[25a-11F-1~11] 3.11 Nanoscale optical science and near-field optics (formerly 3.12)

9:30 AM - 9:45 AM JST(12:30 AM - 12:45 AM UTC)Presentation by Applicant for JSAP Young Scientists Presentation Award

[25a-11F-3] Prediction Algorithms for Determining Parameters of Broadband Metamaterials

〇Hamada Kenta¹, Kubo Wakana¹ (1.Tokyo Univ. of Agri. Technol.)

Keywords:

optimization,machine learning,deep learning

Metamaterial (MA) thermoelectric conversion is a new mechanism that enables thermoelectric power generation in a uniform thermal radiation environment. In this mechanism, the thermal radiation absorption property of MA determines the thermoelectric performance, and therefore, it is desirable for MA to have high absorption in a broad band. In this study, we optimized a parallel structure of hyperbolic metamaterials (HMMs), which exhibit broadband absorption, and evaluated their absorption characteristics. By using deep reinforcement learning, HMMs exhibiting broadband absorption were successfully designed.