Personal tools
You are here: Home › Research Trends & Opportunities › New Materials Technology and Applications › AI and Computational Materials
 

AI and Computational Materials

Cornell University_011121C
[Cornell University]


Simulate Today, Innovate Tomorrow: 

The Future of Materials Science

 

- Overview

Computational materials science uses computers to understand and predict the properties and structures of materials. It's based on fundamental physics, thermodynamics, kinetics, mechanics, and numerical algorithms. 

Artificial intelligence (AI) uses sophisticated algorithms and data-driven models for problem solving and decision making. AI and computational materials science can be combined to:

  • Accelerate discovery and development: AI can improve the efficiency of hypothesis generation, testing, and data analysis.
  • Gain fundamental understanding: AI and machine learning (ML) models can extract patterns at spatiotemporal scales that were previously impossible.
  • Open new windows: Computational materials science and advanced experimental techniques can open new windows into the materials world.

 

Please refer to the following for more details:

 

- AI in Materials Science

Artificial intelligence (AI) is used in materials science to help with a variety of tasks, including:  

  • Material discovery: AI can help identify promising materials from large databases of known materials.
  • Material design: AI can help with advanced materials design.
  • Experimental data analysis: AI can help with analyzing experimental data.
  • Text and data mining: AI can help with retrieving information from text and data.
  • Material selection: AI can help with selecting the right material for a project.
  • Material optimization: AI can help optimize material properties.
  • Material sustainability: AI can help develop more sustainable materials by reducing the use of rare and expensive ingredients.
 

Some AI methods used in materials science include: Classical regression models, Bayesian optimization, Deep learning techniques, and Active learning approaches. 

Here are some examples of AI in materials science:

  • Pacific Northwest National Laboratory (PNNL): PNNL developed an AI model that can identify patterns in electron microscope images of materials.
  • Fictiv: Fictiv created Materials.AI, an AI assistant that helps users navigate the landscape of plastic and metal materials.
  • Johns Hopkins APL: Johns Hopkins APL developed a pathway that links alloy phases to their mechanical properties

In addition to predicting and optimizing material properties, AI has also been applied in the discovery of new materials. Through the use of ML (ML) algorithms, researchers have been able to identify promising material candidates from vast databases of known materials and predict their potential properties. 

 

[More to come ...]



Document Actions