一种基于神经网络提取旋波材料参数的新方法

提出了一种从测试的S参数中提取旋波材料参数的新方法——神经网络法。针对旋波材料参数的自由空间测试方法，本文采用BP神经网络来构建映射模型，利用不同算法对网络进行训练，从而获得所需的网络。结果表明，该方法计算速度快、精度高，解决了求解非线性方程组解的模糊性问题，确保了测试结果的可靠性。该方法所构建的网络可用于材料测试后旋波材料参数的快速提取。

A New Method to Extract Parameter of Chiral Materials Based on Neural Network

A new method based on neural network is presented to extract constitutive parameters of chiral materials from S parameters of measurement. The mapping model is created with the BP neural network for the measurement of chiral materials. And the expected network is obtained after training the network by different algorithms. It is shown that the method is quick and precise, and the problem of the ambiguity for nonlinear equations and the measurement reliability are solved. The network could be used to extract materials parameters from the measurement results quickly.