In 2018, Meta committed to minimizing its environmental footprint and is aiming for net zero emissions for its value chain by 2030. However, it plans to build eight data centers. To reduce the carbon emissions that this will generate, the META team, with the help of Lav Varshney and Nishant Garg from the University of Urbana-Champaign, designed a low-carbon concrete thanks to algorithms of generative machine learning that she tested at the center in Delkab, Illinois.
Concrete has been used for millennia for the construction of buildings or structures. Admittedly, it has evolved a lot and if cement is now one of its ingredients, it also represents the major source of its greenhouse gas emissions.
Concrete is the most widely used construction material in the world due to its low cost and ease of manufacture, and nearly 10 billion tonnes of it are produced each year.
It is made by mixing cement, sand and gravel to which water is added. Cement is obtained by heating a mixture of clay and limestone to over 1,400°C in a kiln, which produces 600 kilograms of carbon dioxide per ton of cement produced.
Cement production is estimated to be the source of 8% of global carbon emissions. Research to reduce this impact is being carried out, whether to replace the fossil fuels used to produce cement (hydrogen, etc.), or to reduce its quantity in the composition of concrete, by replacing it with fly ash or slag (leftovers of the production of iron or steel).
Meta has fully fulfilled its 2018 commitments: to have reduced its greenhouse gas emissions in 2020 by 75% compared to 2017 and to use 100% renewable energy within the same period. It even exceeded these objectives and achieved scopes 1 and 2 of the GHG Protocol (Greenhouse Gas Protocol). To achieve the Scope 3 standard, it is necessary to build its data centers with the most ecological materials possible.
Meta’s low-carbon concrete built using AI
The Meta team asked themselves the following question:
“Is it possible to formulate concrete mixes with standard ingredients so that they have half the embodied carbon of traditional formulations, but are just as strong? »
Answering this question is complex given the number of possible combinations between the various ingredients of concrete.
Amruta Sudhalkar, Méta’s Meta Sustainability Program Manager, said:
“Manually optimizing a concrete formula for results that are both durable and technically sound…is a formidable challenge. Factors that go into making good concrete, especially one that tries to replace its cement content, mean not only considering ingredient ratios, but also external factors such as weather and location. »
the machine learning is the solution to such problems. L’équipe de Meta s’est donc associée au рrоfеѕѕеur Lаv Vаrѕhnеy, du déраrtеmеnt dе génіе élесtrіquе еt іnfоrmаtіquе, еt au рrоfеѕѕеur Nіѕhаnt Gаrg, du déраrtеmеnt dе génіе сіvіl dе l’unіvеrѕіté dе Urbаnа-Сhаmраіgn, pour entraîner un algorithme grâce à un еnѕеmblе dе dоnnéеѕ ѕur ѕіѕtаnсе соmрrеѕѕіоn соmрrеѕѕіоn of the concrete.
1,030 concrete mixes were listed, the algorithm compared them taking into account the criteria of hardening time and carbon footprint. Five mixes were selected by Meta, tested in the laboratory and then adjusted by the concrete manufacturer Ozinga Bros.
The latter then tested a concrete where the cement was replaced at 50% by fly ash and slag in real conditions for a small part of the foundations of the center of Elkab. According to Meta, carbon emissions were found to be 40% lower than regional benchmarks and the concrete stronger in seven- and 28-day strength tests.
This test took place when the temperatures were low, the concrete took a long time to set and harden, which causes delays during a construction. Meta said:
“We strive to improve the initial strength performance of the concrete after three to five days and to take into account the impact that variations in environmental conditions, such as temperature and wind, can have on the performance of the concrete. concrete. »
We would like to give thanks to the author of this short article for this remarkable material
Thanks to artificial intelligence, Meta can build its data centers with low-carbon concrete
Take a look at our social media profiles and also other related pageshttps://www.ai-magazine.com/related-pages/