Using Convolutional Neural Network (CNN) to learn by sensor signal classification critical local areas in a physical flexible shaft rotor system as an example of a complicated mechanical system.

Παπαγεωργίου, Παντελής-Παναγιώτης; Papageorgiou, Pantelis-Panagiotis

dc.contributor.author	Παπαγεωργίου, Παντελής-Παναγιώτης	el
dc.contributor.author	Papageorgiou, Pantelis-Panagiotis	en
dc.date.accessioned	2023-01-18T12:21:36Z
dc.date.available	2023-01-18T12:21:36Z
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/56759
dc.identifier.uri	http://dx.doi.org/10.26240/heal.ntua.24457
dc.description	Εθνικό Μετσόβιο Πολυτεχνείο--Μεταπτυχιακή Εργασία. Διεπιστημονικό-Διατμηματικό Πρόγραμμα Μεταπτυχιακών Σπουδών (Δ.Π.Μ.Σ.) “Ναυτική και Θαλάσσια Τεχνολογία και Επιστήμη”	el
dc.rights	Αναφορά Δημιουργού - Μη Εμπορική Χρήση - Παρόμοια Διανομή 3.0 Ελλάδα	*
dc.rights	Αναφορά Δημιουργού - Παρόμοια Διανομή 3.0 Ελλάδα	*
dc.rights.uri	http://creativecommons.org/licenses/by-sa/3.0/gr/	*
dc.subject	Vibration Signal Analysis	en
dc.subject	Convolution Neural Network	en
dc.subject	Machine Learning	en
dc.subject	Prognostic	en
dc.subject	Diagnostic	en
dc.subject	Προγνωστική	el
dc.subject	Διαγνωστική	el
dc.subject	Μηχανική Μάθηση	el
dc.subject	Ανάλυση Ταλαντώσεων	el
dc.subject	Τεχνητή Νοημοσύνη	el
dc.title	Using Convolutional Neural Network (CNN) to learn by sensor signal classification critical local areas in a physical flexible shaft rotor system as an example of a complicated mechanical system.	el
heal.type	masterThesis
heal.classification	Prognostic & Diagnostic of complex mechanical systems	en
heal.language	en
heal.access	free
heal.recordProvider	ntua	el
heal.publicationDate	2022-07-12
heal.abstract	Complex machinery contains critical areas, such as revolute joints and bearings, that are prone to damage initiation and growth. If not detected early, damage in complex local areas leads to premature failure. The complexity of an integrated system is a factor that limits developed classical methods from detecting early damage in complex local areas. A pure experimental data environment could provide solutions given the broad impact of machine learning. Here an interesting idea is introduced to support a machine-learning framework for damage detection in local critical areas. The vibration field developed in a local area surrounding a ball bearing support of a lab flexible shaft-rotor system was measured by a set of accelerometers to form a dataset environment. It was used as an experience for machine learning by a deep convolutional neural network adapted from the AlexNet Architecture. Our main result is casting a solid mechanics prediction problem into a classification problem and eventually computing a solution by a deep learning technique. Current technology innovations are improving computer speed, data storage media, and graphics processing units. These factors are turning machine learning techniques into state-of-the-art computation-and-prediction tools that can be automated to deal with large volumes of vibration data. Prediction-diagnosis of damage results in improved condition monitoring of complex mechanical systems and this in turn infers economic gains due to estimated low-cost maintenance. Classical condition monitoring techniques have serious limitations such as the inability in learning from datasets the dynamics properties onboard installed machinery units operating under varying environmental conditions.	en
heal.advisorName	Γεωργίου, Ιωάννης	el
heal.committeeMemberName	Γεωργίου, Ιωάννης	el
heal.committeeMemberName	Σπύρου, Κωνσταντίνος	el
heal.committeeMemberName	Ανυφαντής, Κωνσταντίνος	el
heal.academicPublisher	Εθνικό Μετσόβιο Πολυτεχνείο. Σχολή Ναυπηγών Μηχανολόγων Μηχανικών	el
heal.academicPublisherID	ntua
heal.numberOfPages	45 σ.	el
heal.fullTextAvailability	false
heal.fullTextAvailability	false