligent industrial processes, the sparse principal component analysis (SPCA), as a promising process monitoring method, has been widely used in the field of industrial fault detection. However, due to the inadequacy of data preprocessing and the insufficient detection accuracy for minor faults, the SPCA models exhibit obvious limitations in dealing with the processes with dynamic and temporal features. In this study, a Harris Hawk optimization method enhanced variational mode decomposition (HHO-VMD) coupled with the sliding window optimized adaptive SPCA (SWOASPCA) method is proposed to improve the fault detection performance of the SPCA models. In the HHO-VMD-SWOASPCA method, the process data is first preprocessed by adaptively and iteratively optimizing the number of modes and penalty terms in the VMD method via the Harris Hawk Optimization (HHO) method, and then the original SPCA model is combined with the sliding window method and the weight assignment strategy to enhance the model's adaptive capability and accuracy to detect minor faults. Moreover, an improved reconstruction-based contribution (RBC) method is presented to diagnose the detected faults for determining the fault causes. The effectiveness of the proposed method is verified by its application in the industrial sugar production process.

A new reconstruction-based method using local Mahalanobis distance for incipient fault isolation and amplitude estimation

Faulty variable isolation and amplitude estimation are of great importance to support the decision-making for system maintenance but lack sufficient studies, especially concerning the challenge of incipient faults with tiny amplitude. The reconstruction-based contribution (RBC) idea is commonly used for faulty variable isolation and fault amplitude estimation but usually suffers from low accuracy performance when facing the incipient faults challenge due to the use of insensitive detection indexes. Therefore, this paper proposes an improved reconstruction-based approach using the highly sensitive index named local Mahalanobis distance (LMD) for incipient fault isolation and amplitude estimation. The novel RBC approach retains the advantages of LMD, e.g., high sensitivity to incipient faults, robustness to outliers, and ability to handle non-Gaussian data, and is also available for multiple faulty variables isolation. The performance evaluation of the proposed methods using the benchmark case of the Continuous-flow Stirred Tank Reactor (CSTR) process shows that this approach has high isolation and estimation accuracy for both single and multiple faults. Thanks to a comparative study, it can be highlighted that for both faulty variables isolation and fault amplitude estimation tasks, our proposal outperforms the state-of-the-art RBC-based methods using different detection indexes: the combined index of PCA, conventional Mahalanobis distance, and augmented Mahalanobis distance.

Operating performance assessment method for industrial process with slowness principle-based LSTM network

Timely and accurate performance assessment and non-optimal regulation of industrial processes can effectively guarantee product quality. Most industrial processes are highly nonlinear and dynamic, so long short-term memory (LSTM) network is suitable for industrial performance assessment. However, in the network learning, the typical LSTM network focuses on the representation learning of input variables, lacks the representation of comprehensive economic indexes (CEI), and cannot selectively learn essential features, which increases the computational burden and easily mixes redundant information. Thus, a supervised slow feature analysis (SFA)-based LSTM (SSFALSTM) network is proposed for industrial operating performance assessment. By utilizing CEI information and SFA constraints, the network is guided to simultaneously learn features related to CEI and slow-changing features that reflect the inherent dynamics of the process. Further, cascade performance recognition model to construct the complete performance assessment framework. For the non-optimal performance, a reconstruction-based contribution plot method is proposed to identify the main cause variables and guide adjustment operations. Finally, the effectiveness of the proposed method is validated on the dense medium coal preparation process.

Dynamic monitoring and anomaly tracing of the quality in tobacco strip processing based on improved canonical variable analysis and transfer entropy.

Multivariate statistical monitoring methods are proven to be effective for the dynamic tobacco strip manufacturing process. However, the traditional methods are not sensitive enough to small faults and the practical tobacco processing monitoring requires further root cause of quality issues. In this regard, this study proposed a unified framework of detection-identification-tracing. This approach developed a dissimilarity canonical variable analysis (CVA), namely, it integrated the dissimilarity analysis concept into CVA, enabling the description of incipient relationship among the process variables and quality variables. We also adopted the reconstruction-based contribution to separate the potential abnormal variable and form the candidate set. The transfer entropy method was used to identify the causal relationship between variables and establish the matrix and topology diagram of causal relationships for root cause diagnosis. We applied this unified framework to the practical operation data of tobacco strip processing from a tobacco factory. The results showed that, compared with traditional contribution plot of anomaly detection, the proposed approach cannot only accurately separate abnormal variables but also locate the position of the root cause. The dissimilarity CVA proposed in this study outperformed traditional CVA in terms of sensitiveness to faults. This method would provide theoretical support for the reliable abnormal detection and diagnosis in the tobacco production process.

