Fault Detection In Industrial Processes Research Articles

Industrial Process Fault Detection Based on IGA‐Combinatorial Model Decision Mechanism

ABSTRACTTo address the challenges of extracting features from complex industrial process data, the reliance of numerous fault detection methodologies on presupposed data distribution types, and the limited generalization capacity of fault detection, this manuscript introduces a sophisticated algorithm for industrial process fault detection. This algorithm harnesses the information gain adaptive (IGA) technique for feature selection and a synergistic model decision mechanism. Initially, the process involves the computation of information gain via decision trees, coupled with the determination of the value through cross‐validation. This strategy enables the adaptive selection of features, thereby facilitating data dimensionality reduction and effective feature extraction. The subsequent phase introduces a ternary statistical measure monitoring group for the detection of linear faults, while autoencoders and one‐class SVM methodologies are applied for the monitoring of nonlinear faults. The culmination of this approach is the development of an innovative weighted decision mechanism, designed to amalgamate the findings from both linear and nonlinear detection avenues, yielding more dependable detection results. The validation of this algorithm employs datasets from the water chillers process and Tennessee Eastman (TE) process, demonstrating the IGA‐combined model's superior performance over isolated linear or nonlinear detection algorithms in terms of detection accuracy and robustness. Notably, the efficacy of this method is not contingent upon specific assumptions regarding data distribution, rendering it a versatile and efficacious tool for the fault detection in industrial processes.

Improved diffusion mapping combined with procrustes analysis for capturing local-global data structures in industrial process monitoring

BackgroundProcess monitoring, by providing early warnings of abnormal operating states resulting from process faults, facilitates the maintenance of normal production and ensures process safety. In the domain of industrial process monitoring, capturing the local-global structural features of data and acquiring an explicit mapping relationship for dimensionality reduction projection holds significant importance for online fault detection in industrial processes. MethodsThis study introduces an Improved Diffusion Mapping and Procrustes analysis (IDM-P) method for this purpose. Initially, considering the multiscale and correlation among industrial data features, the Mahalanobis distance is incorporated to improve the diffusion mapping algorithm. Utilizing this method allows for the concurrent capture of both local and global data structures, leading to a more efficient extraction of data-representative features, which enhances the accuracy of fault detection. Procrustes analysis is then used to obtain an explicit mapping matrix between high-dimensional data and low-dimensional manifolds, improving the efficiency of the key feature extraction of the new samples. Finally, this matrix is utilized to construct process monitoring statistics for fault detection. Significant FindingsThe method's effectiveness was validated through experiments on the TEP dataset and actual industrial data, demonstrating that IDM-P maintains higher accuracy and achieves optimal fault detection compared to other methods.

Industrial process fault detection based on dynamic kernel principal component analysis combined with weighted structural difference

AbstractThe 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.

Fault detection of industrial processes using attention-based gated recurrent unit autoencoder with skip connection

Abstract With the continuous evolution of modern industrial technology, industrial production has grown progressively complex, necessitating the use of various sensors to measure multiple process variables. However, intricate temporal dependencies and nonlinear relationships between data presented by multivariate sequences pose significant challenges to process fault detection. In response to these challenges, this paper proposes an attention-based gated recurrent unit autoencoder with skip connection (SAGRU-AE) model for monitoring large-scale, nonlinear, and multivariate industrial process faults. SAGRU-AE combines gated recurrent units, multi-head self-attention, and autoencoder to extract features from multivariable time series data efficiently. Concurrently, feature reuse is achieved through the skip connection structure, which improves the accuracy of data reconstruction. Based on the implementation of process data feature extraction and input reconstruction in SAGRU-AE, two statistics have been developed, namely the H2 statistic and square prediction error (SPE) statistic, for fault detection tasks. Ultimately, the feasibility and effectiveness of the proposed algorithms are validated through experimentation on the TE process.

