Material Fracturing and Failure Simulation Datasets.

Area of Science:

• Physics
• Materials Science
• Computational Mechanics

Background:

• Fracture phenomena are critical in diverse fields like infrastructure, aerospace, and geophysics.
• Predicting material failure requires robust simulation data.
• Existing simulation methods vary in fidelity and efficiency.

Purpose of the Study:

• To present a comprehensive dataset of simulated fracture evolution and material failure.
• To support the development of machine learning models for failure prediction.
• To provide a foundation for future research in computational fracture mechanics.

Main Methods:

• Generated fracture data using two distinct numerical solvers: phase-field and finite-discrete element method (FDEM).
• Simulated five materials: PBX, anisotropic shale, tungsten, aluminum, and steel.
• Conducted extensive simulations under various loading conditions (uniaxial and biaxial tension) with randomized initial fractures.

Main Results:

• Created a diverse dataset encompassing 490,000 simulation cases.
• Captured temporal fracture propagation dynamics for each simulation.
• The dataset includes detailed information on fracture evolution under different material properties and stress states.

Conclusions:

• The presented dataset is a valuable resource for advancing computational material failure analysis.
• It enables the development and validation of machine learning approaches for predicting fracture.
• Facilitates future research in fundamental physics and engineering applications of fracture mechanics.