Reconstructing Rayleigh–Bénard flows out of temperature-only measurements using Physics-Informed Neural Networks

article Reconstructing Rayleigh–Bénard flows out of temperature-only measurements using Physics-Informed Neural Networks published yes Patricio Clark Di Leoni author Lokahith N Agasthya author cd100965-0804-11ed-9c55-f4878ff4e877 Michele Buzzicotti author Luca Biferale author CaMu department We investigate the capabilities of Physics-Informed Neural Networks (PINNs) to reconstruct turbulent Rayleigh–Bénard flows using only temperature information. We perform a quantitative analysis of the quality of the reconstructions at various amounts of low-passed-filtered information and turbulent intensities. We compare our results with those obtained via nudging, a classical equation-informed data assimilation technique. At low Rayleigh numbers, PINNs are able to reconstruct with high precision, comparable to the one achieved with nudging. At high Rayleigh numbers, PINNs outperform nudging and are able to achieve satisfactory reconstruction of the velocity fields only when data for temperature is provided with high spatial and temporal density. When data becomes sparse, the PINNs performance worsens, not only in a point-to-point error sense but also, and contrary to nudging, in a statistical sense, as can be seen in the probability density functions and energy spectra. Springer Nature2023 eng The European Physical Journal E 1292-8941 1292-895X 2301.07769 36939938 00095638720000110.1140/epje/s10189-023-00276-9 463 Clark Di Leoni P, Agasthya LN, Buzzicotti M, Biferale L. 2023. Reconstructing Rayleigh–Bénard flows out of temperature-only measurements using Physics-Informed Neural Networks. The European Physical Journal E. 46(3), 16. P. Clark Di Leoni, L. N. Agasthya, M. Buzzicotti, and L. Biferale, “Reconstructing Rayleigh–Bénard flows out of temperature-only measurements using Physics-Informed Neural Networks,” The European Physical Journal E, vol. 46, no. 3. Springer Nature, 2023. Clark Di Leoni, P., Agasthya, L. N., Buzzicotti, M., & Biferale, L. (2023). Reconstructing Rayleigh–Bénard flows out of temperature-only measurements using Physics-Informed Neural Networks. The European Physical Journal E. Springer Nature. <a href="https://doi.org/10.1140/epje/s10189-023-00276-9">https://doi.org/10.1140/epje/s10189-023-00276-9</a> P. Clark Di Leoni, L.N. Agasthya, M. Buzzicotti, L. Biferale, The European Physical Journal E 46 (2023). Clark Di Leoni, Patricio, Lokahith N Agasthya, Michele Buzzicotti, and Luca Biferale. “Reconstructing Rayleigh–Bénard Flows out of Temperature-Only Measurements Using Physics-Informed Neural Networks.” The European Physical Journal E. Springer Nature, 2023. <a href="https://doi.org/10.1140/epje/s10189-023-00276-9">https://doi.org/10.1140/epje/s10189-023-00276-9</a>. Clark Di Leoni P, Agasthya LN, Buzzicotti M, Biferale L. Reconstructing Rayleigh–Bénard flows out of temperature-only measurements using Physics-Informed Neural Networks. The European Physical Journal E. 2023;46(3). doi:<a href="https://doi.org/10.1140/epje/s10189-023-00276-9">10.1140/epje/s10189-023-00276-9</a> Clark Di Leoni, Patricio, et al. “Reconstructing Rayleigh–Bénard Flows out of Temperature-Only Measurements Using Physics-Informed Neural Networks.” The European Physical Journal E, vol. 46, no. 3, 16, Springer Nature, 2023, doi:<a href="https://doi.org/10.1140/epje/s10189-023-00276-9">10.1140/epje/s10189-023-00276-9</a>. 127912023-04-02T22:01:11Z2025-04-23T08:52:35Z