Multi-fault diagnosis for series-connected lithium-ion battery pack with reconstruction-based contribution based on parallel PCA-KPCA

Various faults of the lithium-ion battery threaten the safety and performance of the battery system. The early faults are difficult to detect and isolate owing to unobvious abnormality and the nonlinear time-varying characteristics of the battery. Herein, a multi-fault diagnosis strategy is proposed that focuses on detecting and isolating different types of faults, and estimating fault waveforms of the battery, including inconsistency evaluation, virtual connection fault, and external short circuit. First, the principal component analysis (PCA) model of the battery is established and the contribution is employed to detect the abnormity in the battery pack. Once the fault is detected, the parallel kernel principal component analysis (KPCA) technology is adopted to reconstruct the fault waveform of the battery parameters, including ohmic resistance, terminal voltage, and open-circuit voltage. These parameters are jointly taken as fault indexes improving the reliability of fault diagnosis. Finally, the proposed method is verified using amounts of tested data of eight cells in series. The results indicate that the contribution-based PCA method can accurately detect the fault. Furthermore, the reconstruction-based parallel PCA-KPCA can accurately estimate the fault waveform of the faulty battery, which helps investigate the fault degree and causes.

Cycle temporal algorithm-based multivariate statistical methods for fault diagnosis in chemical processes

Multivariate statistical process monitoring methods are often used in chemical process fault diagnosis. In this article, (I) the cycle temporal algorithm (CTA) combined with the dynamic kernel principal component analysis (DKPCA) and the multiway dynamic kernel principal component analysis (MDKPCA) fault detection algorithms are proposed, which are used for continuous and batch process fault detections, respectively. In addition, (II) a fault variable identification model based on reconstructed-based contribution (RBC) model that paves the way for determining the cause of the fault are proposed. The proposed fault diagnosis model was applied to Tennessee Eastman (TE) process and penicillin fermentation process for fault diagnosis. And compare with other fault diagnosis methods. The results show that the proposed method has better detection effects than other methods. Finally, the reconstruction-based contribution (RBC) model method is used to accurately locate the root cause of the fault and determine the fault path.

An incipient fault diagnosis methodology using local Mahalanobis distance: Fault isolation and fault severity estimation

Recently, a detection methodology based on anchor estimation and the Mahalanobis distance, named the local Mahalanobis distance (LMD) technique, has been proposed for incipient fault detection. As an extension, this paper proposes the LMD based faulty variable isolation and fault severity estimation methods to offer an integrated incipient fault diagnosis solution. The LMD technique has been proved to be highly sensitive to incipient faults, robust to outliers, and free of distribution assumptions. Within the LMD framework, the faulty variable is recognized by analyzing the relative position of faulty samples and their corresponding anchors. Then an analytical expression of fault severity derived from the LMD index is established for the fault severity estimation task. Performance evaluations of the proposed methods based on a benchmark case of the Continuous-flow Stirred Tank Reactor (CSTR) process are provided in this work. The result shows that even for tiny deviation, such as 5dB fault to noise ratio, the total isolation performance still achieves 100% accuracy, and the relative error of two tricky faults in the case of severity estimation is less than 10%. The comparison study of different methods indicates that our solution outperforms divergence based techniques and reconstruction based contribution (RBC) methods for the two tasks.

Implementation of fault isolation for molten salt reactor using PCA and contribution analysis

In order to implement the real-time monitoring of the system and equipment, principal component analysis (PCA) and contribution analysis are combined for fault isolation in this paper. The molten salt reactor (MSR) is taken as the research object. PCA and contribution analysis are applicable even in the case of lack of expert knowledge or small amount of data, with advantages in processing a large number of variables of complex systems. PCA is used for detecting faults. After an anomaly is detected, contribution analysis is used to diagnose the signal that causes the anomaly. The undetectable intervals of PCA models are performed quantitatively for obtaining better detection effect. Since the traditional contribution analysis method is prone to misjudgment due to smearing effect, the reconstruction-based contribution (RBC) is modified to improve the diagnostic accuracy. The three contribution analysis methods are compared and analyzed. Methods mentioned are evaluated through simulation tests and fault conditions in actual operation. Results demonstrate that the method proposed in this paper is accurate, effective and highly applicable.

