Adjoint Sensitivity Analysis Based on Well-Conditioned Asymptotic Waveform Evaluation for Broadband Antenna Topology Optimization Acceleration

The well-conditioned asymptotic waveform evaluation (WCAWE) method is used for adjoint sensitivity analysis (ASA), with the aim of expediting gradient-based topology optimization of broadband antenna. The equations of WCAWE-based ASA are derived when S-parameters and radiation characteristics are served as objectives, which can cater to the design requirements of antennas with diverse performance. The WCAWE method fully leverages solutions from multiple expanded frequencies to achieve a fast frequency sweep (FFS), thereby effectively reducing the computational time for sensitivity in gradient-based optimization. The excellent acceleration effect of WCAWE-based ASA is validated through an ultrawideband (UWB) monopole antenna. Finally, topology optimization is applied to a broadband antenna with far-field radiation power as the objective function and S-parameters as constraint conditions. The optimized antenna exhibits excellent broadband performance, while the entire optimization process takes 409 min, which is only 36% of that required when using discrete frequency sweep (DFS).