RESEARCH ON REVERSE DESIGN OF BAND GAP FOR ELASTIC METAMATERIALS BASED ON DEEP LEARNING

Elastic metamaterials are the artificial microstructures with the extraordinary mechanical and acoustic properties, possessing unique bandgap characteristics. By dynamically adjusting their bandgap design, they can meet the specific requirements in the aerospace field for vibration reduction and noise reduction performance, and have a good application prospect.. In this article, based on the deep learning methods, research on reverse design of the elastic metamaterial bandgap is conducted. Firstly, considering the advantages of deep learning models in image processing, the single-cell structure is represented in the form of a pixel image and converted into pixel data, and the sample configurations are generated based on the parametric curve description and the expansion function operation, which combines the advantages and disadvantages of the two methods to quickly generate a large amount of sample data with bandgap features and good diversity. Secondly, conditional generative adversarial network (cGAN) is used for the reverse design of elastic metamaterials. The dispersion curves are used as sample conditions. Via the PatchGAN discriminator, the error details of more image regions can be found to improve the network's ability to process image details. In the training of neural networks, the new error evaluation method is introduced to improve the accuracy of inverse design. And then, the reverse design of the band gap of elastic metamaterials is implemented, including widening, generation, and amplification of the band gap. Finally, the reverse design of elastic metamaterial bandgap is carried out, including bandgap broadening, generation and increase, etc, with small error. The frequency response curves of the reverse generated structure are obtained by numerical calculation. By comparison, the elastic wave attenuation range of elastic metamaterials obtained by the reverse design matches the theoretically designed bandgap region, which verifies the reliability of the reverse design technology. This work provides an effective method for the elastic metamaterials bandgap modulation.