Abstract
Designing thermal radiation metamaterials is challenging especially for problems with high degrees of freedom and complex objectives. In this Letter, we develop a hybrid materials informatics approach which combines the adversarial autoencoder and Bayesian optimization to design narrowband thermal emitters at different target wavelengths. With only several hundreds of training data sets, new structures with optimal properties can be quickly determined in a compressed two-dimensional latent space. This enables the optimal design by calculating far less than 0.001% of the total candidate structures, which greatly decreases the design period and cost. The proposed design framework can be easily extended to other thermal radiation metamaterials design with higher dimensional features.
© 2022 Optica Publishing Group
Full Article | PDF ArticleMore Like This
Jiang Guo, Shenghong Ju, and Junichiro Shiomi
Opt. Lett. 45(2) 343-346 (2020)
Wang Xi, Yida Liu, Jinlin Song, Run Hu, and Xiaobing Luo
Opt. Lett. 46(4) 888-891 (2021)
Zahra Rahimian Omam, Amir Ghobadi, Bahram Khalichi, and Ekmel Ozbay
Opt. Lett. 47(22) 5781-5784 (2022)