Metamaterial absorber using cascaded ring resonators and optimization through machine learning for sensing applications

In this research, a narrowband metamaterial absorber (MA) is designed using a cascaded ring resonator (RR) with a Metal-Insulator-Metal (MIM) structure. Conventional design methodologies such as finite-difference time-domain (FDTD) method often rely on trial-and-error simulations, which are both computationally intensive and time-consuming. To address this challenge, we present a design framework based on the Adaptive Neuro-Fuzzy Inference System (ANFIS). This hybrid approach, which combines neural networks and fuzzy logic, offers a cost-effective and time-efficient alternative to FDTD optical simulations. This type of hybrid network is recognized for its capability to represent highly nonlinear and complex samples. To predict the absorption through the ANFIS, four geometrical parameters of the structure have been utilized as input variables. The performance of this model is assessed by employing a root mean square error (RMSE) and mean absolute error (MAE) as a criteria for the evaluation, with two membership functions (MF) and a 5-fold cross-validation. The RMSE (0.0811) and MAE (0.062) indicate that absorption can be predicted with high precision. Based on our results, ANFIS achieves approximately 90% time savings compared to the conventional FDTD method, while maintaining acceptable prediction accuracy. Furthermore, the comparison of our method with Random Forest and k-Nearest Neighbor (KNN) shows that ANFIS yields lower prediction error. The results indicate that the proposed MA can achieve nearly perfect absorption at specific values of the geometrical parameters. As a key application, the proposed structure has been utilized as a sensor. The optimal results for two key factors are a sensitivity of 600 nm per refractive index unit (RIU) and a full-width at half-maximum (FWHM) of 4 nm. Due to its compact size, perfect absorption, and tunability, the proposed MA offers diverse applications in solar cells, sensors, and stealth technology.