Over a million developers have joined DZone.

How AI Is Supporting Materials Science

DZone's Guide to

How AI Is Supporting Materials Science

AI is increasingly being used to explore for interesting possible protein combinations in medical research. Similar work is also underway in other fields.

· AI Zone ·
Free Resource

Insight for I&O leaders on deploying AIOps platforms to enhance performance monitoring today. Read the Guide.

AI is increasingly being used to explore for interesting possible protein combinations in medical research, but while that has perhaps gained most exposure, similar work is also underway in other fields.

A recent paper from a team of Japanese researchers highlights how such an approach can help with the design of advanced materials. The researchers used an algorithm that had previously been deployed to find the best moves in a video game to analyze the best combination of atoms within a structure.

Materials Design

The team utilized a method known as materials design using tree search (MDTS) using the Monte Carlo tree search. It's an approach that's often used in video games to determine the best strategy in a video game.

The researchers attempted to find the best way to design silicon-germanium alloy structures, which are invaluable in the conducting of heat. Such materials, which have minimal thermal conductance, are great as they recover waste heat from industrial processes and can use it as an energy source. They're especially useful in areas such as computing as they draw heat away from the CPU.

The algorithm went through a series of iterative learning phases to explore all possible positions for the best placing for the silicon or germanium to achieve the best level of thermal conductance.

"MDTS is a practical tool that material scientists can easily deploy in their own problems and has the potential to become a standard choice," the authors say.

The project is part of a growing number that use AI to further our understanding of materials. For instance, the Material Genome Project aims to predict the lifespan of any machine.

Similarly, a project from researchers at the Lawrence Berkeley National Laboratory developed algorithms to predict defects in certain intermetallic compounds with extremely high levels of accuracy. The hope is that it will support the development of advanced alloys and lightweight materials for industries ranging from aerospace to automotive.

It will form part of the Materials Project, which is an open source initiative that provides open, web-based access to information about materials, together with an open-source, Python-based toolkit to allow people to model point defects in things such as semiconductors.

TrueSight is an AIOps platform, powered by machine learning and analytics, that elevates IT operations to address multi-cloud complexity and the speed of digital transformation.

materials science ,ai ,algorithms

Published at DZone with permission of

Opinions expressed by DZone contributors are their own.

{{ parent.title || parent.header.title}}

{{ parent.tldr }}

{{ parent.urlSource.name }}