Intelligent and Efficient Optimization Design for Ultra-Wideband High-Performance Complex Wave-Absorbing Structures

This paper presents a intelligent approach for designing and optimizing multi-layer resistive film absorbing structures across various frequency ranges. Initially, a deep neural network (DNN) is trained to accurately forecast the wave absorption performance of the model. Subsequently, utilizing the genetic algorithm and particle swarm optimization (GA-PSO), parameters are optimized for absorption within the target frequency band and the overall thickness of the model, establishing a rapid and effective design methodology tailored to diverse requirements. The proposed algorithm is employed to simulate and validate the optimization outcomes for the 8-100GHz and 2-20GHz frequency bands, achieving an average absorption rate exceeding 90% throughout the bands. This method simplifies the design process significantly and facilitates intelligent optimization for intricate absorption structures.