Skip to content Skip to navigation

Inverse Design of Mechanical Metamaterials with Target Nonlinear Response via a Neural Accelerated Evolution Strategy

A team at the Harvard MRSEC led by Bertoldi and Rycroft  has  developed  a  framework  to  design mechanical  metamaterials  with  target  nonlinear response. Neural networks were used to accurately learn  the  relationship  between  the  geometry  and nonlinear   mechanical   response   of   these metamaterials.   Next,   neural   networks   were combined  with  an  evolution  strategy  to  efficiently identify  geometries  that  exhibit  target  nonlinear stress-strain  behaviors.  Their  neural  accelerated evolution  strategy  holds  potential  for  a  range  of applications that benefit from systems with a target nonlinear mechanical behavior, as demonstrated by the design of energy absorbing systems, soft robots and morphing structures.