Naichen Shi, Hao Yan, Shenghan Guo, Raed Al Kontar
{"title":"多物理场仿真指导下的生成扩散模型在流体和热动力学中的应用","authors":"Naichen Shi, Hao Yan, Shenghan Guo, Raed Al Kontar","doi":"arxiv-2407.17720","DOIUrl":null,"url":null,"abstract":"In this paper, we present a generic physics-informed generative model called\nMPDM that integrates multi-fidelity physics simulations with diffusion models.\nMPDM categorizes multi-fidelity physics simulations into inexpensive and\nexpensive simulations, depending on computational costs. The inexpensive\nsimulations, which can be obtained with low latency, directly inject contextual\ninformation into DDMs. Furthermore, when results from expensive simulations are\navailable, MPDM refines the quality of generated samples via a guided diffusion\nprocess. This design separates the training of a denoising diffusion model from\nphysics-informed conditional probability models, thus lending flexibility to\npractitioners. MPDM builds on Bayesian probabilistic models and is equipped\nwith a theoretical guarantee that provides upper bounds on the Wasserstein\ndistance between the sample and underlying true distribution. The probabilistic\nnature of MPDM also provides a convenient approach for uncertainty\nquantification in prediction. Our models excel in cases where physics\nsimulations are imperfect and sometimes inaccessible. We use a numerical\nsimulation in fluid dynamics and a case study in heat dynamics within\nlaser-based metal powder deposition additive manufacturing to demonstrate how\nMPDM seamlessly integrates multi-idelity physics simulations and observations\nto obtain surrogates with superior predictive performance.","PeriodicalId":501215,"journal":{"name":"arXiv - STAT - Computation","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2024-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Multi-physics Simulation Guided Generative Diffusion Models with Applications in Fluid and Heat Dynamics\",\"authors\":\"Naichen Shi, Hao Yan, Shenghan Guo, Raed Al Kontar\",\"doi\":\"arxiv-2407.17720\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"In this paper, we present a generic physics-informed generative model called\\nMPDM that integrates multi-fidelity physics simulations with diffusion models.\\nMPDM categorizes multi-fidelity physics simulations into inexpensive and\\nexpensive simulations, depending on computational costs. The inexpensive\\nsimulations, which can be obtained with low latency, directly inject contextual\\ninformation into DDMs. Furthermore, when results from expensive simulations are\\navailable, MPDM refines the quality of generated samples via a guided diffusion\\nprocess. This design separates the training of a denoising diffusion model from\\nphysics-informed conditional probability models, thus lending flexibility to\\npractitioners. MPDM builds on Bayesian probabilistic models and is equipped\\nwith a theoretical guarantee that provides upper bounds on the Wasserstein\\ndistance between the sample and underlying true distribution. The probabilistic\\nnature of MPDM also provides a convenient approach for uncertainty\\nquantification in prediction. Our models excel in cases where physics\\nsimulations are imperfect and sometimes inaccessible. We use a numerical\\nsimulation in fluid dynamics and a case study in heat dynamics within\\nlaser-based metal powder deposition additive manufacturing to demonstrate how\\nMPDM seamlessly integrates multi-idelity physics simulations and observations\\nto obtain surrogates with superior predictive performance.\",\"PeriodicalId\":501215,\"journal\":{\"name\":\"arXiv - STAT - Computation\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-07-25\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - STAT - Computation\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2407.17720\",\"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 - STAT - Computation","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2407.17720","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Multi-physics Simulation Guided Generative Diffusion Models with Applications in Fluid and Heat Dynamics
In this paper, we present a generic physics-informed generative model called
MPDM that integrates multi-fidelity physics simulations with diffusion models.
MPDM categorizes multi-fidelity physics simulations into inexpensive and
expensive simulations, depending on computational costs. The inexpensive
simulations, which can be obtained with low latency, directly inject contextual
information into DDMs. Furthermore, when results from expensive simulations are
available, MPDM refines the quality of generated samples via a guided diffusion
process. This design separates the training of a denoising diffusion model from
physics-informed conditional probability models, thus lending flexibility to
practitioners. MPDM builds on Bayesian probabilistic models and is equipped
with a theoretical guarantee that provides upper bounds on the Wasserstein
distance between the sample and underlying true distribution. The probabilistic
nature of MPDM also provides a convenient approach for uncertainty
quantification in prediction. Our models excel in cases where physics
simulations are imperfect and sometimes inaccessible. We use a numerical
simulation in fluid dynamics and a case study in heat dynamics within
laser-based metal powder deposition additive manufacturing to demonstrate how
MPDM seamlessly integrates multi-idelity physics simulations and observations
to obtain surrogates with superior predictive performance.