A single Bayesian network classifier for monitoring with unknown classes

Atoui, Mohamed Amine; Cohen, Achraf; Verron, Sylvain; Kobi, Abdessamad

doi:10.1016/j.engappai.2019.07.016

Cited by 41 publications

(24 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…. x n } indicates that the parent node set of X i is multiplied by the probability distribution of Pa(Xi) variables to obtain the joint distribution [30], as shown in formula 2:…”

Section: Bayesian Network (Bn)mentioning

confidence: 99%

Bayesian Network-Based Risk Assessment of Single-Phase Grounding Accidents of Power Transmission Lines

Zhang

Bian

Zhao

et al. 2020

IJERPH

View full text Add to dashboard Cite

With the increasing demand for electricity transmission and distribution, single-phase grounding accidents, which cause great economic losses and casualties, have occurred frequently. In this study, a Bayesian network (BN)-based risk assessment model for representing single-phase grounding accidents is proposed to examine accident evolution from causes to potential consequences. The Bayesian network of single-phase grounding accidents includes 21 nodes that take into account the influential factors of environment, management, equipment and human error. The Bow-tie method was employed to build the accident evolution path and then converted to a BN. The BN conditional probability tables are determined with reference to historical accident data and expert opinion obtained by the Delphi method. The probability of a single-phase grounding accident and its potential consequences in normal conditions and three typical accident scenarios are analyzed. We found that “Storm” is the most critical hazard of single-phase grounding, followed by “Aging” and “Icing”. This study could quantitatively evaluate the single-phase grounding accident in multi-hazard coupling scenarios and provide technical support for occupational health and safety management of power transmission lines.

show abstract

“…. x n } indicates that the parent node set of X i is multiplied by the probability distribution of Pa(Xi) variables to obtain the joint distribution [30], as shown in formula 2:…”

Section: Bayesian Network (Bn)mentioning

confidence: 99%

Bayesian Network-Based Risk Assessment of Single-Phase Grounding Accidents of Power Transmission Lines

Zhang

Bian

Zhao

et al. 2020

IJERPH

View full text Add to dashboard Cite

show abstract

“…In [18], both sensor data and residual data are used as input to a tree augmented naive Bayes fault classifier. In [6], a conditional Gaussian network is proposed to handle both known and unknown fault classes. In [19], feature selection using neural networks is applied before training the fault classifiers.…”

Section: B Related Researchmentioning

confidence: 99%

“…There are multiple methods proposed for one-class classification, for example probabilistic models, one-class support vector machines (OSVM), and isolation forests (iForests) [28]. Probabilistic models use probability distributions to model data from one class and detect outliers, with respect to that class, when the likelihood of a sample is small, see for example [6]. Non-probabilistic models, such as OSVM and iForests, model a decision boundary that encapsulates training data to determine if new data can be explained by that class or not.…”

Section: A Using One-class Classifiers For Modeling Fault Classesmentioning

confidence: 99%

“…The probability of a sample to belong to an unknown fault class, denoted f x , is difficult to model without prior information. In [6], [23], there are no probability models of the unknown fault class f x . Instead, f x is selected when the probabilities of all known fault classes are below some threshold.…”

Section: Fault Classification Of Time-series Data Using Bayesianmentioning

confidence: 99%

“…Another complicating factor in data-driven fault diagnosis is collecting representative training data from all relevant faults. Data collection is an expensive and time-consuming process and not feasible in many applications [5], [6]. Especially, since many faults do not occur until after years of operation.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Data-Driven Open-Set Fault Classification of Residual Data Using Bayesian Filtering

Jung

2020

IEEE Trans. Contr. Syst. Technol.

View full text Add to dashboard Cite

Data-driven fault classification in industrial applications is complicated by unknown fault classes and limited training data. In addition, different faults can have similar effects on sensor outputs resulting in fault classification ambiguities, i.e. multiple fault hypotheses can explain the data. One solution is to identify and rank all plausible fault classes which give useful information, for example at a workshop when performing troubleshooting. A probabilistic fault classification algorithm is proposed for residual data classification combining Weibullcalibrated one-class support vector machines for fault class modeling and Bayesian filtering for time-series analysis. The fault classifier ranks different fault classes and can identify sequences from unknown fault realizations, i.e. faults not represented in training data. Real residual data computed from sensor data and model analysis of an internal combustion engine are used as a case study illustrating the usefulness of the proposed method.

show abstract

Applying Verbal Decision Analysis for Public Research and Technology Advancement

Polevoi

Проничкин

Chernyshova

2023

Software Engineering Application in Systems Design

View full text Add to dashboard Cite

A single Bayesian network classifier for monitoring with unknown classes

Cited by 41 publications

References 34 publications

Bayesian Network-Based Risk Assessment of Single-Phase Grounding Accidents of Power Transmission Lines

Bayesian Network-Based Risk Assessment of Single-Phase Grounding Accidents of Power Transmission Lines

Data-Driven Open-Set Fault Classification of Residual Data Using Bayesian Filtering

Applying Verbal Decision Analysis for Public Research and Technology Advancement

Contact Info

Product

Resources

About