Incipient fault detection based on dense feature ensemble net

With modern industrial processes becoming more and more complex, the occurrence of faults may cause unmitigated disaster. Therefore, incipient fault detection is very important and has attracted increasing attention Recently, a feature ensemble net (FENet) has been proposed based on a deep feature ensemble framework, which can detect partial notorious faults (such as faults 3, 9 and 15 in the Tennessee Eastman process (TEP)). However, the feature transformer layers in FENet can only mine the process information from the upper layer’s output rather than all features from previous layers. Therefore, a novel dense feature ensemble net (DenseFENet) is proposed to deeply extract the features of data for incipient fault detection in industrial process. DenseFENet firstly constructs base detectors with capabilities for extracting nonlinear features, such as one-class support vector machine, isolation forest, one-class back propagation neural network, one-class long short term memory network and one-class temporal convolutional network, etc. Then, a dense net structure with short paths between non-adjacent feature transformer layers is developed, improving the reutilization capability for shallow knowledge. In addition, a numerical simulation and TEP are adopted to demonstrate the performance of the proposed method. The fault detection rate of DenseFENet has improved by 5.05% and 3.27% correspondingly, revealing the superiority of this approach.

Comparison of autoencoder architectures for fault detection in industrial processes

Fault detection constitutes a fundamental task for predictive maintenance, requiring mathematical models that can be conveniently provided by data-driven techniques. Autoencoders are a particular type of unsupervised Artificial Neural Networks that can be suitable for fault detection applications. Diverse architectures might be used for autoencoders, resulting in different fault detection performances, which are usually compared by means of Fault Detection Rates for a fixed threshold of the False Alarm Rate, limiting the conclusions to particular cases. To improve the comparability, the present work uses the area under the receiver operating characteristic curve to compare autoencoder architectures for a range of false alarm rates using the Tennessee Eastman Process benchmark. Performances obtained for shallow and deep autoencoders were compared with those of the denoising and variational autoencoders for undercomplete and sparse structures. Overall, the results indicate better performances for sparse structures, especially for the variational autoencoder and the deep denoising autoencoder, with area under the curve of 98.35%.

Gaussian-kernel weighted neighborhood preserving embedding algorithm and its application in fault detection

Fault detection in industrial processes is essential for enhancing production safety. Despite the application of the neighborhood preserving embedding (NPE) algorithm in fault detection as a manifold learning technique, a notable limitation exists-NPE overlooks local geometric structure, leading to suboptimal fault detection and occasional false alarms. This paper introduces the Gaussian kernel weighted NPE (KW-NPE) algorithm to address this challenge. Specifically designed for precise weight assignment in local structures, KW-NPE strategically employs the Gaussian kernel method to project the spatial neighborhood set and capture comprehensive local structural characteristics. The weight assignment, dependent on feature values, enhances the retention of intrinsic structure during dimensionality reduction. A novel objective function further augments this process.To assess performance, a comprehensive composite index is introduced in a case study, amalgamating the false alarm rate and fault detection rate. The effectiveness of the KW-NPE algorithm is demonstrated through extensive simulations and its application to the Tennessee Eastman process dataset, highlighting its superiority over conventional approaches.

Deep Probabilistic Principal Component Analysis for Process Monitoring.

Probabilistic latent variable models (PLVMs), such as probabilistic principal component analysis (PPCA), are widely employed in process monitoring and fault detection of industrial processes. This article proposes a novel deep PPCA (DePPCA) model, which has the advantages of both probabilistic modeling and deep learning. The construction of DePPCA includes a greedy layer-wise pretraining phase and a unified end-to-end fine-tuning phase. The former establishes a hierarchical deep structure based on cascading multiple layers of the PPCA module to extract high-level features. The latter builds an end-to-end connection between the raw inputs and the final outputs to further improve the representation of the model to high-level features. After constructing the model structure of DePPCA, we first present the detailed training processes of the pretraining and fine-tuning stages, then clarify the theoretical merits of the proposed model from the perspective of variational inference. For process monitoring purposes, we develop two statistics based on the established DePPCA. The monitoring performance of these two statistics can remain superior even if the features extracted by DePPCA are significantly compressed to univariate. This makes the feature extraction process and online monitoring procedure of DePPCA quite fast. In other words, the proposed DePPCA can achieve accurate and efficient process monitoring by only extracting one feature for each sample. Finally, the effectiveness of DePPCA is evaluated on the Tennessee Eastman (TE) process and the multiphase flow (MPF) facility.

