A Study on Optimal Design of Optical Devices Utilizing Coupled Mode Theory and Machine Learning
A Study on Optimal Design of Optical Devices Utilizing Coupled Mode Theory and Machine Learning
复制标题
DOI：
10.1587/transele.2019esp0002
复制
发表时间：
2020-11
期刊：
影响因子：
--
通讯作者：
Koji Kudo;K. Morimoto;A. Iguchi;Yasuhide Tsuji
中科院分区：
文献类型：
--
作者：
Koji Kudo;K. Morimoto;A. Iguchi;Yasuhide Tsuji
摘要本文提出一种利用耦合模理论（CMT）和神经网络（NN）提高光波导器件优化设计计算效率的新设计方法。近年来，神经网络已开始用于光学器件的高效优化设计。在本文中，使用神经网络跳过了CMT中需要的特征模态分析，并且可以有效地使用进化算法进行优化。为了验证该方法的有效性，给出了波长不敏感3dB耦合器、1:2功率分配器和波长解复用器的优化设计实例，并通过与有限元波束传播方法（FE-BPM）的计算结果进行比较，验证了CMT与NN （NN-CMT）的传输特性。
SUMMARY We propose a new design approach to improve the computational e ffi ciency of an optimal design of optical waveguide devices utilizing coupled mode theory (CMT) and a neural network (NN). Recently, the NN has begun to be used for e ffi cient optimal design of optical devices. In this paper, the eigenmode analysis required in the CMT is skipped by using the NN, and optimization with an evolutionary algorithm can be e ffi ciently carried out. To verify usefulness of our approach, optimal design examples of a wavelength insensitive 3dB coupler, a 1 : 2 power splitter, and a wave-length demultiplexer are shown and their transmission properties obtained by the CMT with the NN (NN-CMT) are verified by comparing with those calculated by a finite element beam propagation method (FE-BPM).