@inproceedings{3f57ab3723394052bc7b6b0b3f4eb938,
title = "Equivalent Electromagnetic Parameter Inversion of Honeycomb Structures Based on BP Neural Network",
abstract = "We present in this paper an effective BP neural network-based method for the equivalent electromagnetic parameter inversion of honeycomb structures. The BP neural network is trained using radar cross section (RCS) of different honeycomb structures and the equivalent dielectric constants of the honeycomb as the input and output variables, respectively. To simplify the sample data generation and reduce the dimension of output variables, the Hashin-Shtrikman (HS) variational theory is used to homogenize the honeycomb structure as homogenous materials. The hybrid finite element-boundary integral-multilevel fast multipole method (FE-BI-MLFMA) is used as the solver for scattering problems after homogenization. Numerical results show that the trained high-quality network model can achieve good accuracy, which provides an effective way for predicting equivalent electromagnetic parameter of microwave absorbing honeycomb structures.",
keywords = "BP neural network, effective electromagnetic parameters, electromagnetic scattering, microwave absorbing honeycomb structure",
author = "Zhang, {Yu Xin} and Yuan, {Xiao Wei} and Yang, {Ming Lin} and Sheng, {Xin Qing}",
note = "Publisher Copyright: {\textcopyright} 2022 IEEE.; 2022 International Applied Computational Electromagnetics Society Symposium, ACES-China 2022 ; Conference date: 09-12-2022 Through 12-12-2022",
year = "2022",
doi = "10.1109/ACES-China56081.2022.10064831",
language = "English",
series = "2022 International Applied Computational Electromagnetics Society Symposium, ACES-China 2022",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
booktitle = "2022 International Applied Computational Electromagnetics Society Symposium, ACES-China 2022",
address = "United States",
}