New deep learning method efficiently handles random material parameters in solid mechanics

Researchers have developed a new physics-driven deep learning framework called the Constitutive Parameterized Deep Energy Method (CPDEM) to address challenges in solid mechanics simulations with random material parameters. Unlike traditional methods that require costly re-computation or extensive datasets for each parameter variation, CPDEM reformulates the strain energy density functional to encode a latent representation of stochastic constitutive parameters. This allows for real-time, zero-shot inference of displacement fields for unknown material parameters without retraining, making it efficient for complex nonlinear mechanics problems. AI

IMPACT Enables faster and more efficient simulations for solid mechanics problems with variable material properties.

RANK_REASON The cluster contains a research paper detailing a novel method for solid mechanics simulations. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• Constitutive Parameterized Deep Energy Method
• deep energy method
• finite element method
• Zhangyong Liang

• paper
• infra

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →