Jagiellonian University Repository

Predictive maintenance of induction motors using ultra-low power wireless sensors and compressed recurrent neural networks

pcg.skipToMenu

Predictive maintenance of induction motors using ultra-low power wireless sensors and compressed recurrent neural networks

Show full item record

dc.contributor.author Markiewicz, Michał [SAP13009640] pl
dc.contributor.author Wielgosz, Maciej pl
dc.contributor.author Bocheński, Mikołlaj pl
dc.contributor.author Tabaczyński, Waldemar pl
dc.contributor.author Konieczny, Tomasz pl
dc.date.accessioned 2020-02-12T12:59:48Z
dc.date.available 2020-02-12T12:59:48Z
dc.date.issued 2019 pl
dc.identifier.uri https://ruj.uj.edu.pl/xmlui/handle/item/148809
dc.language eng pl
dc.rights Udzielam licencji. Uznanie autorstwa 4.0 Międzynarodowa *
dc.rights.uri http://creativecommons.org/licenses/by/4.0/pl/legalcode *
dc.title Predictive maintenance of induction motors using ultra-low power wireless sensors and compressed recurrent neural networks pl
dc.type JournalArticle pl
dc.description.physical 178891-178902 pl
dc.abstract.en In real-world applications - to minimize the impact of failures - machinery is often monitored by various sensors. Their role comes down to acquiring data and sending it to a more powerful entity, such as an embedded computer or cloud server. There have been attempts to reduce the computational effort related to data processing in order to use edge computing for predictive maintenance. The aim of this paper is to push the boundaries even further by proposing a novel architecture, in which processing is moved to the sensors themselves thanks to decrease of computational complexity given by the usage of compressed recurrent neural networks. A sensor processes data locally, and then wirelessly sends only a single packet with the probability that the machine is working incorrectly. We show that local processing of the data on ultra-low power wireless sensors gives comparable outcomes in terms of accuracy but much better results in terms of energy consumption that transferring of the raw data. The proposed ultra-low power hardware and firmware architecture makes it possible to use sensors powered by harvested energy while maintaining high confidentiality levels of the failure prediction previously offered by more powerful mains-powered computational platforms. pl
dc.subject.en computational complexity pl
dc.subject.en electric machine analysis computing pl
dc.subject.en energy consumption pl
dc.subject.en failure analysis pl
dc.subject.en firmware pl
dc.subject.en induction motors pl
dc.subject.en low-power electronics pl
dc.subject.en maintenance engineering pl
dc.description.volume 7 pl
dc.identifier.doi 10.1109/ACCESS.2019.2953019 pl
dc.identifier.eissn 2169-3536
dc.title.journal IEEE Access pl
dc.language.container eng pl
dc.affiliation Wydział Matematyki i Informatyki : Instytut Informatyki i Matematyki Komputerowej pl
dc.subtype Article pl
dc.rights.original CC-BY; otwarte czasopismo; ostateczna wersja wydawcy; w momencie opublikowania; 0 pl
dc.identifier.project ROD UJ / OP pl
.pointsMNiSW [2019 A]: 100


Files in this item

This item appears in the following Collection(s)

Udzielam licencji. Uznanie autorstwa 4.0 Międzynarodowa Except where otherwise noted, this item's license is described as Udzielam licencji. Uznanie autorstwa 4.0 Międzynarodowa