Matilde Fiore, Enrico Saccaggi, Lilla Koloszar, Yann Bartosiewicz, Miguel Alfonso Mendez
{"title":"数据驱动的湍流热通量建模,可输入多种保真度数据","authors":"Matilde Fiore, Enrico Saccaggi, Lilla Koloszar, Yann Bartosiewicz, Miguel Alfonso Mendez","doi":"arxiv-2409.03395","DOIUrl":null,"url":null,"abstract":"Data-driven RANS modeling is emerging as a promising methodology to exploit\nthe information provided by high-fidelity data. However, its widespread\napplication is limited by challenges in generalization and robustness to\ninconsistencies between input data of varying fidelity levels. This is\nespecially true for thermal turbulent closures, which inherently depend on\nmomentum statistics provided by low or high fidelity turbulence momentum\nmodels. This work investigates the impact of momentum modeling inconsistencies\non a data-driven thermal closure trained with a dataset with multiple fidelity\n(DNS and RANS). The analysis of the model inputs shows that the two fidelity\nlevels correspond to separate regions in the input space. It is here shown that\nsuch separation can be exploited by a training with heterogeneous data,\nallowing the model to detect the level of fidelity in its inputs and adjust its\nprediction accordingly. In particular, a sensitivity analysis and verification\nshows that such a model can leverage the data inconsistencies to increase its\nrobustness. Finally, the verification with a CFD simulation shows the potential\nof this multi-fidelity training approach for flows in which momentum statistics\nprovided by traditional models are affected by model uncertainties.","PeriodicalId":501125,"journal":{"name":"arXiv - PHYS - Fluid Dynamics","volume":"33 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Data-driven turbulent heat flux modeling with inputs of multiple fidelity\",\"authors\":\"Matilde Fiore, Enrico Saccaggi, Lilla Koloszar, Yann Bartosiewicz, Miguel Alfonso Mendez\",\"doi\":\"arxiv-2409.03395\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Data-driven RANS modeling is emerging as a promising methodology to exploit\\nthe information provided by high-fidelity data. However, its widespread\\napplication is limited by challenges in generalization and robustness to\\ninconsistencies between input data of varying fidelity levels. This is\\nespecially true for thermal turbulent closures, which inherently depend on\\nmomentum statistics provided by low or high fidelity turbulence momentum\\nmodels. This work investigates the impact of momentum modeling inconsistencies\\non a data-driven thermal closure trained with a dataset with multiple fidelity\\n(DNS and RANS). The analysis of the model inputs shows that the two fidelity\\nlevels correspond to separate regions in the input space. It is here shown that\\nsuch separation can be exploited by a training with heterogeneous data,\\nallowing the model to detect the level of fidelity in its inputs and adjust its\\nprediction accordingly. In particular, a sensitivity analysis and verification\\nshows that such a model can leverage the data inconsistencies to increase its\\nrobustness. Finally, the verification with a CFD simulation shows the potential\\nof this multi-fidelity training approach for flows in which momentum statistics\\nprovided by traditional models are affected by model uncertainties.\",\"PeriodicalId\":501125,\"journal\":{\"name\":\"arXiv - PHYS - Fluid Dynamics\",\"volume\":\"33 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-09-05\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - PHYS - Fluid Dynamics\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2409.03395\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - PHYS - Fluid Dynamics","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.03395","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Data-driven turbulent heat flux modeling with inputs of multiple fidelity
Data-driven RANS modeling is emerging as a promising methodology to exploit
the information provided by high-fidelity data. However, its widespread
application is limited by challenges in generalization and robustness to
inconsistencies between input data of varying fidelity levels. This is
especially true for thermal turbulent closures, which inherently depend on
momentum statistics provided by low or high fidelity turbulence momentum
models. This work investigates the impact of momentum modeling inconsistencies
on a data-driven thermal closure trained with a dataset with multiple fidelity
(DNS and RANS). The analysis of the model inputs shows that the two fidelity
levels correspond to separate regions in the input space. It is here shown that
such separation can be exploited by a training with heterogeneous data,
allowing the model to detect the level of fidelity in its inputs and adjust its
prediction accordingly. In particular, a sensitivity analysis and verification
shows that such a model can leverage the data inconsistencies to increase its
robustness. Finally, the verification with a CFD simulation shows the potential
of this multi-fidelity training approach for flows in which momentum statistics
provided by traditional models are affected by model uncertainties.