A Hybrid Algorithm for Electromagnetic Optimization Utilizing Neural Networks

2018 IEEE 27th Conference on Electrical Performance of Electronic Packaging and Systems (EPEPS) Pub Date : 2018-10-01 DOI:10.1109/EPEPS.2018.8534264

Yanan Liu, Tianjian Lu, Ken Wu, Jianming Jin

引用次数: 3

Abstract

We propose a hybrid algorithm for electromagnetic optimization. The algorithm combines the genetic algorithm and gradient based optimization, leveraging the power of neural networks in both. The effectiveness and efficiency of the proposed algorithm are demonstrated using antenna and microwave filter design examples.

查看原文本刊更多论文

一种基于神经网络的电磁优化混合算法

提出了一种用于电磁优化的混合算法。该算法结合了遗传算法和基于梯度的优化，利用了神经网络在这两方面的强大功能。通过天线和微波滤波器的设计实例验证了该算法的有效性和高效性。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

2018 IEEE 27th Conference on Electrical Performance of Electronic Packaging and Systems (EPEPS)

自引率

0.00%

发文量