Artificial intelligence and machine learning in design of mechanical materials.

Area of Science:

• Materials Science
• Mechanical Engineering
• Computational Science

Background:

• Materials and mechanical engineering increasingly utilize artificial intelligence (AI), specifically machine learning (ML) and deep learning (DL).
• Complex material structures create vast design spaces, challenging conventional design and optimization methods.
• AI offers powerful tools for predicting material properties, designing new materials, and uncovering underlying mechanisms.

Purpose of the Study:

• To review the state-of-the-art ML models and algorithms applied in materials science and mechanical engineering.
• To highlight the critical role of data collection, generation, and preprocessing in ML-based material design.
• To summarize applications of ML in mechanical property prediction, de novo material design, and computational methods.

Main Methods:

• Introduction to current ML models, algorithms, and structures.
• Discussion on data requirements, including collection, generation, and domain-specific preprocessing.
• Summary of ML applications in predicting mechanical properties and designing materials.

Main Results:

• ML models trained on large, curated datasets enable rapid exploration of complex material design spaces.
• AI techniques provide new avenues for overcoming long-standing challenges in mechanics.
• Breakthroughs in ML facilitate the development of unprecedented materials design strategies.

Conclusions:

• ML and DL are transformative tools in materials and mechanical engineering, enabling efficient design and discovery.
• The success of ML-based approaches hinges on high-quality data and appropriate model selection.
• The field presents significant opportunities and open challenges for future research and development.