Enhanced canonical variate analysis with slow feature for dynamic process status analytics

SFNet: A slow feature extraction network for parallel linear and nonlinear dynamic process monitoring

Fault detection and fault identification are consecutive steps of multivariate statistical process monitoring. In recent years, increasing attention has been paid to process dynamics. For dynamic process modeling, canonical variate analysis (CVA) extracts process dynamics effectively. However, process noises are not well analyzed in traditional CVA and corresponding fault identification methods are less studied. To solve these issues, a variational Bayesian CVA (VBCVA) model is proposed for dynamic process monitoring. Through a probabilistic perspective, the inevitable noises in realistic industrial processes can be captured in the new model. Moreover, the proposed model is further extended in the variational Bayesian framework to overcome the common problems in probabilistic methods. Besides, an improved fault identification approach based on fault relevance is introduced, which avoids the smearing effect caused by data reconstruction. Finally, the feasibility of the proposed process monitoring scheme is verified on the TE benchmark and a real wastewater treatment process.

For the problem that the early fault information of diesel engine system is weak and difficult to identify and diagnose, an early fault diagnosis method based on cross-point frequency response and static and dynamic information fusion was proposed for the assembly quality of diesel engine system. The dynamic vibration response signal and static cross-point frequency response signal of the diesel engine system were collected by reasonable layout of measuring points. After CEEMD reconstruction and de-noising, the sample entropy and approximate entropy were extracted as characteristic parameters of the dynamic signal, and the frequency response features were extracted from the static signal. The static and dynamic information of the two kinds of information was integrated by PCA. The optimized support vector machine is used to identify the dynamic information and the static and dynamic fusion information respectively. The results show that this method can effectively detect the assembly quality of key components of diesel engine system, and the accuracy of diagnosis is up to 95%, and the recognition rate after static and dynamic information fusion is better than that of dynamic information. The method presented in this paper has a good application prospect in the assembly quality inspection and early fault diagnosis of diesel engine system.

A new soft-sensor algorithm with concurrent consideration of slowness and quality interpretation for dynamic chemical process

Physics-informed probabilistic slow feature analysis

A declarative approach to detecting design patterns from Java execution traces and source code

Modern industrial processes are characterized by increasing complexity, often exhibiting pronounced dynamic behaviors and significant nonlinearity. Addressing these dynamic and nonlinear characteristics is essential for effective process monitoring. However, many existing methods for process monitoring and fault diagnosis are insufficient in handling these challenges. In this article, we present a novel process monitoring approach, CVA-DisDAE, which integrates an improved Denoising Autoencoder (DAE) with Canonical Variate Analysis (CVA) to address the challenges posed by dynamic behaviors and nonlinear relationships in industrial processes. First, CVA is employed to reduce data dimensionality and minimize information redundancy by maximizing correlations between past and future observations, thereby effectively capturing process dynamics. Following this, we introduce a discriminative DAE model (DisDAE) designed to serve as a semi-supervised denoising autoencoder for precise feature extraction. This is achieved by incorporating both between-class separability and within-class variability into the traditional DAE framework. The key distinction between the proposed DisDAE and the conventional DAE lies in the integration of a linear discriminant analysis (LDA) penalty into the DAE’s loss function, resulting in extracted features that are more conducive to fault classification. Finally, we validate the effectiveness of the proposed semi-supervised dynamic process monitoring approach through its application to the Tennessee Eastman benchmark process, demonstrating its superior performance.

Data‐driven soft sensor models have been extensively utilized in industrial processes. Batch processes are usually employed to manufacture low‐volume and high value‐added products in chemical, materials, and pharmaceutical industries. The most distinctive features of batch process lie in nonlinear, repetition, and slow time varying characteristics. In this paper, a data‐driven soft sensor modelling method based on linear slow feature analysis (LSFA) and least squares support vector regression (LSSVR) is proposed. In this method, LSFA was used to effectively capture the driving force behind the data features that change as slowly as possible. Then, a LSSVR model was constructed between the extracted slow feature variables and quality variables. Finally, a numerical example, industrial penicillin fermentation processes, and cobalt oxalate synthesis process were utilized to confirm the prediction accuracy and model reliability of the proposed approach.

