journal article·2023-03-31

Normalizing Flows for Efficient Inverse Design of Thermophotovoltaic Emitters

Jia-Qi Yang, YuCheng Xu, Kebin Fan, Jingbo Wu, Caihong Zhang, De-Chuan Zhan, Biao-Bing Jin, Willie J. Padilla

Publisher page

Abstract

Metasurfaces have shown flexibility in realizing various functionalities via shaping the geometry on the subwavelength scale. However, with increased design complexity, the flexibility makes the traditional paradigm based on expert knowledge less effective. Due to the ability to learn knowledge from raw data, deep-learning techniques have made remarkable progress in the automatic design of high-performance nanophotonic devices. However, deep-learning-based methods require a large amount of training data, which is very expensive for metasurface design. Therefore, there is still a need for efficient inverse design methods with lower data complexity and higher performance. In this Article, we modeled the inverse design problem as a probabilistic distribution learning problem and introduce normalizing flow as a flexible model to explicitly learn the distribution. To efficiently explore prior distributions without generating extra training data, we proposed a prior reshaping method that reweighs the original training data according to their fitness to the design target. The algorithm is implemented for designing niobium-based emitters in a GaSb thermophotovoltaic (TPV) cell, achieving an in-band emittance efficiency close to 99%. To further improve the data efficiency of inverse design, we explored the transfer learning ability of the proposed normalizing flow model. A new tungsten-based emitter for a GaInAsSb TPV cell was obtained with comparable performance using only 5% of data compared to the original design problem. The demonstrated approach is not only applicable to TPV device designs, but may also be used in various data-driven inverse design problems, such as artificially structured materials, synthetic materials, and mechanical devices.
Source: openalex_inverted_index · Quality: baseline

Research context

OpenAlex topics: Thermal Radiation and Cooling Technologies · Metamaterials and Metasurfaces Applications · Acoustic Wave Phenomena Research

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

Affiliations

State Key Laboratory for Novel Software Technology, Nanjing University, Nanjing 210023, China

Research Institute of Superconductor Electronics (RISE) School of Electronic Science and Engineering, Nanjing University, Nanjing 210023, China

Purple Mountain Laboratories, Nanjing 211111, China

Duke University, Department of Electrical and Computer Engineering, Durham, North Carolina 27708, United States

Funding

Central University Basic Research Fund of China · National Natural Science Foundation of China · Jiangsu Provincial Key Laboratory of Advanced Manipulating Technique of Electromagnetic Wave

References (60)

