Accelerating Auxetic Metamaterial Design with Deep Learning. Issue 5 (20th May 2020)

• Record Type: Journal Article
• Authors: Wilt, Jackson K.; Yang, Charles; Gu, Grace X.
• Is Part Of: Advanced engineering materials. Volume 22:Issue 5(2020)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2020-05-20
• Subjects: Materials -- Periodicals; 620.11
• DOI: 10.1002/adem.202070018
• Languages: English
• ISSNs: 1438-1656

Abstract

Abstract : Auxetic metamaterials are designed using a machine learning workflow. The 2D metamaterial lattice is simulated computationally to contain gradients of Poisson's ratio values for each unit cell according to pseudorandomized image generation. Thousands of simulations are used to train the neural network resulting in a model for predicting the deformation deviation of potential solutions. Further information can be found in the article, number 1901266, by Grace X. Gu and co‐workers.