Comparing PCA-based fault detection methods for dynamic processes with correlated and Non-Gaussian variables

Michalski, Miguel Ângelo de Carvalho; Souza, Gilberto Francisco Martha de

doi:10.1016/j.eswa.2022.117989

Cited by 18 publications

(7 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This opens up applicability in many systems where PCA and related methods are used for change detection. While traffic is among these systems [12], other prominent examples for fault detection are chemical plants [43] and industrial machinery [9,44]. A first step to include multiple operational conditions in PCA-based failure detection for heat pumps was already undertaken by Zhang et al [45].…”

Section: Discussionmentioning

confidence: 99%

Sensitivity of principal components to system changes in the presence of non-stationarity

Bette,

Schreckenberg,

Guhr

2023

J. Stat. Mech.

View full text Add to dashboard Cite

Non-stationarity affects the sensitivity of change detection in correlated systems described by sets of measurable variables. We study this by projecting onto different principal components. Non-stationarity is modeled as multiple normal states that exist in the system even before a change occurs. The studied changes occur in mean values, standard deviations or correlations of the variables. Monte Carlo simulations are performed to test the sensitivity for change detection with and without knowledge about non-stationarity for different system dimensions and numbers of normal states. A comparison clearly shows that knowledge about the non-stationarity of the system greatly improves change detection sensitivity for all principal components. This improvement is largest for those components that already provide the greatest possibility for change detection in the stationary case. We illustrate our results with an example using real traffic flow data, in which we detect a weekend and a bank holiday start as anomalies.

show abstract

Section: Discussionmentioning

confidence: 99%

Sensitivity of principal components to system changes in the presence of non-stationarity

Bette,

Schreckenberg,

Guhr

2023

J. Stat. Mech.

View full text Add to dashboard Cite

show abstract

“…The domain of qualitative designs methodologies pertaining to FDI encompasses a noteworthy discourse centred on the subject of multivariate statistical procedure tracking. Within this particular framework, methodologies that involve principal component analysis (PCA) [ 24 ] as well as partial least squares (PLS) [ 25 ] have garnered significant recognition in the realm of industrial application due to their remarkable effectiveness in detecting and diagnosing faults. These methodologies utilise a dual-phase procedure: firstly, the multivariate as well as collinear information is projected onto a smaller space of reduced dimensions, subsequently leading to the formulation of test statistics including T 2 and SPE, which serve as efficient monitors for the multivariate data.…”

Section: Introductionmentioning

confidence: 99%

“…By utilizing the ACSA, the proposed approach takes advantage of its ability to dynamically adapt the search parameters, striking a balance between exploration and exploitation. This adaptive nature allows our method to effectively handle the complexities and variations present in real-world industrial systems, leading to improved fault detection capabilities as compared to other statistical approaches presented in [ 24 , 50 , 51 ].…”

Section: Introductionmentioning

confidence: 99%

“…This adaptive modification ensures that the chosen attributes are in line with the changing traits of the TEP, hence improving the overall resilience of our approach. Comparative assessments comparing statistical approaches in [ 24 , 50 , 51 , 56 ] procedures with ACSA demonstrate the shortcomings of traditional methods in adjusting to dynamic industrial processes, hence emphasizing the suitability of ACSA in our specific situation.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

An adaptive metaheuristic optimization approach for Tennessee Eastman process for an industrial fault tolerant control system