A Self-Attention Mechanism Integrating Adaptive Double Subspace for Fault Detection in Industrial Processes

A Self-Attention Mechanism Integrating Adaptive Double Subspace for Fault Detection in Industrial Processes

Online fault detection for industrial processes through Kalman filter

Industrial processes suffer from a wide range of damages including normal wear, environmental changes, physical structural defects and so on. This paper describes the possibility of system health management based on a prediction model, i.e., state space model realized by Kalman filter. The categorical target was mapped to numerical values in advance for this purpose. To deal with the time-varying and streaming characteristics of the industrial process, the model is applied in an online fashion. Comparing with conventional fault detection techniques, this model has the advantages of monitoring not only the production process of interests through observation equation, but also the structural anomalies described via unseen states estimation. In addition, the process and measurement noises provide valuable information about the unstructured uncertainties caused by other reasons. Experiments have been conducted to valid the effectiveness of the proposed method.

Modified q–[formula omitted] rule for incipient fault detection in industrial processes on analysis of intermittent process variation

Incipient faults usually have small impacts on systems. They are easily affected by noise and control effects. Thus, state variables will not deviate from their mean values in a sustained way, but in a fluctuating form. That is, fault samples will return to control limits in some time. I.e., the intermittent process variation occurs. Therefore, there will be lots of missed detection and high detection delay in traditional methods. It even causes the illusion that the system returns to normal. To address this issue, canonical variate residuals are firstly generated. The aim is to capture the system dynamic. Then the modified q–σ rule is proposed, it sets control limits for all variables. And it also monitors the intermittent process variation with fluctuating growth. When tested on a simulated process and a real industrial process, the proposed method is sensitive to the intermittent process variation in incipient faults and has a shorter detection delay.

Decentralized adaptively weighted stacked autoencoder-based incipient fault detection for nonlinear industrial processes

Modern industrial processes often exhibit large-scale and nonlinear characteristics. Incipient fault detection for industrial processes is a big challenge because of the faint fault signature. To improve the performance of incipient fault detection for large-scale nonlinear industrial processes, a decentralized adaptively weighted stacked autoencoder (DAWSAE) -based fault detection method is proposed. First, the industrial process is divided into several sub-blocks and local adaptively weighted stacked autoencoder (AWSAE) is established for each sub-block to mine local information and obtain local adaptively weighted feature vectors and residual vectors. Second, the global AWSAE is established for the whole process to mine global information and obtain global adaptively weighted feature vectors and residual vectors. Finally, local statistics and global statistics are constructed based on local and global adaptively weighted feature vectors and residual vectors to detect the sub-blocks and the whole process, respectively. The advantages of proposed method are verified by a numerical example and Tennessee Eastman process (TEP).

Long–short-term memory encoder–decoder with regularized hidden dynamics for fault detection in industrial processes

The ability of recurrent neural networks (RNN) to model nonlinear dynamics of high dimensional process data has enabled data-driven RNN-based fault detection algorithms. Previous studies have focused on detecting faults by identifying the discrepancies in data distribution between the faulty and normal data, as reflected in prediction errors generated by RNN models. However, in industrial processes, variations in data distribution can also result from changes in normal control setpoints and compensatory control adjustments in response to disturbances, making it hard to differentiate between normal and faulty conditions. This paper proposes a fault detection method utilizing a long short-term memory (LSTM) encoder–decoder structure with regularized hidden dynamics and reversible instance normalization (RevIN) to compactly represent high-dimensional measurements for effective monitoring. During training, the hidden states of the model are regularized to form a low-dimensional latent space representation of the original multivariate time series data. As a result, the prediction errors of the latent states can be used to monitor the abnormal dynamic variations, while the reconstruction errors of the measured variables are used to monitor the abnormal static variations. Furthermore, the proposed indices can reflect operating conditions, even when the distribution of test data changes, which helps distinguish faults from normal adjustments and disturbances that controllers can settle. Data from numerical simulation and the Tennessee Eastman process are used to illustrate the effectiveness of the proposed fault detection method.

