A fault detection system based on unsupervised techniques for industrial control loops

Jove, Esteban; Casteleiro‐Roca, José‐Luis; Quintián, Héctor; Méndez‐Pérez, Juan Albino; Calvo-Rolle, José Luís

doi:10.1111/exsy.12395

Cited by 38 publications

(17 citation statements)

References 40 publications

(36 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…BHL has also been previously employed in the analysis of the internal structure of a series of datasets [35,36], providing a clear projection of the original dataset. More specifically, it has been successfully applied to Android malware datasets [33,34], where its task was to characterize Android malware families.…”

Section: Literature Reviewmentioning

confidence: 99%

Intrusion Detection with Unsupervised Techniques for Network Management Protocols over Smart Grids

et al. 2020

Self Cite

View full text Add to dashboard Cite

The present research work focuses on overcoming cybersecurity problems in the Smart Grid. Smart Grids must have feasible data capture and communications infrastructure to be able to manage the huge amounts of data coming from sensors. To ensure the proper operation of next-generation electricity grids, the captured data must be reliable and protected against vulnerabilities and possible attacks. The contribution of this paper to the state of the art lies in the identification of cyberattacks that produce anomalous behaviour in network management protocols. A novel neural projectionist technique (Beta Hebbian Learning, BHL) has been employed to get a general visual representation of the traffic of a network, making it possible to identify any abnormal behaviours and patterns, indicative of a cyberattack. This novel approach has been validated on 3 different datasets, demonstrating the ability of BHL to detect different types of attacks, more effectively than other state-of-the-art methods.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Intrusion Detection with Unsupervised Techniques for Network Management Protocols over Smart Grids

et al. 2020

Self Cite

View full text Add to dashboard Cite

show abstract

“…In this part, a numerical simulation model proposed by Acala et al in 2009 is used to test the fault detection power of the proposed method [30]. The model is constructed as (13):…”

Section: B Numerical Simulationmentioning

confidence: 99%

“…The testing data are generated by the following formula (14): where x * is the normal data obtained by formula (13). The direction ξ i is randomly chosen from six possible variable directions with uniform probability, and f is the fault magnitude following N ∼ (5, 0.1).…”

Section: B Numerical Simulationmentioning

confidence: 99%

“…Unsupervised process monitoring methods have been researched in recent decades. Jove et al described a novel approach using a visual tool for the detection of faults in industrial processes via unsupervised and projectionist techniques [13]. Escobar et al proposed a combined generative topographic mapping and graph theory approach for unsupervised nonlinear data visualization and fault identification [14].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Graph-Based Method for Fault Detection in the Iron-Making Process

Yang

Pan

2020

IEEE Access

View full text Add to dashboard Cite

Since the iron-making process is performed in complicated environments and controlled by operators, observation labeling is difficult and time-consuming. Therefore, unsupervised fault detection methods are a promising research topic. Recently, an unsupervised graph-based change point detection method has been introduced, and the graph of observations is constructed by the minimum spanning tree. In this paper, a novel fault detection method based on the graph for an iron-making process is proposed, and a weight calculation method for constructing the minimum spanning tree is introduced. The Euclidean distance and Mahalanobis distance are combined to calculate the weights in the minimum spanning tree, which contain important relations of variables. The distance calculation method is determined by the correlation coefficients of variables. Each testing observation is set as a change point candidate, and a change point candidate divides the observations into two groups. The number of a special type of edge in the minimum spanning tree is used as a fault detection statistic. That special edge connects two observations from two different groups. The minimum number of that type of edge corresponding to the change point candidate is a true change point. Finally, numerical simulation is used to test the power of the proposed method, and a real iron-making process including low stock, cooling, and slip faults is implemented to illustrate the effectiveness of fault detection in industrial processes.

show abstract

“…In the paper by Jove, Casteleiro‐Roca, Quintián, Méndez‐Pérez, and Calvo‐Rolle (), a novel approach for fault detection in industrial processes, by means of unsupervised and projectionist techniques, is proposed. The proposed system is tested in a laboratory plant built with industrial equipment where anomalies have to be detected without any previous knowledge of the data.…”

Section: Contents Of the Special Issuementioning

confidence: 99%

Soft computing applications in the field of industrial and environmental enterprises

Jiménez

Herrero

2019

Expert Systems

View full text Add to dashboard Cite

A fault detection system based on unsupervised techniques for industrial control loops

Cited by 38 publications

References 40 publications

Intrusion Detection with Unsupervised Techniques for Network Management Protocols over Smart Grids

Intrusion Detection with Unsupervised Techniques for Network Management Protocols over Smart Grids

Graph-Based Method for Fault Detection in the Iron-Making Process

Soft computing applications in the field of industrial and environmental enterprises

Contact Info

Product

Resources

About