Deep learning for the modeling and inverse design of radiative heat transfer
arXiv:2109.03114 · doi:10.1103/PhysRevApplied.16.064006
Abstract
Deep learning is having a tremendous impact in many areas of computer science and engineering. Motivated by this success, deep neural networks are attracting an increasing attention in many other disciplines, including physical sciences. In this work, we show that artificial neural networks can be successfully used in the theoretical modeling and analysis of a variety of radiative heat transfer phenomena and devices. By using a set of custom-designed numerical methods able to efficiently generate the required training datasets, we demonstrate this approach in the context of three very different problems, namely, (i) near-field radiative heat transfer between multilayer systems that form hyperbolic metamaterials, (ii) passive radiate cooling in photonic-crystal slab structures, and (iii) thermal emission of subwavelength objects. Despite their fundamental differences in nature, in all three cases we show that simple neural network architectures trained with datasets of moderate size can be used as fast and accurate surrogates for doing numerical simulations, as well as engines for solving inverse design and optimization in the context of radiative heat transfer. Overall, our work shows that deep learning and artificial neural networks provide a valuable and versatile toolkit for advancing the field of thermal radiation.
arXiv comment16 pages, 11 figures
References in corpus (29)
- Machine learning and the physical sciences
- Training deep neural networks for the inverse design of nanophotonic structures
- A high-bias, low-variance introduction to Machine Learning for physicists
- The discrete dipole approximation: an overview and recent developments
- Deep neural networks for the evaluation and design of photonic devices
- Global optimization of dielectric metasurfaces using a physics-driven neural network
- Towards a black body for near-field thermal radiation
- Broadband super-Planckian thermal emission from hyperbolic metamaterials
- Near-field Radiative Heat Transfer in Many-Body Systems
- Machine-Learning-Assisted Metasurface Design for High-Efficiency Thermal Emitter Optimization
- Trace formulae for non-equilibrium Casimir interactions, heat radiation and heat transfer for arbitrary objects
- Non-equilibrium electromagnetic fluctuations: Heat transfer and interactions
- Magnetic-field control of near-field radiative heat transfer and the realization of highly tunable hyperbolic thermal emitters
- Super-Planckian Near-Field Thermal Emission with Phonon-Polaritonic Hyperbolic Metamaterials
- Simulator-based training of generative models for the inverse design of metasurfaces
- Enhancing Near-Field Radiative Heat Transfer with Si-based Metasurfaces
- Fluctuating surface-current formulation of radiative heat transfer: theory and applications
- Efficient Computation of Power, Force, and Torque in BEM Scattering Calculations
- The effectiveness of thin films in lieu of hyperbolic metamaterials in the near field
- Fluctuating volume-current formulation of electromagnetic fluctuations in inhomogeneous media: incandecence and luminescence in arbitrary geometries
- Heat radiation from long cylindrical objects
- Thermal discrete dipole approximation for the description of thermal emission and radiative heat transfer of magneto-optical systems
- Generalized scattering-matrix approach for magneto-optics in periodically patterned multilayer systems
- Super-Planckian Far-Field Radiative Heat Transfer
- Near-field heat transfer between multilayer hyperbolic metamaterials
- Thermalization via Heat Radiation of an Individual Object Thinner than the Thermal Wavelength
- Descending through a Crowded Valley - Benchmarking Deep Learning Optimizers
- Near-field radiative heat transfer between one-dimensional magneto-photonic crystals
- Tailoring the local density of states of non-radiative field at the surface of nanolayered materials