Feature learning based on entropy estimation density peak clustering and stacked autoencoder for industrial process monitoring

AbstractRecently, stacked autoencoder (SAE)‐based deep learning has been widely employed for industrial process fault detection, which can extract representative low‐dimensional features. However, SAE fails to take the distribution structure of raw data into consideration in the unsupervised self‐reconstruction learning, and this may lead to limited monitoring performance. Focusing on this issue, this paper proposes a new fault detection method based on entropy estimation density peak clustering and stacked autoencoder (EEDPC‐SAE) for industrial process monitoring. First, an improved automatic clustering algorithm, namely, entropy estimation density peak clustering (EEDPC), is presented to analyze the distribution structure of process data. Then, the data distribution information is integrated into the learning procedure of SAE to capture the data intrinsic structure. EEDPC‐SAE can describe the distribution characteristics of process data, and more informative high‐level features can be learned for fault detection. Finally, two statistics, that is, Hotelling's T‐squared () and squared prediction error (SPE), are established based on the encoder features and residual features generated by EEDPC‐SAE. The effectiveness of the proposed method is evaluated by the Tennessee Eastman (TE) process and fed‐batch fermentation penicillin (FBFP) process.

Fault detection in multimodal processes based on the local entropy double subspace

Recently, principal component analysis (PCA) and independent component analysis (ICA) have attracted increasing attention for fault detection in industrial processes. However, industrial processes usually have multimodal characteristics, in which case, the performance of traditional PCA and ICA methods is poor. A fault detection method based on the local entropy double subspace (LEDS) is proposed that effectively supervises the multimodal process with multiple steady modes. The proposed method uses the local probability density (LPD) to eliminate the multimodal characteristics of the data. Furthermore, local entropy (LE) is applied to mine the statistical information associated with the process data. The process variables are classified into Gaussian and non-Gaussian subspaces by the Kolmogorov–Smirnov (KS) test. PCA and ICA models are built in Gaussian and non-Gaussian subspaces, respectively, for fault detection. A Bayesian decision method is employed to transform the detection results into fault probabilities, and the detection results are combined with the final statistical information. This method effectively eliminates the multicentre structure of the data and improves the fault detection performance of traditional PCA and ICA. The proposed method is applied to the Tennessee Eastman multimodal process. The simulation results show that the fault detection rate of LEDS is better than the rates of traditional methods when the false alarm rate is low for most faults.

Applying Intelligent Multi-Agents to Reduce False Alarms in Wind Turbine Monitoring Systems

This study proposes a method for improving the capability of a data-driven multi-agent system (MAS) to perform condition monitoring and fault detection in industrial processes. To mitigate the false fault-detection alarms, a co-operation strategy among software agents is proposed because it performs better than the individual agents. Few steps transform this method into a valuable procedure for improving diagnostic certainty. First, a failure mode and effects analysis are performed to select physical monitoring signals of the industrial process that allow agents to collaborate via shared signals. Next, several artificial neural network (ANN) models are generated based on the normal behavior operation conditions of various industrial subsystems equipped with monitoring sensors. Thereafter, the agents use the ANN-based expected behavior models to prevent false alarms by continuously monitoring the measurement samples of physical signals that deviate from normal behavior. Finally, this method is applied to a wind turbine. The system and tests use actual data from a wind farm in Spain. The results show that the collaboration among agents facilitates the effective detection of faults and can significantly reduce false alarms, indicating a notable advancement in the industrial maintenance and monitoring strategy.

