Scientists argue that as machine learning accelerates the creation of new materials for everything from electric motors to carbon capture, AI-designed paints could reduce the impact of urban swelling urban thermal islands and reduce air conditioning bills.
Material experts are using artificial intelligence to formulate new coatings that can keep a cooler building from 5c to 20c than regular paint after exposure to the noon sun. It can also be applied to cars, trains, electrical equipment and other objects. These objects require more cooling in a hot world.
According to a peer-reviewed study published in Science Journal Nature, researchers from universities in the US, China, Singapore and Sweden have designed new paintworks that are tailored to best reflect the sunlight and emit heat.
This is the latest example of AI being used to leaps through traditional trial and error approaches to scientific advancement. Last year, British company Matnex used AI to create a new kind of permanent magnet used in electric vehicle motors to avoid the use of carbon-intensive rare earth metals.
Microsoft has released AI tools to help researchers quickly design new inorganic materials. It is a crystal structure that is often used in solar panels and medical implants. And there is hope for new materials to better capture carbon in the atmosphere and create more efficient batteries.
Paint studies were conducted by scholars at the University of Texas Austin, Shanghai Ziaoton University, Singapore National University, and Umeo University in Sweden. In hot climates such as Rio de Janeiro and Bangkok, we found that applying one of the new AI-enabled paints to the roof of a four-storey apartment block could save 15,800 kilowatt hours of electricity per year. When paint is applied to 1,000 blocks, it saves enough power to power over 10,000 air conditioning units a year.
“We are a scientist at Texas University of Texas, and a co-leader of research,” said Yuebbing Zeng. “Our machine learning framework represents a key leaps in the design of thermal meta-emitters. By automating processes and expanding the design space, we can create materials with superior performance that were previously unimaginable.”
He said a month of work using AI to design new materials in just a few days, creating new materials that may not have been discovered through trial and error.
“Now we follow the machine learning output, [its instructions for] It can be done right without going through many design and manufacturing test cycles, as well as many design and manufacturing test cycles. ”
Dr. Alex Ganoce, a lecturer at Imperial College, London, also stated: “We also use machine learning to design new materials.” “Things are moving very quickly in this area. Last year or so, there were so many startups trying to use generative AI for materials.”
He said the process of designing new materials could involve calculations of millions of potential combinations. AI allows materials scientists to push past limitations on computing power. It also allows scientists to communicate the necessary traits to AI in advance, thus reversing the traditional process of creating materials and testing those traits.