Online reconstruction and diagnosibility analysis of multiplicative fault models for process-related faults

Multiplicative faults generally refer to the change of process parameters or structures which are well-suited to represent the process-related anomalies. Unlike sensor faults and external disturbances that are added into process observations and independent with process states, process-related faults directly influence process states such that it is more challenge to reconstruct and diagnose. To address the process-related fault diagnosis, an online fault reconstruction method based on the multiplicative fault model is proposed with the commonly used multivariate statistical process monitoring framework. The fault reconstruction strategy based on the multiplicative fault representation is given by minimizing reconstruction errors. The diagnosability of the proposed reconstruction method is guaranteed for the change of a single parameter, also known as a unidimensional fault. Moreover, the reconstruction-based contribution is derived for providing heuristic references when diagnosing multidimensional faults. Experiments on a numerical example and a simulated continuous stirred tank heater process benchmark are carried out to investigate the effectiveness of the proposed method. The results show that this method can accurately diagnose the faulty variable or loop and further reconstruct the faulty samples into normal ones.

Application of just-in-time-learning CCA to the health monitoring of a real cold source system

The stable and efficient operation of the cold source system can significantly reduce building energy consumption and improve indoor comfort. Health monitoring of the cold source system is a necessary means to ensure such a requirement. Due to the variations of ambient temperature and cold load, the cold source system often running in multiple modes, which limits the application of conventional health monitoring methods with poor performance, e.g. high false alarm rate. Therefore, this paper uses our previously proposed method, just-in-time-learning aided canonical correlation analysis (JITL-CCA), to monitor the health condition of the real cold source system. The application steps are detailed, including how to select monitoring variables and how to use reconstruction-based contribution (RBC) to assist JITL-CCA in fault diagnosis. The on-site data experiments show the applicability of the JITL-CCA method and its superior performance when comparing with the conventional CCA method.

Fault Detection and Diagnosis in Industrial Processes with Variational Autoencoder: A Comprehensive Study.

This work considers industrial process monitoring using a variational autoencoder (VAE). As a powerful deep generative model, the variational autoencoder and its variants have become popular for process monitoring. However, its monitoring ability, especially its fault diagnosis ability, has not been well investigated. In this paper, the process modeling and monitoring capabilities of several VAE variants are comprehensively studied. First, fault detection schemes are defined in three distinct ways, considering latent, residual, and the combined domains. Afterwards, to conduct the fault diagnosis, we first define the deep contribution plot, and then a deep reconstruction-based contribution diagram is proposed for deep domains under the fault propagation mechanism. In a case study, the performance of the process monitoring capability of four deep VAE models, namely, the static VAE model, the dynamic VAE model, and the recurrent VAE models (LSTM-VAE and GRU-VAE), has been comparatively evaluated on the industrial benchmark Tennessee Eastman process. Results show that recurrent VAEs with a deep reconstruction-based diagnosis mechanism are recommended for industrial process monitoring tasks.

Multivariate/minor fault diagnosis with severity level based on Bayesian decision theory and multidimensional RBC

Fault diagnosis(FD) is vital for monitoring of industrial processes. Actually, both the multiple fault variables and minor faults are inclined to bring wrong diagnosis results. In this paper, a fault variable identification method is proposed for multivariate/minor fault diagnosis. The deviation factor is adopted as the characteristic of the sample, and Bayesian decision theory is adopted to calculate the possibility of the variable being faulty, then multi-dimensional reconstruction-based contribution (MRBC) is used to determine the fault source variables. This method not only improves the diagnosis rate for multiple/minor faults with less computation time, but also can indicate severity level of the different fault variables according to the fault occurrence probability. A Numerical example and Tennessee Eastman process are given to show the efficiency of the proposed method.

