Isogeometric boundary element method analysis for dielectric target shape optimization in electromagnetic scattering.

Area of Science:

• Computational electromagnetics
• Numerical analysis
• Applied mathematics

Background:

• Electromagnetic scattering analysis is crucial for designing advanced systems.
• Optimizing scattering forms requires accurate geometric representation and efficient computation.
• Existing methods may face challenges with geometric fidelity and computational cost during optimization.

Purpose of the Study:

• To develop a robust method for optimizing electromagnetic scattering forms in 2D dielectric media.
• To ensure geometric accuracy and prevent mesh over-refinement during the optimization process.
• To create an efficient and accurate surrogate model for multi-frequency scattering optimization.

Main Methods:

• Frequency-domain boundary element method integrated with isogeometric analysis (IGABEM).
• Discretization of boundary integrals using non-uniform rational B-splines (NURBS) for accuracy and speed.
• Development of a gray wolf optimizer-based back-propagation neural network (GWO-ANN) as a surrogate model.
• Utilizing radar cross-section (RCS) as the objective function for optimization.

Main Results:

• The IGABEM ensures geometric correctness and avoids mesh over-refinement.
• NURBS-based discretization enables rapid and accurate numerical computation.
• The GWO-ANN serves as an effective surrogate model for shape optimization.
• The proposed algorithm efficiently and accurately solves multi-frequency electromagnetic scattering problems.

Conclusions:

• The developed IGABEM combined with GWO-ANN provides an efficient and accurate approach for electromagnetic scattering shape optimization.
• This method enhances geometric control and computational performance in complex scattering problems.
• The approach is suitable for multi-frequency optimization tasks, offering significant advantages in design processes.