Optimizing Spectral Prediction in MXene-Based Metasurfaces Through Multi-Channel Spectral Refinement and Savitzky-Golay Smoothing

The prediction of electromagnetic spectra for MXene-based solar absorbers is a computationally intensive task, traditionally addressed using full-wave solvers. This study introduces an efficient deep learning framework incorporating transfer learning, multi-channel spectral refinement (MCSR), and Savitzky-Golay smoothing to accelerate and enhance spectral prediction accuracy. The proposed architecture leverages a pretrained MobileNetV2 model, fine-tuned to predict 102-point absorption spectra from $64\times64$ metasurface designs. Additionally, the MCSR module processes the feature map through multi-channel convolutions, enhancing feature extraction, while Savitzky-Golay smoothing mitigates high-frequency noise. Experimental evaluations demonstrate that the proposed model significantly outperforms baseline Convolutional Neural Network (CNN) and deformable CNN models, achieving an average root mean squared error (RMSE) of 0.0245, coefficient of determination \( R^2 \) of 0.9578, and peak signal-to-noise ratio (PSNR) of 32.98 dB. The proposed framework presents a scalable and computationally efficient alternative to conventional solvers, positioning it as a viable candidate for rapid spectral prediction in nanophotonic design workflows.

• Publication year

  2026

• Venue

  Unknown venue

• Publication date

  2026-02-09

• Fields of study

  Materials Science, Physics, Computer Science, Engineering, Environmental Science

• Identifiers
  arXiv 2602.08406
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Predictive Modeling of MXene-Based Solar Absorbers Using Deep Neural Network

  2025influential reference
• Machine-Learning-Assisted Design of Multiband Terahertz Metamaterial Absorber

  2023cited by this paper
• Machine learning in prediction of MXenes-based metasurface absorber for maximizing solar spectral absorption

  2023influential reference
• Guest Editorial Special Issue on Metamaterials, Metadevices, and Applications

  2023cited by this paper
• Topological Photonic Crystals: Physics, Designs, and Applications

  2022cited by this paper
• Plasmonic Hot-hole Injection Combined With Patterned Substrate for Performance Improvement in Trapezoidal PIN GaN Microwire Self-Powered Ultraviolet Photodetector

  2022cited by this paper
• Optical spin-symmetry breaking for high-efficiency directional helicity-multiplexed metaholograms

  2021cited by this paper
• SERSNet: Surface-Enhanced Raman Spectroscopy Based Biomolecule Detection Using Deep Neural Network

  2021cited by this paper
• Switchable and Tunable Deep Beamformer Using Adaptive Instance Normalization for Medical Ultrasound

  2021cited by this paper
• Recent Advances in Plasmonic Nanostructures for Enhanced Photocatalysis and Electrocatalysis

  2020cited by this paper
• Searching for MobileNetV3

  2019cited by this paper
• Deformable ConvNets V2: More Deformable, Better Results

  2018cited by this paper
• MobileNetV2: Inverted Residuals and Linear Bottlenecks

  2018influential reference
• Training Deep Neural Networks for the Inverse Design of Nanophotonic Structures

  2017cited by this paper
• Deformable Convolutional Networks

  2017cited by this paper
• Metalenses: Versatile multifunctional photonic components

  2017cited by this paper
• Adam: A Method for Stochastic Optimization

  2014influential reference
• Controlling Vector Bessel Beams with Metasurfaces

  2014cited by this paper
• A Highly Efficient Domain Decomposition Method Applied to 3-D Finite-Element Analysis of Electromechanical and Electric Machine Problems

  2012cited by this paper
• Understanding the difficulty of training deep feedforward neural networks

  2010cited by this paper
• A black body absorber from vertically aligned single-walled carbon nanotubes

  2009cited by this paper
• Stable implementation of the rigorous coupled-wave analysis for surface-relief gratings: enhanced transmittance matrix approach

  1995cited by this paper
• Smoothing and differentiation of data by simplified least square procedure.

  1964cited by this paper

• No citing papers are available for this paper.