StressNet - Deep learning to predict stress with fracture propagation in brittle materials

Engineering & Materials Science

• Brittleness 100%
• Deep learning 81%
• Residual stresses 79%
• Time-to-failure 18%
• Long short-term memory 17%
• Coalescence 17%
• Convolutional neural networks 13%
• Fracture toughness 13%
• Costs 11%
• Cracks 10%

Chemical Compounds

• Error 68%
• Coalescence 33%
• Time 33%
• Resistance 19%