Process monitoring of incipient faults, as opposed to abrupt faults, in an industrial process is increasingly becoming more important. These are slowly developing faults that may eventually lead to severe abnormal conditions, and ultimately, failure of a critical component. Data-driven multivariate statistical process monitoring (MSPM) methods are extensively studied and widely used for abrupt fault detection. One such method is Canonical Variate Analysis (CVA), which has strength for dynamic process monitoring. This paper now aims to demonstrate the effectiveness of CVA in detecting incipient faults — in particular, catalyst decay and heat transfer fouling occurring in a temperature-controlled CSTR, simulated separately and simultaneously. Performance is evaluated by noting detection delays and investigating the effects of process control. The results show the sensitivity of CVA to incipient faults, providing an early detection even in the case of multiple faults that mask each other.

This paper proposes novel framework for facial expressions analysis using dynamic and static information in video sequences. First, based on incremental formulation, discriminative deformable face alignment method is adapted to locate facial points to correct in‐plane head rotation and break up facial region from background. Then, spatial‐temporal motion local binary pattern (LBP) feature is extracted and integrated with Gabor multiorientation fusion histogram to give descriptors, which reflect static and dynamic texture information of facial expressions. Finally, a one‐versus‐one strategy based multiclass support vector machine (SVM) classifier is applied to classify facial expressions. Experiments on Cohn‐Kanade (CK) + facial expression dataset illustrate that integrated framework outperforms methods using single descriptors. Compared with other state‐of‐the‐art methods on CK+, MMI, and Oulu‐CASIA VIS datasets, our proposed framework performs better.

How to fuse static and dynamic information is a key issue in event analysis. In this paper, a top-down motion guided fusing method is proposed for recognizing events in an unconstrained news video. In the method, the static information is represented as a Bag-of-SIFT-features and motion information is employed to generate event specific attention map to direct the sampling of the interest points. We build class-specific motion histograms for each event so as to give more weight on the interest points that are discriminative to the corresponding event. Experimental results on TRECVID 2005 video corpus demonstrate that the proposed method can improve the mean average accuracy of recognition.

Dynamic analysis extracts information that describes the structure of the runtime behavior of a software system. This information, typically represented in the form of execution traces, is useful in many software modernization activities, such as reverse engineering and program analysis. The Architecture-Driven Modernization initiative has defined standards to support the modernization process in the context of Model-Driven Engineering (MDE). However, to date, there is no standard for representing runtime information that allows the specification of dynamic information independent of any programming language. In light of this, we present TRACEM, a metamodel for representing trace information under a standard representation that would allow traces to become a new domain in software engineering. The purpose of defining TRACEM is to complement an MDE framework for software modernization that aims to integrate static and dynamic analysis techniques during the reverse engineering process. This article contains a description of TRACEM and a case study that illustrates how dynamic information combined with static information allows for improving the whole reverse engineering process.

A new dissimilarity method integrating multidimensional mutual information and independent component analysis for non-Gaussian dynamic process monitoring

Modeling in the observable or latent space? A comparison of dynamic latent variable based monitoring methods for sensor fault detection

Real-time monitoring of source zone of light non-aqueous liquids (LNAPLs) is important for preventing accidental pollution and taking effective underground pollution remediation measures. As a high-precision monitoring technology, cross-hole electrical resistivity tomography (CHERT) has been widely used to obtain static information regarding underground stratigraphic structures and dynamic information regarding fluid motion. The time-lapse data processing method can be used to obtain pollution zone dynamic change information. Since the leakage rate directly affects the diffusion range of pollution, this study simulated the initial evolution process of LNAPL pollution source regions under two different leakage rates. The time-lapse monitoring of the above process was performed using CHERT. The test results show that CHERT can be used to observe the migration of LNAPLs and the initial evolution of the contaminated area. Differences in leakage rate will result in variation in soil wettability and fluid distribution, which will cause changes in soil resistivity in the corresponding region. The low-saturation LNAPL-contaminated area may exhibit low-resistivity characteristics and is easily overlooked in field investigations. In addition, the degree of contamination in the saturated zone can be quantitatively evaluated by CHERT; however, the pollution range and oil saturation value determined by CHERT are underestimated. The results showed the electrical variation characteristics of the initial evolution process of the fresh pollution source area and provide data that will enable early warnings of LNAPL leakage. This shows that CHERT is a promising tool for monitoring LNAPL pollution source leakage even if further research is needed to fully understand the effect of hydrological processes on electronic signals.