高取向性石墨烯/磁性粒子混合填充复合材料是目前吸波、电磁屏蔽材料研究的热点之一，但对其电磁特性的建模与设计仍然受困于理论研究的严重滞后。为此，本项目在“背景-粒子-网络”三层级理论框架下，拟：①提出一种高取向性石墨烯复合材料等效介电常数计算方法，通过修正“粒子”的非完美界面模型和构建“网络”的等效电路模型，以解决现有理论研究在单一填充条件下仍然存在的问题；②进而以此为基础，提出使用完美界面模型、进一步修正的非完美界面模型和混合等效电路模型来实现“背景-粒子-网络”的三层级理论建模，最终将理论研究扩展至高取向性石墨烯/磁性粒子混合填充的情况；③基于机器学习方法，挖掘等效电磁参数闭式表达式中的待定系数与本征材料、结构参数之间的隐含关系，以期服务于这些未知系数的高可信性估计，从而避免理论研究的可行性悖论；④在上述研究的基础上，实现对此类材料电磁特性的优化设计。研究成果将为实际工程提供理论指导。

The nanocomposite with aligned-graphene and magnetic particle hybrid fillers is one of the research hotspots in the field of electromagnetic wave absorbing/shielding materials. However, modeling and design of electromagnetic properties for this kind of nanocomposite is suffering from the serious lag of corresponding theoretical research. Hence, based on the three-level frame, i.e., the “background-particle-network” frame, this project is organized as follows. ①The effective permittivity of aligned-graphene composite is calculated by modifying the imperfect interface model of "particle" and constructing the equivalent circuit model of "network". ②The three-level theoretical modeling for nanocomposite with hybrid fillers is proposed by using the perfect interface model, the further modified imperfect interface model and the hybrid equivalent circuit model. ③The machine learning is applied to mine the underlying relationships between the intrinsic material and structure parameters and the unknown coefficients in the closed-form expressions of effective electromagnetic parameters. ④The proposed methods are used to optimize the electromagnetic properties for this kind of nanocomposite. This project can provide theoretical guidance for practical engineering.