Real Time Predictions of VGF-GaAs Growth Dynamics by LSTM Neural Networks

dc.bibliographicCitation.firstPage138eng
dc.bibliographicCitation.issue2eng
dc.bibliographicCitation.journalTitleCrystals : open access journaleng
dc.bibliographicCitation.volume11eng
dc.contributor.authorDropka, Natasha
dc.contributor.authorEcklebe, Stefan
dc.contributor.authorHolena, Martin
dc.date.accessioned2022-01-18T09:15:24Z
dc.date.available2022-01-18T09:15:24Z
dc.date.issued2021
dc.description.abstractThe aim of this study was to assess the aptitude of the recurrent Long Short-Term Memory (LSTM) neural networks for fast and accurate predictions of process dynamics in vertical-gradient-freeze growth of gallium arsenide crystals (VGF-GaAs) using datasets generated by numerical transient simulations. Real time predictions of the temperatures and solid–liquid interface position in GaAs are crucial for control applications and for process visualization, i.e., for generation of digital twins. In the reported study, an LSTM network was trained on 1950 datasets with 2 external inputs and 6 outputs. Based on network performance criteria and training results, LSTMs showed the very accurate predictions of the VGF-GaAs growth process with median root-mean-square-error (RMSE) values of 2 × 10−3. This deep learning method achieved a superior predictive accuracy and timeliness compared with more traditional Nonlinear AutoRegressive eXogenous (NARX) recurrent networks.eng
dc.description.versionpublishedVersioneng
dc.identifier.urihttps://oa.tib.eu/renate/handle/123456789/7839
dc.identifier.urihttps://doi.org/10.34657/6880
dc.language.isoengeng
dc.publisherBasel : MDPIeng
dc.relation.doihttps://doi.org/10.3390/cryst11020138
dc.relation.essn2073-4352
dc.rights.licenseCC BY 4.0 Unportedeng
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/eng
dc.subject.ddc540eng
dc.subject.otherCrystal growtheng
dc.subject.otherDigital twinseng
dc.subject.otherGaAseng
dc.subject.otherNeural networkseng
dc.subject.otherProcess controleng
dc.titleReal Time Predictions of VGF-GaAs Growth Dynamics by LSTM Neural Networkseng
dc.typeArticleeng
dc.typeTexteng
tib.accessRightsopenAccesseng
wgl.contributorIKZeng
wgl.contributorLIKATeng
wgl.subjectChemieeng
wgl.typeZeitschriftenartikeleng
Files
Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
Real time predictions of VGF-GaAs growth dynamics by LSTM neural networks.pdf
Size:
1.16 MB
Format:
Adobe Portable Document Format
Description:
Collections