基于RBF神经网络的异向介质基本单元分析

利用径向基神经网络(RBF神经网络)对异向介质的基本单元进行分析,建立异向介质等效介电常数及等效磁导率与介质敏感结构参数之间的神经网络模型.实验结果表明,该模型所得结果与全波分析相吻合,分析时间为138.465793s,训练平方和误差为0.0374879,满足工程要求.验证了该模型的可靠性及合理性,减少数值分析法由于厚度谐振等问题而引起的结果错误问题,解决了异向介质分析精度与效率难以共存的问题.

Analysis of metamaterial elemental cell based on RBF neural network

To analyze the basic unit of metamaterials by use of the radial basis function neural network (RBFNN),the neural network model is constructed to represent the relations of the equivalent dielectric constant and the equivalent permeability of the metamaterials and the medium's sensitive structural parameters. The experimental results obtained from the model indicate good agreement with the full wave analysis; the analysis time is 138.465793 seconds and the error of the training sum of squares is 0.0374879,wh...