Bayesian Networks in Fault Diagnosis

Cai, Baoping; Huang, Lei; Xie, Min

doi:10.1109/tii.2017.2695583

Cited by 382 publications

(110 citation statements)

References 121 publications

Supporting

Mentioning

108

Contrasting

Unclassified

Order By: Relevance

“…The BN is intuitive as we can see that a sensor value in S is influenced by both F 1 and F 2 . From this BN, we may consider the dependencies between variables to correspond to causal relationships, as it is logical reasoning to suggest that underlying faults in a system will have an impact on sensor values [12]. Fig.…”

Section: B Bayesian Networkmentioning

confidence: 99%

A Bayesian network based learning system for modelling faults in large-scale manufacturing

Carbery¹,

Woods²,

Marshall³

2018

2018 IEEE International Conference on Industrial Technology (ICIT)

View full text Add to dashboard Cite

Abstract-Manufacturing companies can benefit from the early prediction and detection of failures to improve their product yield and reduce system faults through advanced data analytics. Whilst an abundance of data on their processing systems exist, they face difficulties in using it to gain insights to improve their systems. Bayesian networks (BNs) are considered here for diagnosing and predicting faults in a large manufacturing dataset from Bosch. Whilst BN structure learning has been performed traditionally on smaller sized data, this work demonstrates the ability to learn an appropriate BN structure for a large dataset with little information on the variables, for the first time. This paper also demonstrates a new framework for creating an appropriate probabilistic model for the Bosch dataset through the selection of statistically important variables on the response; this is then used to create a BN network which can be used to answer probabilistic queries and classify products based on changes in the sensor values in the production process.

show abstract

Section: B Bayesian Networkmentioning

confidence: 99%

A Bayesian network based learning system for modelling faults in large-scale manufacturing

Carbery¹,

Woods²,

Marshall³

2018

2018 IEEE International Conference on Industrial Technology (ICIT)

View full text Add to dashboard Cite

show abstract

“…The advantage with respect to a classical probabilistic temporal model like Markov chains is that the dynamic Bayesian networks are stochastic transition models factored over a number of random variables, over which a set of conditional dependency assumption is defined [4][5][6]. We adopt dynamic Bayesian networks to predict the future state of variables taking into consideration the observation of variables up to now.…”

Section: Series Parallel and Voting Systemmentioning

confidence: 99%

A novel critical infrastructure resilience assessment approach using dynamic Bayesian networks

Cai

Xie

Liu

et al. 2017

AIP Conference Proceedings

Self Cite

View full text Add to dashboard Cite

Abstract. The word resilience originally originates from the Latin word "resiliere", which means to "bounce back". The concept has been used in various fields, such as ecology, economics, psychology, and society, with different definitions. In the field of critical infrastructure, although some resilience metrics are proposed, they are totally different from each other, which are determined by the performances of the objects of evaluation. Here we bridge the gap by developing a universal critical infrastructure resilience metric from the perspective of reliability engineering. A dynamic Bayesian networks-based assessment approach is proposed to calculate the resilience value. A series, parallel and voting system is used to demonstrate the application of the developed resilience metric and assessment approach.

show abstract

“…Normally, the computer software locates the liquid level by extracting the reflected pulse, so for an algorithm, the ability of identifying the acoustic pulse is what we are concerned about. In order to quantitatively reveal the identification ability of FFT and ACF, the Crest Factor is introduced as an evaluation index [46,47]. Crest Factor (aka Peak-to-Average Ratio) is defined as peak value divided by the effective value of a signal [48]:…”

Section: Experimental Schemesmentioning

confidence: 99%

A Novel Acoustic Liquid Level Determination Method for Coal Seam Gas Wells Based on Autocorrelation Analysis

Zhang

Fan

et al. 2017

Energies

View full text Add to dashboard Cite

Abstract:In coal seam gas (CSG) wells, water is periodically removed from the wellbore in order to keep the bottom-hole flowing pressure at low levels, facilitating the desorption of methane gas from the coal bed. In order to calculate gas flow rate and further optimize well performance, it is necessary to accurately monitor the liquid level in real-time. This paper presents a novel method based on autocorrelation function (ACF) analysis for determining the liquid level in CSG wells under intense noise conditions. The method involves the calculation of the acoustic travel time in the annulus and processing the autocorrelation signal in order to extract the weak echo under high background noise. In contrast to previous works, the non-linear dependence of the acoustic velocity on temperature and pressure is taken into account. To locate the liquid level of a coal seam gas well the travel time is computed iteratively with the non-linear velocity model. Afterwards, the proposed method is validated using experimental laboratory investigations that have been developed for liquid level detection under two scenarios, representing the combination of low pressure, weak signal, and intense noise generated by gas flowing and leakage. By adopting an evaluation indicator called Crest Factor, the results have shown the superiority of the ACF-based method compared to Fourier filtering (FFT). In the two scenarios, the maximal measurement error from the proposed method was 0.34% and 0.50%, respectively. The latent periodic characteristic of the reflected signal can be extracted by the ACF-based method even when the noise is larger than 1.42 Pa, which is impossible for FFT-based de-noising. A case study focused on a specific CSG well is presented to illustrate the feasibility of the proposed approach, and also to demonstrate that signal processing with autocorrelation analysis can improve the sensitivity of the detection system.

show abstract

Bayesian Networks in Fault Diagnosis

Cited by 382 publications

References 121 publications

A Bayesian network based learning system for modelling faults in large-scale manufacturing

A Bayesian network based learning system for modelling faults in large-scale manufacturing

A novel critical infrastructure resilience assessment approach using dynamic Bayesian networks

A Novel Acoustic Liquid Level Determination Method for Coal Seam Gas Wells Based on Autocorrelation Analysis

Contact Info

Product

Resources

About