基于U-Net架构与球坐标系物理规律的地球-电离层波导三维电磁场预测方法及应用
By combining the physical information neural network of spherical coordinate system physical laws and impedance boundary condition constraints with the U-Net architecture, the problems of high efficiency, accuracy and physical consistency in predicting the three-dimensional electric field distribution of the Earth-ionospheric waveguide are solved, and it is suitable for electromagnetic field analysis in complex dynamic scenarios.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- ZHEJIANG UNIV
- Filing Date
- 2026-05-13
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies cannot simultaneously meet the requirements of high accuracy, high efficiency, strong generalization and physical consistency in predicting the three-dimensional electric field distribution of the Earth-ionospheric waveguide, especially in complex dynamic scenarios where accurate electromagnetic field distribution analysis is difficult to achieve.
A physical information neural network based on the U-Net architecture is constructed, which combines the physical laws of the spherical coordinate system, the ground impedance boundary conditions, and the ionospheric impedance boundary conditions as constraints. The end-to-end prediction of the three-dimensional electromagnetic field is achieved through iterative training.
It achieves high-precision and rapid prediction of electromagnetic field distribution in complex ionospheric environments, adapts to the global spherical geometry of the Earth-ionospheric waveguide, and improves the model's generalization ability and physical consistency.
Smart Images
Figure CN122174701B_ABST