Enhancing Electromagnetic Parameter Modeling of Irregularly Shaped Composite Structures with Partitioned Equivalent Modeling and Near-Field Scanning

In aerospace applications, composite materials are widely used due to their flexibility and performance advantages. Therefore, it is necessary to model the composite structure parametrically to evaluate the overall electromagnetic shielding effectiveness (SE) of the aircraft. Composite materials used in aircraft structures often exhibit irregular geometries, such as curvatures and slots. These geometric complexities lead to spatial variations in electromagnetic parameters, which makes it difficult to accurately evaluate the SE by using conventional modeling methods. To address the challenge of obtaining low-frequency electromagnetic parameters such as the dielectric constant \varepsilon for irregular structures, the partitioned equivalent modeling (PEM) method is proposed in this study. The PEM method consists of a partition modeling framework for multidirectional field interactions, an enhanced near-field scanning method for localized parameter extraction, and a physics-guided deep neural network (DNN) for coupling analysis. The proposed PEM method decomposes irregular composite materials into homogeneous partitions, and electromagnetic parameter reconstruction is performed by using S-parameter-based inversion algorithms. Numerical simulations have verified the effectiveness of the method in the low-frequency domain, which is the primary concern of this study. To further substantiate the method’s robustness and generality, additional simulations are conducted to evaluate its performance in the high-frequency range. A DNN model is employed to predict equivalent electromagnetic parameters, with results demonstrating strong agreement with reference data. Experimental validation using a near-field measurement system in low-frequency range demonstrates the practical applicability of the methodology, which addresses gaps in low-frequency near-field characterization. This study proposes a comprehensive framework for electromagnetic parameterization of irregular composite structures by integrating theoretical modeling, numerical simulation and experimental validation. The proposed PEM method reliably captures localized electromagnetic properties, offering an effective solution for low-frequency SE analysis of geometrically complex composites.