Deep neural operator-driven real-time inference to enable digital twin solutions for nuclear energy systems
Abstract This paper focuses on the feasibility of deep neural operator network (DeepONet) as a robust surrogate modeling method within the context of digital twin (DT) enabling technology for nuclear energy systems. Machine learning (ML)-based prediction algorithms that need extensive retraining for...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2024-01-01
|
| Series: | Scientific Reports |
| Online Access: | https://doi.org/10.1038/s41598-024-51984-x |
| _version_ | 1797276541764239360 |
|---|---|
| author | Kazuma Kobayashi Syed Bahauddin Alam |
| author_facet | Kazuma Kobayashi Syed Bahauddin Alam |
| author_sort | Kazuma Kobayashi |
| collection | DOAJ |
| description | Abstract This paper focuses on the feasibility of deep neural operator network (DeepONet) as a robust surrogate modeling method within the context of digital twin (DT) enabling technology for nuclear energy systems. Machine learning (ML)-based prediction algorithms that need extensive retraining for new reactor operational conditions may prohibit real-time inference for DT across varying scenarios. In this study, DeepONet is trained with possible operational conditions and that relaxes the requirement of continuous retraining - making it suitable for online and real-time prediction components for DT. Through benchmarking and evaluation, DeepONet exhibits remarkable prediction accuracy and speed, outperforming traditional ML methods, making it a suitable algorithm for real-time DT inference in solving a challenging particle transport problem. DeepONet also exhibits generalizability and computational efficiency as an efficient surrogate tool for DT component. However, the application of DeepONet reveals challenges related to optimal sensor placement and model evaluation, critical aspects of real-world DT implementation. Addressing these challenges will further enhance the method’s practicality and reliability. Overall, this study marks an important step towards harnessing the power of DeepONet surrogate modeling for real-time inference capability within the context of DT enabling technology for nuclear systems. |
| first_indexed | 2024-03-07T15:29:39Z |
| format | Article |
| id | doaj.art-5890b529102e46fe8af4663a3fa7228c |
| institution | Directory Open Access Journal |
| issn | 2045-2322 |
| language | English |
| last_indexed | 2024-03-07T15:29:39Z |
| publishDate | 2024-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj.art-5890b529102e46fe8af4663a3fa7228c2024-03-05T16:29:20ZengNature PortfolioScientific Reports2045-23222024-01-0114111110.1038/s41598-024-51984-xDeep neural operator-driven real-time inference to enable digital twin solutions for nuclear energy systemsKazuma Kobayashi0Syed Bahauddin Alam1Nuclear, Plasma & Radiological Engineering, University of Illinois Urbana-ChampaignNuclear, Plasma & Radiological Engineering, University of Illinois Urbana-ChampaignAbstract This paper focuses on the feasibility of deep neural operator network (DeepONet) as a robust surrogate modeling method within the context of digital twin (DT) enabling technology for nuclear energy systems. Machine learning (ML)-based prediction algorithms that need extensive retraining for new reactor operational conditions may prohibit real-time inference for DT across varying scenarios. In this study, DeepONet is trained with possible operational conditions and that relaxes the requirement of continuous retraining - making it suitable for online and real-time prediction components for DT. Through benchmarking and evaluation, DeepONet exhibits remarkable prediction accuracy and speed, outperforming traditional ML methods, making it a suitable algorithm for real-time DT inference in solving a challenging particle transport problem. DeepONet also exhibits generalizability and computational efficiency as an efficient surrogate tool for DT component. However, the application of DeepONet reveals challenges related to optimal sensor placement and model evaluation, critical aspects of real-world DT implementation. Addressing these challenges will further enhance the method’s practicality and reliability. Overall, this study marks an important step towards harnessing the power of DeepONet surrogate modeling for real-time inference capability within the context of DT enabling technology for nuclear systems.https://doi.org/10.1038/s41598-024-51984-x |
| spellingShingle | Kazuma Kobayashi Syed Bahauddin Alam Deep neural operator-driven real-time inference to enable digital twin solutions for nuclear energy systems Scientific Reports |
| title | Deep neural operator-driven real-time inference to enable digital twin solutions for nuclear energy systems |
| title_full | Deep neural operator-driven real-time inference to enable digital twin solutions for nuclear energy systems |
| title_fullStr | Deep neural operator-driven real-time inference to enable digital twin solutions for nuclear energy systems |
| title_full_unstemmed | Deep neural operator-driven real-time inference to enable digital twin solutions for nuclear energy systems |
| title_short | Deep neural operator-driven real-time inference to enable digital twin solutions for nuclear energy systems |
| title_sort | deep neural operator driven real time inference to enable digital twin solutions for nuclear energy systems |
| url | https://doi.org/10.1038/s41598-024-51984-x |
| work_keys_str_mv | AT kazumakobayashi deepneuraloperatordrivenrealtimeinferencetoenabledigitaltwinsolutionsfornuclearenergysystems AT syedbahauddinalam deepneuraloperatordrivenrealtimeinferencetoenabledigitaltwinsolutionsfornuclearenergysystems |