Practical options for selecting data-driven or physics-based prognostics algorithms with reviews

This paper is to provide practical options for prognostics so that beginners can select appropriate methods for their fields of application. To achieve this goal, several popular algorithms are first reviewed in the data-driven and physics-based prognostics methods. Each algorithm’s attributes and p...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	Reliability engineering & system safety Ročník 133; s. 223 - 236
Hlavní autoři:	An, Dawn, Kim, Nam H., Choi, Joo-Ho
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	Oxford Elsevier Ltd 01.01.2015 Elsevier
Témata:	Algorithms Applied sciences Bayesian inference Bias Complexity Data-driven prognostics Exact sciences and technology Fatigue failure Fracture mechanics (crack, fatigue, damage...) Fundamental areas of phenomenology (including applications) Gaussian process regression Mathematics Neural network Neural networks Noise Operational research and scientific management Operational research. Management science Parametric inference Particle filter Physics Physics-based prognostics Probability and statistics Reliability engineering Reliability theory. Replacement problems Sampling theory, sample surveys Sciences and techniques of general use Solid mechanics Statistics Structural and continuum mechanics Physics-based prognostics Gaussian process regression Neural network Bayesian inference Data-driven prognostics Particle filter Availability Bayes estimation Monte Carlo method Parameter estimation Contour line Regression analysis Review Data driven modelling Modeling Fatigue crack Noise level Crack propagation Particle method Gaussian process Model matching Covariance Physical model Damaging
ISSN:	0951-8320, 1879-0836
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Popis
Shrnutí:	This paper is to provide practical options for prognostics so that beginners can select appropriate methods for their fields of application. To achieve this goal, several popular algorithms are first reviewed in the data-driven and physics-based prognostics methods. Each algorithm’s attributes and pros and cons are analyzed in terms of model definition, model parameter estimation and ability to handle noise and bias in data. Fatigue crack growth examples are then used to illustrate the characteristics of different algorithms. In order to suggest a suitable algorithm, several studies are made based on the number of data sets, the level of noise and bias, availability of loading and physical models, and complexity of the damage growth behavior. Based on the study, it is concluded that the Gaussian process is easy and fast to implement, but works well only when the covariance function is properly defined. The neural network has the advantage in the case of large noise and complex models but only with many training data sets. The particle filter and Bayesian method are superior to the former methods because they are less affected by noise and model complexity, but work only when physical model and loading conditions are available. •Practical review of data-driven and physics-based prognostics are provided.•As common prognostics algorithms, NN, GP, PF and BM are introduced.•Algorithms’ attributes, pros and cons, and applicable conditions are discussed.•This will be helpful to choose the best algorithm for different applications.
Bibliografie:	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23
ISSN:	0951-8320 1879-0836
DOI:	10.1016/j.ress.2014.09.014