AI was used to create concrete with a service life of up to 200 years

20 Jan, 2025
Image: Yelena Afonina/TASS

Specialists from the Ogarev Mordovian State University have developed self-compacting, shrinkage-free fine-grained concrete that is two to four times superior to traditional heavy concrete in terms of useful properties.

"The design of concrete was carried out using machine learning methods - one of the areas of artificial intelligence. To solve the tasks, a database of experimental data was collected and analyzed, including over a thousand different compositions of cement material," the university press service noted.

The development is a dry construction mixture consisting of Portland cement, quartz sand and a complex organomineral additive based on surfactants, waste from metallurgical production and common sedimentary rocks. When mixed with water, it acquires the ability to self-compact and compensate for shrinkage deformations.

"Due to low permeability, high strength and corrosion resistance, the service life of the developed concrete under normal operating conditions can exceed 150-200 years. The cost price of the dry mix is 10-15% lower than existing analogues due to the reduced cement capacity of the composition, the use of local raw materials and industrial waste," said Associate Professor of the Department of Building Structures Artemy Balykov.

Machine learning models used in the development process can monitor predictive factors, learning as new data is received. This makes it possible to obtain information on the impact of various formulation and process parameters on the structure and performance of fine-grained concrete.

Source: TASS

Image: Yelena Afonina/TASS

Other news