Researchers have developed an extended deep energy method to simulate thermo-mechanical crack propagation, a complex phenomenon involving coupled heat conduction and crack growth. This new method uses two neural networks to model temperature and displacement, with the crack represented by a scalar embedding function. The approach incorporates Williams' expansion for displacement near the crack tip and Monte Carlo integration for energy calculations. The method's accuracy has been validated against existing solutions for various crack scenarios, showing strong agreement. AI
RANK_REASON The cluster contains a research paper detailing a new computational method. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →