Optical Instruments, Volume. 45, Issue 5, 72(2023)

Infrared radiation computation and inverse design of multilayer thin film structures based on neural network

Yinggang CHEN1,2, Qian ZHU1,2, Wenyi DENG1,2, Xinyu WU1,2, and Yinan ZHANG1、*
Author Affiliations
  • 1Institute of Photonic Chips, University of Shanghai for Science and Technology, Shanghai 200093, China
  • 2School of Optical-Electrical and Computer Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China
  • show less
    Figures & Tables(8)
    Schematic diagram of multilayer thin film structure and its infrared emission spectrum based on neural network
    Forward simulation of multilayer dielectric thin film structures based on neural network
    The training loss function of the 5-layer thin film structure
    Neural network inverse design of multilayer dielectric thin film structures
    Neural network inverse optimization of multilayer dielectric thin film structures
    Net cooling power of two types radiative coolers at night
    Net cooling power of two types radiative coolers in daytime
    • Table 1. Neural numbers of various layer thin film structures and its cross validation results

      View table
      View in Article

      Table 1. Neural numbers of various layer thin film structures and its cross validation results

      层数神经元 个数 训练集 平均损失/% 验证集 平均损失/% 测试集 平均损失/%
      31000.1040.2260.028
      42000.0860.3560.030
      52000.0800.3240.028
      62250.0660.0780.068
      72250.0700.0670.041
    Tools

    Get Citation

    Copy Citation Text

    Yinggang CHEN, Qian ZHU, Wenyi DENG, Xinyu WU, Yinan ZHANG. Infrared radiation computation and inverse design of multilayer thin film structures based on neural network[J]. Optical Instruments, 2023, 45(5): 72

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category:

    Received: Feb. 2, 2023

    Accepted: --

    Published Online: Dec. 27, 2023

    The Author Email: ZHANG Yinan (zhangyinan@usst.edu.cn)

    DOI:10.3969/j.issn.1005-5630.2023.005.009

    Topics