Integrated Diagnostic Framework for Process and Sensor Faults in Chemical Industry.

This study considers the problem of distinguishing between process and sensor faults in nonlinear chemical processes. An integrated fault diagnosis framework is proposed to distinguish chemical process sensor faults from process faults. The key idea of the framework is to embed the cycle temporal algorithm into the dynamic kernel principal component analysis to improve the fault detection speed and accuracy. It is combined with the fault diagnosis method based on the reconstruction-based contribution graph to diagnose the fault variables and then distinguish the two fault types according to their characteristics. Finally, the integrated fault diagnosis framework is applied to the Tennessee Eastman process and acid gas absorption process, and its effectiveness is proved.

An Improved Fault Diagnosis Strategy for Process Monitoring Using Reconstruction Based Contributions

Air pollution has become the fourth leading cause of premature death on Earth. Air pollution causes poor health and death; about one case out of every ten deaths worldwide is caused by air pollution, which is six times more than malaria. Human activities are the main cause of air pollution, such as chemical industries, road traffic, and fossil fuel power plants. Over the span of several years, monitoring air quality has become an exigent and essential task. In order to limit the health impact of air pollution and to ensure safe operation of chemical processes, it is necessary to quickly detect and locate instrumentation defects. There are several process monitoring techniques in the literature. Among these techniques is the one selected for this work for the detection and location of sensor faults: the kernel principal component analysis (KPCA) method, which was selected for its primary advantages of easy employment and less necessity for prior knowledge. Using the KPCA method for monitoring nonlinear systems, the calculation cost and the memory size are related to the number of initial data. This is currently a major limitation of the KPCA method, especially in industrial environments. In order to remedy this limitation, in this paper we propose a new method of detection and localization. The key idea of this approach is to extend the method of localization based on the principle of reconstruction-based contributions (RBC) by downsizing the kernel matrix in the characteristic space. The proposed technique is named reconstruction-based contribution reduced rank kernel principal component analysis(RBC-RRKPCA). The approach is demonstrated using real air quality monitoring network data and simulated data from the Tennessee Eastman process (TEP) as a challenging benchmark problem. We also present a comparative study of the performances of the conventional diagnostic technique RBC-KPCA and the proposed technique RBC-RRKPCA. The results in this paper reveal that the proposed technique achieves the highest detection and localization accuracy.

Low-Rank Joint Embedding and Its Application for Robust Process Monitoring

Industrial data are in general corrupted by noises and outliers. In this context, robustness to the contaminated data is a challenging issue in process monitoring. In this article, a novel method named low-rank joint embedding is proposed for robust process monitoring. By learning a low-rank coefficient matrix, low-rank joint embedding can capture the global structure of the original data and alleviate the negative effect of outliers, making the monitoring results more reliable. Moreover, a manifold regularization is introduced to preserve the local geometric structure of data, which enables the extracted low-dimensional representation of data to be more faithful and informative to enhance the monitoring capability. Based on projection learning, the low-rank joint embedding can learn an explicit projection that transforms the data not involved in the training data into the low-dimensional space, avoiding the out-of-sample problem. Furthermore, a reconstruction-based contribution plots based on the low-rank joint embedding is developed to identify the potential faulty variables. Case studies on the Tennessee Eastman process and a real industrial application demonstrate the effectiveness of the proposed monitoring approach.

A multi‐block NMF model for non‐Gaussian process monitoring based on the adaptive partition non‐negative matrix factorization and Bayesian inference

AbstractNon‐negative matrix factorization (NMF) is a novel technique for dimension‐reduction, which can be used to process data of non‐Gaussian and Gaussian efficiently. A global NMF model is inappropriate for the whole process, since it neglects the local information and monitoring results are often hard to be interpreted. On the basis of adaptive partition non‐negative matrix factorization (APNMF) and Bayesian inference, a multi‐block NMF model for non‐Gaussian process monitoring is put forward to detect and isolate the faults effectively. Using APNMF method, the original variables in different fault states can be adaptively divided into multiple sub‐blocks, and on this basis, the NMF monitoring model of each sub‐block is formed. Then, two new statistics are constructed by Bayesian inference to supply an intuitive display. Finally, a weighted reconstruction‐based contribution (RBC) plot method is presented to reduce the smearing effect and find out the main causes of these faults. This method makes full use of the local and global information of process data and improves the effectiveness of process monitoring. The validity and feasibility of the proposed method will be proved by an example of a numerical process, a Tennessee Eastman (TE) benchmark process and a continuous stirred‐tank reactor (CSTR) process.

