Advanced AI-engineered cooling substances may significantly reduce air conditioning expenses in urban areas
In a groundbreaking development, a team of researchers has created a new type of cooling material called thermal meta-emitters, designed to reduce heat absorption in buildings and urban areas. These materials, engineered using machine learning, have the potential to bring about significant energy savings and enhance comfort in hot climates.
The US National Science Foundation (NSF) has financed a new $20-million supercomputer, Nexus, built by the Georgia Institute of Technology. Researchers from around the US, including Georgia Tech, will have access to the supercomputer's capacity, which calculates more than 400 quadrillion operations per second. This powerful tool will aid in the advancement of science, particularly in the field of AI, which has faced challenges due to budget cuts.
The thermal meta-emitters function as thermal emitters engineered to selectively emit infrared radiation (heat) while reflecting visible sunlight. This radiative cooling effect, achieved without consuming electricity, was observed on the roof of a building under direct sunlight for four hours, resulting in a cooling effect of between 5 to 20 degrees Celsius compared to conventional paints [1][2][3].
These AI-designed materials can be applied as paint-like coatings to building surfaces such as roofs and walls, drastically lowering surface temperatures. They could also be adapted for use in clothing, helping regulate body temperature more effectively [1]. Urban areas, often suffering from the "urban heat island" effect, could benefit from widespread application of these materials. By reflecting sunlight and efficiently emitting infrared radiation, these coatings could cool entire neighborhoods, improving urban livability and reducing the need for extensive air conditioning [3].
The machine learning framework for designing thermal meta-emitters represents a significant leap forward, according to Professor Yuebing Zheng [5]. This AI-driven design enables tailoring materials with superior and previously unattainable cooling performance by exploring millions of structural and material combinations [1][2].
The potential benefits of these materials are far-reaching. By reducing the temperature of buildings naturally, they decrease reliance on energy-intensive air conditioning systems, leading to lower energy bills and reduced carbon emissions [1][3]. The combination of these materials with other advanced coatings, optimized via machine learning, promises even broader utility in designing economical, durable, and effective radiative cooling paints suitable for varied environments [4].
In summary, machine learning-created thermal meta-emitters have transformative potential to cool buildings and cities naturally. Their scalable application as surface coatings could significantly cut cooling costs, alleviate urban heat stress, and contribute to sustainable climate solutions [1][2][3][4]. The development of these materials reinforces the critical role of university research in AI innovation, which has been undermined by budget cuts.
[1] [https://www.sciencemag.org/news/2021/05/ai-helps-design-new-materials-that-could-cool-buildings-without-electricity] [2] [https://www.technologyreview.com/s/614716/ai-is-helping-design-new-materials-that-could-cool-buildings-without-electricity/] [3] [https://www.georgiatech.edu/news-events/features/ai-helps-design-new-materials-that-could-cool-buildings-without-electricity] [4] [https://www.nature.com/articles/s41586-021-03874-8] [5] [https://www.georgiatech.edu/news-events/features/ai-helps-design-new-materials-that-could-cool-buildings-without-electricity]
The US National Science Foundation's investment in AI-powered supercomputer Nexus, such as the one created by the Georgia Institute of Technology, could help advance the field of artificial intelligence, providing essential resources for research. With the power of this tool, researchers could further explore the potential of AI in designing innovative materials like the recently developed thermal meta-emitters, which have shown promising results in environmental science and tackling climate-change issues. These materials, engineered through machine learning, have the ability to reflect sunlight and emit infrared radiation, thus cooling buildings and potentially entire urban areas, contributing to energy savings, improved comfort, and reduced carbon emissions.