Mustafa,

Ahmed,

Basit

et al. 2024

PLoS ONE

View full text Add to dashboard Cite

The Tennessee Eastman Process (TEP) is widely recognized as a standard reference for assessing the effectiveness of fault detection and false alarm tracking methods in intricate industrial operations. This paper presents a novel methodology that employs the Adaptive Crow Search Algorithm (ACSA) to improve fault identification capabilities and mitigate the occurrence of false alarms in the TEP. The ACSA is an optimization approach that draws inspiration from the observed behavior of crows in their natural environment. This algorithm possesses the capability to adapt its search behavior in response to the changing dynamics of the optimization process. The primary objective of our research is to devise a monitoring strategy that is adaptable in nature, with the aim of efficiently identifying faults within the TEP while simultaneously minimizing the occurrence of false alarms. The ACSA is applied in order to enhance the optimization of monitoring variables, alarm thresholds, and decision criteria selection and configuration. When compared to traditional static approaches, the ACSA-based monitoring strategy is better at finding faults and reducing false alarms because it adapts well to changes in process dynamics and disturbances. In order to assess the efficacy of our suggested methodology, we have conducted comprehensive simulations on the TEP dataset. The findings suggest that the monitoring strategy based on ACSA demonstrates superior fault identification rates while concurrently mitigating the frequency of false alarms. In addition, the flexibility of ACSA allows it to efficiently manage process variations, disturbances, and uncertainties, thereby enhancing its robustness and reliability in practical scenarios. To validate the effectiveness of our proposed approach, extensive simulations were conducted on the TEP dataset. The results indicate that the ACSA-based monitoring strategy achieves higher fault detection rates while simultaneously reducing the occurrence of false alarms. Moreover, the adaptability of ACSA enables it to effectively handle process variations, disturbances, and uncertainties, making it robust and reliable for real-world applications. The contributions of this research extend beyond the TEP, as the adaptive monitoring strategy utilizing ACSA can be applied to other complex industrial processes. The findings of this study provide valuable insights into the development of advanced fault detection and false alarm monitoring techniques, offering significant benefits in terms of process safety, reliability, and operational efficiency.

show abstract

“…The majority of traditional process monitoring methods are rooted in multivariate statistical analysis techniques, including Principal Component Analysis (PCA), Partial Least Squares (PLS), Canonical Variate Analysis (CVA), Canonical Variate Dissimilarity Analysis (CVDA), and Canonical Correlation Analysis (CCA), which have been extensively researched by a wide range of scholars and have yielded noteworthy research achievements and a plethora of successful applications [1]. For instance, references [2][3][4][5][6][7] have, respectively, investigated and improved the PCA method to varying extents and achieved successful applications in the industry; Ding et al applied an enhanced PLS to predict and diagnose key performance indicators of industrial hot-rolled strip steel mills [8], and a series of researches were similarly carried out for the PLS method in [9][10][11][12][13]; Ruiz-Cárcel and colleagues achieved satisfactory experimental results in the process monitoring of multiphase flow facilities using the CVA method [14]; subsequently, Pilario et al proposed the CVDA method and its extended version based on CVA, offering new insights into incipient fault detection in dynamic systems [15][16][17]; as for the CCA method, Chen et al pioneered the use of data-driven CCA technology for generating residuals based on canonical correlation, applying it to fault detection in both static and dynamic processes [18]; following this, the CCA method has gradually attracted attention from scholars in the field of process monitoring and fault detection, undergoing extensive research and improvement [19][20][21][22][23][24][25][26][27][28].…”

Section: Introductionmentioning

confidence: 99%

Gap-MK-DCCA-Based Intelligent Fault Diagnosis for Nonlinear Dynamic Systems

Wu,

Zhang,

Chen

2024

Processes

View full text Add to dashboard Cite

In intelligent process monitoring and fault detection of the modern process industry, conventional methods mostly consider singular characteristics of systems. To tackle the problem of suboptimal incipient fault detection in nonlinear dynamic systems with non-Gaussian distributed data, this paper proposes a methodology named Gap-Mixed Kernel-Dynamic Canonical Correlation Analysis. Initially, the Gap metric is employed for data preprocessing, followed by fault detection utilizing the Mixed Kernel-Dynamic Canonical Correlation Analysis. Ultimately, fault identification is conducted through a contribution method based on the T2 statistic. Furthermore, a comparative analysis was conducted using Canonical Variate Analysis, Dynamic Canonical Correlation Analysis, and Mixed Kernel-Dynamic Canonical Correlation Analysis on the Tennessee Eastman process. Experimental results indicate varying degrees of improvements in the detection rate, false alarm rate, missed detection rate, and detection time compared to the comparative methods, demonstrating the industrial value and academic significance of the method.

show abstract

Comparing PCA-based fault detection methods for dynamic processes with correlated and Non-Gaussian variables

Cited by 18 publications

References 56 publications

Sensitivity of principal components to system changes in the presence of non-stationarity

Sensitivity of principal components to system changes in the presence of non-stationarity

An adaptive metaheuristic optimization approach for Tennessee Eastman process for an industrial fault tolerant control system

Gap-MK-DCCA-Based Intelligent Fault Diagnosis for Nonlinear Dynamic Systems

Contact Info

Product

Resources

About