Viscoelastic Behavior
Viscoelastic behavior, the ability of materials to exhibit both elastic (spring-like) and viscous (fluid-like) properties, is a key area of research focusing on accurately modeling and predicting material response under various conditions. Current efforts leverage machine learning, particularly neural networks (including convolutional, recurrent, and physics-informed architectures), to efficiently identify governing equations and design materials with programmable viscoelastic responses from limited data, overcoming challenges posed by complex material behavior and experimental limitations. This research is crucial for advancing diverse fields, including material science, robotics (especially soft robotics), and biomechanics, enabling the design of novel materials with tailored properties and improved simulation of complex systems.
Papers
Physics3D: Learning Physical Properties of 3D Gaussians via Video Diffusion
Fangfu Liu, Hanyang Wang, Shunyu Yao, Shengjun Zhang, Jie Zhou, Yueqi Duan
Physics-Informed Neural Network based inverse framework for time-fractional differential equations for rheology
Sukirt Thakur, Harsa Mitra, Arezoo M. Ardekani
Design and Characterization of Viscoelastic McKibben Actuators with Tunable Force-Velocity Curves
Michael J. Bennington, Tuo Wang, Jiaguo Yin, Sarah Bergbreiter, Carmel Majidi, Victoria A. Webster-Wood
Real-time simulation of viscoelastic tissue behavior with physics-guided deep learning
Mohammad Karami, Hervé Lombaert, David Rivest-Hénault
Data-driven anisotropic finite viscoelasticity using neural ordinary differential equations
Vahidullah Tac, Manuel K. Rausch, Francisco Sahli-Costabal, Adrian B. Tepole