Show source-provided references
1. 10.1088/0034-4885/79/7/07640110.1088/0034-4885/79/7/076401 ↗
2. 10.1109/TAP.2020.296879510.1109/TAP.2020.2968795 ↗
3. 10.1515/nanoph-2017-011810.1515/nanoph-2017-0118 ↗
4. 10.1515/nanoph-2017-012910.1515/nanoph-2017-0129 ↗
5. 10.1002/smtd.20160006410.1002/smtd.201600064 ↗
6. 10.1038/s42254-021-00394-310.1038/s42254-021-00394-3 ↗
7. 10.1038/s41377-022-00879-510.1038/s41377-022-00879-5 ↗
8. Padilla W. J. · 202210.1007/978-3-031-03765-8 ↗
9. 10.1038/s41586-022-04473-y10.1038/s41586-022-04473-y ↗
10. 10.1515/nanoph-2016-001110.1515/nanoph-2016-0011 ↗
11. 10.1088/0268-1242/18/5/30310.1088/0268-1242/18/5/303 ↗
12. 10.1103/PhysRevLett.107.04590110.1103/PhysRevLett.107.045901 ↗
13. 10.1002/aenm.20140033410.1002/aenm.201400334 ↗
14. 10.1016/j.solmat.2013.12.01210.1016/j.solmat.2013.12.012 ↗
15. 10.1021/acs.nanolett.8b0332210.1021/acs.nanolett.8b03322 ↗
16. 10.1016/j.ijheatmasstransfer.2021.12203910.1016/j.ijheatmasstransfer.2021.122039 ↗
17. 10.1038/nphoton.2008.14610.1038/nphoton.2008.146 ↗
18. 10.1038/ncomms622110.1038/ncomms6221 ↗
19. 10.1103/PhysRevA.71.01381710.1103/PhysRevA.71.013817 ↗
20. 10.1364/OE.27.02752310.1364/OE.27.027523 ↗
21. 10.1038/s41566-018-0246-910.1038/s41566-018-0246-9 ↗
22. 10.1007/s00340005034610.1007/s003400050346 ↗
23. 10.1364/OE.21.02169310.1364/OE.21.021693 ↗
24. 10.1021/acs.nanolett.7b0108210.1021/acs.nanolett.7b01082 ↗
25. 10.1021/acsphotonics.6b0006610.1021/acsphotonics.6b00066 ↗
26. 10.1021/acsphotonics.8b0152210.1021/acsphotonics.8b01522 ↗
27. 10.1021/acsphotonics.0c0105810.1021/acsphotonics.0c01058 ↗
28. 10.1364/OME.9.00184210.1364/OME.9.001842 ↗
29. 10.1002/adfm.20210174810.1002/adfm.202101748 ↗
30. Goodfellow, I. J.; Pouget-Abadie, J.; Mirza, M.; Xu, B.; Warde-Farley, D.; Ozair, S.; Courville, A. C.; Bengio, Y. Generative Adversarial Nets. Advances in Neural Information Processing Systems 27: Annual Conference on Neural Information Processing Systems 2014, December 8–13, 2014, Montreal, Quebec, Canada, 2014; pp 2672–2680.
31. 10.1021/acs.nanolett.8b0317110.1021/acs.nanolett.8b03171 ↗
32. 10.48550/arXiv.1312.611410.48550/arXiv.1312.6114 ↗
33. 10.1002/adma.20190111110.1002/adma.201901111 ↗
34. 10.48550/arXiv.1511.0564410.48550/arXiv.1511.05644 ↗
35. 10.1063/1.513479210.1063/1.5134792 ↗
36. 10.1364/OE.41913810.1364/OE.419138 ↗
37. 10.1021/acsphotonics.1c0188810.1021/acsphotonics.1c01888 ↗
38. Kingma, D. P.; Dhariwal, P. Glow: Generative Flow with Invertible 1 × 1 Convolutions. Advances in Neural Information Processing Systems 31: Annual Conference on Neural Information Processing Systems 2018, NeurIPS 2018; December 3–8, 2018, Montréal, Canada, 2018; pp 10236–10245.
39. 10.1109/TPAMI.2020.299293410.1109/TPAMI.2020.2992934 ↗
40. Papamakarios G. · J. Mach. Learn. Res. · 2021 · 22 · 57:1
41. 10.1039/D1NR08346E10.1039/D1NR08346E ↗
42. 10.48550/arXiv.2101.1076310.48550/arXiv.2101.10763 ↗
43. Wang C. A. · AIP Conf. Proc. · 2004 · 738 · 25510.1063/1.1841902 ↗
44. 10.1364/OE.21.0A103510.1364/OE.21.0A1035 ↗
45. Salomon D. · 2006
46. 10.1364/OL.29.00228810.1364/OL.29.002288 ↗
47. 10.1609/aaai.v36i4.2039010.1609/aaai.v36i4.20390 ↗
48. 10.48550/arXiv.1605.0880310.48550/arXiv.1605.08803 ↗
49. 10.48550/arXiv.1912.0004210.48550/arXiv.1912.00042 ↗
50. 10.1007/978-3-030-71278-5_2710.1007/978-3-030-71278-5_27 ↗
51. 10.48550/arXiv.1609.0474710.48550/arXiv.1609.04747 ↗
52. 10.1002/adom.20220074810.1002/adom.202200748 ↗
53. 10.1103/PhysRevApplied.18.02402210.1103/PhysRevApplied.18.024022 ↗
54. 10.1109/JPROC.2020.300455510.1109/JPROC.2020.3004555 ↗
55. 10.48550/arXiv.1805.0167710.48550/arXiv.1805.01677 ↗
56. 10.48550/arXiv.1911.1327010.48550/arXiv.1911.13270 ↗
57. 10.48550/arXiv.2108.1136810.48550/arXiv.2108.11368 ↗
58. 10.1038/s41467-021-23087-y10.1038/s41467-021-23087-y ↗
59. van der Maaten L. · Journal of Machine Learning Research · 2008 · 9 · 2579
60. Thanh-Tung, H.; Tran, T. Catastrophic Forgetting and Mode Collapse in GANs. 2020 International Joint Conference on Neural Networks, IJCNN 2020, Glasgow, United Kingdom, July 19–24, 2020, IEEE, 2020; pp 1–10.10.1109/IJCNN48605.2020.9207181 ↗