Nanostructure design method based on deep learning

Abstract

The invention provides a nanostructure design method based on deep learning, and the method comprises the steps: building a training data set which comprises any kind of nanostructure information; preprocessing the training data set to transpose and normalize the data in the training data set; constructing a spectrum prediction network SPN and a geometrical shape prediction network GPN; integrating the spectrum prediction network SPN and the geometrical shape prediction network GPN into a neural network model, wherein the output of the spectrum prediction network SPN is connected with the input of the geometrical shape prediction network GPN; training the integrated neural network model by using the preprocessed training data set, and finishing the training of the neural network model whenthe loss function reaches a preset value; and inputting the frequency spectrums and the material properties in the X and Y polarization directions corresponding to the to-be-established nanostructureinto the trained neural network model to obtain eight nanostructure characterization points of the nanostructure, thereby obtaining the geometrical shape of the nanostructure.

Description

technical field [0001] The invention belongs to the technical field of nanophotonics and deep learning, and in particular relates to a nanostructure design method based on deep learning. Background technique [0002] In recent years, the field of nanophotonics has revolutionized the field of optics by manipulating photon-matter interactions on subwavelength structures. However, nanostructured materials require extensive manual fabrication, so the spectrum and structure of envisioned metasurfaces must be accurately predicted in advance. This is limited by complex iterative processes, cyclic modeling, nanofabrication and nanocharacterization. The fundamental reason for this is that the complex physical mechanisms that describe these light-matter interactions at the nanoscale cannot be resolved using generalized theories, thus predicting the optical properties and approximate structures of materials relies on finite element modeling (FEM) or time-limited Advanced iterative co...

AI Technical Summary

Problems solved by technology

[0009] The technical problem to be solved by the present invention is to provide a...

Images