A Survey on Diagnosis Methods Combining Dynamic Systems Structural Analysis and Machine Learning
Résumé
This paper reviews diagnosis methods that com-
bine dynamic systems structural analysis and ma-
chine learning. A corpus of related articles
has been constituted using a thorough research
methodology. Three main families of recent re-
search papers have been identified: residual se-
lection methods, residual generation techniques
and methods using the structural analysis output
to train a machine learning model. A detailed ex-
planation of how each article tackles the diagnosis
problem is given. The way these methods make
up for structural analysis and machine learning
drawbacks by combining them is analyzed.
Domaines
Informatique [cs]
Fichier principal
A_survey_on_diagnosis_methods_combining_dynamic_systems_structural_analysis_and_machine_learning.pdf (554.56 Ko)
Télécharger le fichier
Origine | Fichiers produits par l'(les) auteur(s) |
---|