Real Time Predictions of VGF-GaAs Growth Dynamics by LSTM Neural Networks
The 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...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-01-01
|
| Series: | Crystals |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4352/11/2/138 |
| _version_ | 1797406075690942464 |
|---|---|
| author | Natasha Dropka Stefan Ecklebe Martin Holena |
| author_facet | Natasha Dropka Stefan Ecklebe Martin Holena |
| author_sort | Natasha Dropka |
| collection | DOAJ |
| description | The 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<sup>−3</sup>. This deep learning method achieved a superior predictive accuracy and timeliness compared with more traditional Nonlinear AutoRegressive eXogenous (NARX) recurrent networks. |
| first_indexed | 2024-03-09T03:21:02Z |
| format | Article |
| id | doaj.art-000bf34d29da4bdf9fe9da1b8eb643d6 |
| institution | Directory Open Access Journal |
| issn | 2073-4352 |
| language | English |
| last_indexed | 2024-03-09T03:21:02Z |
| publishDate | 2021-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Crystals |
| spelling | doaj.art-000bf34d29da4bdf9fe9da1b8eb643d62023-12-03T15:09:37ZengMDPI AGCrystals2073-43522021-01-0111213810.3390/cryst11020138Real Time Predictions of VGF-GaAs Growth Dynamics by LSTM Neural NetworksNatasha Dropka0Stefan Ecklebe1Martin Holena2Leibniz-Institut für Kristallzüchtung, Max Born St. 2, 12489 Berlin, GermanyInstitute of Control Theory, TU Dresden, Georg Schumann St. 7a, 01187 Dresden, GermanyLeibniz Institute for Catalysis, Albert Einstein St. 29A, 18069 Rostock, GermanyThe 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<sup>−3</sup>. This deep learning method achieved a superior predictive accuracy and timeliness compared with more traditional Nonlinear AutoRegressive eXogenous (NARX) recurrent networks.https://www.mdpi.com/2073-4352/11/2/138neural networkscrystal growthGaAsprocess controldigital twins |
| spellingShingle | Natasha Dropka Stefan Ecklebe Martin Holena Real Time Predictions of VGF-GaAs Growth Dynamics by LSTM Neural Networks Crystals neural networks crystal growth GaAs process control digital twins |
| title | Real Time Predictions of VGF-GaAs Growth Dynamics by LSTM Neural Networks |
| title_full | Real Time Predictions of VGF-GaAs Growth Dynamics by LSTM Neural Networks |
| title_fullStr | Real Time Predictions of VGF-GaAs Growth Dynamics by LSTM Neural Networks |
| title_full_unstemmed | Real Time Predictions of VGF-GaAs Growth Dynamics by LSTM Neural Networks |
| title_short | Real Time Predictions of VGF-GaAs Growth Dynamics by LSTM Neural Networks |
| title_sort | real time predictions of vgf gaas growth dynamics by lstm neural networks |
| topic | neural networks crystal growth GaAs process control digital twins |
| url | https://www.mdpi.com/2073-4352/11/2/138 |
| work_keys_str_mv | AT natashadropka realtimepredictionsofvgfgaasgrowthdynamicsbylstmneuralnetworks AT stefanecklebe realtimepredictionsofvgfgaasgrowthdynamicsbylstmneuralnetworks AT martinholena realtimepredictionsofvgfgaasgrowthdynamicsbylstmneuralnetworks |