Between-class difference analysis based multidimensional RBC for multivariate fault isolation of industrial processes

In industrial process, fault isolation technology will identify the major variables leading to faults. The existing fault isolation methods usually have ”smearing effect”, cannot identify the sensitivity of the fault variables, or require too much calculation. To solve the above problems, a multivariate fault isolation method based on between-class difference analysis and multidimensional reconstruction-based contribution (RBC) is proposed in this work. First, two kind of strategies, i.e.,principle component analysis (PCA) and fisher discriminant analysis (FDA), are adopted to obtain the sensitivity indicators of the variables to the faults respectively, which is taken as the basis for determining the optimal reconstruction direction. Then, multidimensional RBC is used to determine the number of fault variables based on the selected reconstruction direction. Finally, the primary and secondary fault variables are isolated according to their fault sensitivities. The feasibility of the proposed method is illustrated by a numerical simulation example and TE process. To sum up, the proposed method provide sensitivity information of each fault variables and high diagnosis rate. Moreover, it also has an advantage in reducing the calculation time because of the preliminary selection of fault variables.

Dynamic statistical process monitoring based on generalized canonical variate analysis

A novel generalized canonical variate analysis (GCVA) algorithm is formulated and then applied for data-driven dynamic process monitoring. The proposed GCVA algorithm seeks for different projecting bases for the time-serial samples, so that the sum of squared canonical correlation coefficients between all pairs of the projected latent variables could be maximized. The corresponding dynamic process monitoring scheme first utilizes GCVA to explicitly extract dynamic and static latent variables from the time-serial data, simultaneously. Second, a multivariate regression model is employed for describing the time-serial relationship between the dynamic latent variables, the model residual then services as a good indicator for the inconsistency in the defined time-serial mechanism. For online monitoring purposes, two combined monitoring indices are proposed for detecting abnormalities in the time-dependent and time-independent variations, respectively. Additionally, reconstruction-based contribution indices are also derived for fault diagnosis accordingly. Finally, the capability of the GCVA algorithm in exploiting the time-serial correlation inherited in the given data is demonstrated, the effectiveness and superiority of the proposed GCVA-based approach over other counterparts are validated as well, through comparisons on two dynamic industrial processes.

Locally linear back-propagation based contribution for nonlinear process fault diagnosis

This paper proposes a novel locally linear backpropagation based contribution (LLBBC) for nonlinear process fault diagnosis. As a method based on the deep learning model of auto-encoder (AE), LLBBC can deal with the fault diagnosis problem through extracting nonlinear features. When the on-line fault diagnosis task is in progress, a locally linear model is firstly built at the current fault sample. According to the basic idea of reconstruction based contribution (RBC), the propagation of fault information is described by using back-propagation (BP) algorithm. Then, a contribution index is established to measure the correlation between the variable and the fault, and the final diagnosis result is obtained by searching variables with large contributions. The smearing effect, which is an important factor affecting the performance of fault diagnosis, can be suppressed as well, and the theoretical analysis reveals that the correct diagnosis can be guaranteed by LLBBC. Finally, the feasibility and effectiveness of the proposed method are verified through a nonlinear numerical example and the Tennessee Eastman benchmark process.

Canonical correlation analysis-based explicit relation discovery for statistical process monitoring

Different from the latent variables which characterize the implicit relation, the proposed method focuses on discovery and description of the explicit relation between measured variables, based on which a novel statistical process monitoring approach is then derived for both static and dynamic processes. First, the canonical correlation analysis (CCA) algorithm is employed to find a set of interacted variables for every single variable individually, a regression model is then used to describe the explicit relation between the interacted variables. Second, on the basis of an ensemble representation of the mathematically defined explicit relation, fault detection and reconstruction-based contribution for fault diagnosis through the residual can be implemented. Finally, the effectiveness and superiority of the proposed approach are validated through comparisons with other state-of-the-art methods that based on latent variable models.