Interval-valued data correlation modeling approach for uncertain nonlinear and non-Gaussian process monitoring

Process monitoring is a crucial part of ensuring the safety and quality of industrial production, and fault detection is a particularly critical step. As a departure from the dimensionality reduction strategy commonly used in fault detection methods, this paper aims to create a statistical model by directly extracting complex correlations among variables with nonlinearity and non-Gaussian properties. Uncertainties in measurement data in an actual process can significantly impact the control decision based on a monitoring model, so interval-valued description strategy is introduced to effectively take the uncertainties into account. Moreover, we improved upon the traditional interval-valued data generation method using moving window technology combine the receiver-operator characteristic curve to construct intervals based on sample mean and standard deviation (SD), which makes full use of the data information. This paper proposes a mean-SD interval vine copula (MSIVC) model for complex industrial process fault detection. The high density region and density quantile theory are introduced to determine the control boundary. The process monitoring performance of the MSIVC method is evaluated by a numerical example and the Tennessee-Eastman process. The results show that the proposed model is stable, sensitive to process faults, and yields effective monitoring results.

MoniNet With Concurrent Analytics of Temporal and Spatial Information for Fault Detection in Industrial Processes

Modern industrial plants generally consist of multiple manufacturing units, and the local correlation within each unit can be used to effectively alleviate the effect of spurious correlation and meticulously reflect the operation status of the process system. Therefore, the local correlation, which is called spatial information here, should also be taken into consideration when developing the monitoring model. In this study, a cascaded monitoring network (MoniNet) method is proposed to develop the monitoring model with concurrent analytics of temporal and spatial information. By implementing convolutional operation to each variable, the temporal information that reveals dynamic correlation of process data and spatial information that reflects local characteristics within individual operation unit can be extracted simultaneously. For each convolutional feature, a submodel is developed and then all the submodels are integrated to generate a final monitoring model. Based on the developed model, the operation status of the newly collected sample can be identified by comparing the calculated statistics with their corresponding control limits. Similar to the convolutional neural network (CNN), the MoniNet can also expand its receptive field and capture deeper information by adding more convolutional layers. Besides, the filter selection and submodel development in MoniNet can be replaced to generalize the proposed network to many existing monitoring strategies. The performance of the proposed method is validated using two real industrial processes. The illustration results show that the proposed method can effectively detect process anomalies by concurrent analytics of temporal and spatial information.

Feature extraction and fault detection scheme via improved locality preserving projection and SVDD

Manifold learning is widely adopted for the fault detection of industrial processes. However, the quality of low-dimensional embedding coordinates can be adversely affected by ill-constructed graph Laplacian. An improved locality preserving projection (ILPP) scheme is proposed. ILPP is built on a geometrically inspired Laplacian, and the Riemannian metric is used to find the suitable bandwidth parameter. The proposed approach combines the advantages of ILPP in preserving manifold data structures and those of support vector data description (SVDD) in handling complex process data distributions. Case studies on helix data, hot strip mill, and Pensim benchmark processes demonstrate the utility and feasibility of the proposed approach. The average fault detection rate for proposed ILPP is 99%, which is higher than locality preserving projection (LPP; 87.8%), local tangent space alignment (LTSA; 74.9%), and principal component analysis (PCA; 90.6%).

Spectral radius-based interval principal component analysis (SR-IPCA) for fault detection in industrial processes with imprecise data

Data-driven process monitoring approaches like principal component analysis (PCA) have been widely used in many industrial processes, most of which assume that the data are precise and reliable. However, the data collected from actual processes are usually contaminated with uncertainties due to measurement noise, severe working scenarios and other reasons. Aiming at fault detection in the complex industrial processes with imprecise measurement data, a spectral radius-based interval PCA (SR-IPCA) method is proposed in this work. First, kernel density estimation (KDE)-based measurement error estimation strategy is presented to transform imprecise single-valued data into interval-valued data. Then, a spectral radius based eigen-decomposition method for interval matrix is developed to extract the process features by projecting high-dimensional interval data to low-dimensional space. Moreover, four monitoring statistics are defined to analyze process status and identify the fault. The proposed SR-IPCA method can effectively tackle the uncertainties by defining an interval domain and extract correlation structure with relative low computation complexity. The simulation results on the synthetic data sets and the Tennessee Eastman process (TEP) reveal that SR-IPCA significantly reduces the false alarm rate and the missed detection rate compared with PCA, vertices PCA (V-PCA) and centers PCA (C-PCA).