Bayes linear variance structure learning for inspection of large scale physical systems

Randell, David; Goldstein, Michael; Jonathan, Philip

doi:10.1177/1748006x13492955

Cited by 2 publications

(2 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This method is applied in an opinion pooling framework by Wisse et al, 8 while Quigley et al 9 demonstrate how the Bayes linear method can be used to estimate correlation in a setting with an underlying Poisson process. Randell et al 10 discuss application of this method to the inspection of large-scale physical systems. However, it is commonly regarded that specification of a full probability distribution over uncertain events is possible, as discussed, for instance, in French.…”

Section: Introductionmentioning

confidence: 99%

Reliability updating in linear opinion pooling for multiple decision makers

Bolger

Houlding

2016

Proceedings of the Institution of Mechanical Engineers, Part O:

View full text Add to dashboard Cite

Accurate information sources are vital prerequisites for good decision making. In this thesis we consider a multiple participant setting, where all decision makers (DMs) have a collection of neighbours with whom they share their beliefs about some common relevant uncertain quantity. When determining which course of action to follow a DM takes into account all the information received from her neighbours. Over time, in light of the returns observed from choices made, DMs update their own beliefs over the uncertain event, and also adjust the degree of consideration that they afford to the opinions of each neighbour based on the level of reliability that the information they provide is ascertained to have. Much of this thesis is concerned with constructing a method that incorporates both of these learning facets in a dynamic fashion. This technique, termed the Plug-in approach, is motivated and derived, and attempts are made to justify its use by consideration of some attractive properties it obeys, in addition to studies conducted using both simulated and real data which compared its performance to some rational alternatives. Generalisations of this method are also provided to a setting where DMs specify their opinions nonparametrically rather than using probability distributions, as well as in a group setting where utilities as well as opinions must be amalgamated. Two subjective approaches are also briefly discussed, before we conclude with numerous suggestions for further research in this field.

show abstract

Section: Introductionmentioning

confidence: 99%

Reliability updating in linear opinion pooling for multiple decision makers

Bolger

Houlding

2016

Proceedings of the Institution of Mechanical Engineers, Part O:

View full text Add to dashboard Cite

show abstract

“…As electronic systems are large and complex, anomalies and faults are governed by different failure mechanisms. 3 Because of the dynamic behavior, the multiple functions, the complex structures, and the numerous non-line and uncertain parameters, 4 an entire fault diagnostic analysis with direct inference can lead to a large inefficient knowledge base and poor diagnostic quality. 5 Therefore, it is often difficult and inefficient to perform detailed fault diagnostics directly on complete complex electronic systems.…”

Section: Introductionmentioning

confidence: 99%

Sensory information fusion–based fault diagnostics for complex electronic systems

2015

Proceedings of the Institution of Mechanical Engineers, Part O:

View full text Add to dashboard Cite

As the structure of complex electronic systems is large, there is inherent uncertainty, so it is often inefficient or infeasible to perform detailed fault diagnostics directly. Fault diagnostics for complex electronic systems need to be performed based on the sensor monitoring data, so in this article, sensory information fusion is introduced to minimize diagnostic uncertainty and computational complexity. Sensory information fusion combines the sensor data by measuring the mutuality of the information and dynamically selecting the most diagnostic decision-relevant sensor subset. This article proposes a sensory information fusion-based diagnostic methodology and framework, which is made up of a dynamical sensor selection that combines sensory information and detailed fault diagnostics based on the sensory information fusion results. A numerical example is conducted to illustrate the proposed sensory information fusion-based diagnostics. The diagnostic results as well as comparisons with different training data sizes confirm that the proposed sensory information fusion-based diagnostics is superior when dealing with uncertain and incomplete information. Furthermore, the sensory information fusion-based diagnostics is shown to simplify the diagnostic complexity and accurately and efficiently solve the diagnostic problems for complex electronic systems. The proposed methodology is also flexible enough to accommodate other fault diagnostic methods so as to perform applicable diagnostics for other complex electronic systems of varying sizes.

show abstract

Bayes linear variance structure learning for inspection of large scale physical systems

Cited by 2 publications

References 30 publications

Reliability updating in linear opinion pooling for multiple decision makers

Reliability updating in linear opinion pooling for multiple decision makers

Sensory information fusion–based fault diagnostics for complex electronic systems

Contact Info

Product

Resources

About