Vis enkel innførsel

dc.contributor.authorDjenouri, Youcef
dc.contributor.authorBelhadi, Asma
dc.contributor.authorSrivastava, Gautam
dc.contributor.authorGhosh, Uttam
dc.contributor.authorChatterjee, Pushpita
dc.contributor.authorLin, Jerry Chun-Wei
dc.date.accessioned2022-07-15T07:54:49Z
dc.date.available2022-07-15T07:54:49Z
dc.date.created2021-12-24T22:28:17Z
dc.date.issued2021
dc.identifier.citationIEEE Internet of Things Journal.en_US
dc.identifier.issn2327-4662
dc.identifier.urihttps://hdl.handle.net/11250/3005663
dc.description.abstractThis paper introduced a new deep learning framework for fault diagnosis in electrical power systems. The framework integrates the convolution neural network and different regression models to visually identify which faults have occurred in electric power systems. The approach includes three main steps, data preparation, object detection, and hyper-parameter optimization. Inspired by deep learning, evolutionary computation techniques, different strategies have been proposed in each step of the process. In addition, we propose a new hyper-parameters optimization model based on evolutionary computation that can be used to tune parameters of our deep learning framework. In the validation of the framework’s usefulness, experimental evaluation is executed using the well known and challenging VOC 2012, the COCO datasets, and the large NESTA 162-bus system. The results show that our proposed approach significantly outperforms most of the existing solutions in terms of runtime and accuracy.en_US
dc.language.isoengen_US
dc.titleFast and Accurate Deep Learning Framework for Secure Fault Diagnosis in the Industrial Internet of Thingsen_US
dc.typePeer revieweden_US
dc.typeJournal articleen_US
dc.description.versionsubmittedVersionen_US
dc.source.pagenumber10.en_US
dc.source.journalIEEE Internet of Things Journalen_US
dc.identifier.doi10.1109/JIOT.2021.3092275
dc.identifier.cristin1971953
dc.relation.projectNatural Sciences and Engineering Research Council of Canada: RGPIN-2020-05363en_US
cristin.ispublishedtrue
cristin.fulltextpreprint
cristin.qualitycode2


Tilhørende fil(er)

Thumbnail

Denne innførselen finnes i følgende samling(er)

Vis enkel innførsel