基于神经网络的纳米光子结构逆设计

doi:10.13190/j.jbupt.2021-281

北京邮电大学学报 ›› 2022, Vol. 45 ›› Issue (3): 112-116.doi: 10.13190/j.jbupt.2021-281

• 研究报告 • 上一篇

基于神经网络的纳米光子结构逆设计

田亮, 潘鹏, 刘玉敏

北京邮电大学电子信息工程学院, 北京 100876

收稿日期:2021-12-03 出版日期:2022-06-28 发布日期:2022-06-01
作者简介:田亮(1987—),男,博士生,邮箱:tianliang667@sina.com;刘玉敏(1976—),男,教授,博士生导师。

The Inverse Design of Nanophotonic Structure Based on Neural Network

TIAN Liang, PAN Peng, LIU Yumin

School of Electronic Information Engineering, Beijing University of Posts and Telecommunications, Beijing 100876, China

Received:2021-12-03 Online:2022-06-28 Published:2022-06-01

摘要/Abstract

摘要： 同扫描参数优化的方法相比,神经网络模型可以在设计误差很小的情况下大大提高结构逆设计的效率。在模拟器与发生器串联的复合模型基础上,提出了2种有效的神经网络复合模型:基于生成对抗思想引入评价模块的复合模型和基于多任务学习增加模拟器模块的复合模型。在传输矩阵法模拟的数据集上进行有效训练后,将2个模型应用于多层薄膜吸收光谱的逆设计,并用具有特殊形状的目标光谱验证了2个模型的逆设计结果。最后,利用多任务串联网络模型设计了多层太阳能吸收器结构参数,在300~1500nm范围内,平均吸收率可达96%。

关键词: 纳米光子结构, 对抗生成, 多任务学习, 复合模型

Abstract: Compared with the method based on scanning parameters and optimization,neural network model can greatly improve the efficiency of structural inverse design with low design error. Two kinds of efficient neural network composite models are proposed: the composite model that introduces critic module based on generation adversarial idea and the composite model that adds simulator module based on multi-task learning. The two models to the inverse design problem of multilayer thin film absorption spectra after effective training on the data set simulated by the transfer matrix method, and the results of the inverse design of the two models are verified by the target spectra with special shapes. Finally, multi-task tandem network model is used to design a set of solar absorber structural parameters, which achieves an average absorption rate of 96% between 300nm and 1500nm.

Key words: nanophotonic structure, adversarial generation, multi-task learning, composite model

中图分类号:

TP183

田亮, 潘鹏, 刘玉敏. 基于神经网络的纳米光子结构逆设计[J]. 北京邮电大学学报, 2022, 45(3): 112-116.

TIAN Liang, PAN Peng, LIU Yumin. The Inverse Design of Nanophotonic Structure Based on Neural Network[J]. Journal of Beijing University of Posts and Telecommunications, 2022, 45(3): 112-116.

参考文献

[1] PEURIFOY J, SHEN Y C, LI J, et al. Nanophotonic particle simulation and inverse design using artificial neural networks[J/OL]. Science Advances, 2018, 4(6)[2021-09-15]. https://www.science.org/doi/10.1126/sciadv.aar4206.
[2] ASANO T, NODA S. Optimization of photonic crystal nanocavities based on deep learning[J]. Optics Express, 2018, 26(25):2122-2133.
[3] LIU D J, TAN Y X, KHORAM E, et al. Training deep neural networks for the inverse design of nanophotonic structures[J]. Acs Photonics, 2018, 5(4):1365-1369.
[4] GOODFELLOW I J, POUGET-ABADIE J, MIRZA M, et al. Generative adversarial networks[J]. Advances in Neural Information Processing Systems, 2014, 3(10):2672-2680.
[5] MA X, ZHAO L Q, HUANG G, et al. Entire space multi-task model:an effective approach for estimating post-click conversion rate[C]//The 41st International ACM SIGIR Conference. Miami:ACM, 2018:1137-1140.
[6] MANASWI, KUMAR N. Deep learning with applications using Python understanding and working with Keras[M]. Berkeley:Apress, 2018:31-43.
[7] YAO L, FANG Z P, XIAO Y Q, et al. An intelligent fault diagnosis method for lithium battery systems based on grid search support vector machine[J/OL]. Energy, 2021, 214(C):[2021-09-15]. https://doi.org/10.1016/j.energy.2020.118866.
[8] SAFFARI M, KHODAYAR M, EBRAHIMI S, et al. Maximum relevance minimum redundancy dropout with informative kernel determinantal point process[J]. Sensors, 2021, 21(5):1846-1852.
[9] HERMANS M, BURM M, VAERENBERGH T, et al. Trainable hardware for dynamical computing using error backpropagation through physical media[J]. Nature Communacations, 2015, 6(3):6729-6737.
[10] LIU V, FAN S. A free electromagnetic solver for layered periodic structures[J]. Computer Physics Communications, 2012, 183(10):2233-2244.
[11] The National Renewable Energy Laboratory. Standard tables for reference solar spectral irradiances:ASTM G173-03(2020)-2020[S/OL]. West Conshohocken:American Society for Testing and Materials, 2020:21[2021-09-15]. https://www.nrel.gov/grid/solar-resource/spectra-am1.5.html.

基于神经网络的纳米光子结构逆设计

The Inverse Design of Nanophotonic Structure Based on Neural Network

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 3

编辑推荐

Metrics

本文评价

[1]	张天骐汪锐安泽亮方竹王雪怡. 基于多任务学习的MIMO-OFDM信号联合信噪比估计与调制识别[J]. 北京邮电大学学报, 2022, 45(6): 98-104.
[2]	王玉峰, 肖灿彬, 陈焱, 金群. 一种利用多任务学习的短期住宅负荷预测方案[J]. 北京邮电大学学报, 2021, 44(3): 47-52.
[3]	傅群超, 王枞. 用于文本分类的多探测任务语言模型微调[J]. 北京邮电大学学报, 2019, 42(6): 76-83.