Researchers have introduced a mesh-based neural energy method (M-NEM) to improve the simulation of heterogeneous materials. This new approach addresses limitations of existing physics-informed neural networks, such as the deep energy method (DEM), which can lead to errors and high computational costs. The M-NEM utilizes a mesh discretization and shape functions to enforce kinematic constraints, significantly reducing stress errors and increasing speed compared to conventional methods. The study also found that radial basis function neural networks (RBFNNs) perform optimally within the M-NEM framework for handling sharp material gradients. AI
IMPACT This method could lead to more accurate and efficient simulations for complex materials in engineering and scientific research.
RANK_REASON The item is a research paper published on arXiv detailing a new computational method. [lever_c_demoted from research: ic=1 ai=1.0]
- arXiv
- deep energy method
- Kolmogorov-Arnold Networks
- mesh-based neural energy method
- Multi-Layer Perceptrons
- neural energy method
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
