The RUSAL Engineering and Technology Center has developed an innovative system for analyzing aluminum ingots, based on machine vision technologies and neural network algorithms. The use of artificial intelligence has made it possible to significantly speed up the process of studying the microstructure of metal and increase the accuracy of the assessment, the company’s press service reported.
Aluminum ingots are used in mechanical engineering, construction and other key industries. Their quality directly affects the performance characteristics of the final product, as well as the durability of equipment working with this material.
To assess eight parameters of the ingots, including grain size, amount and size of impurities, experts created individual neural network models. The system analyzes data in 15 minutes, producing results as accurate as traditional laboratory tests, which take from 1.5 to 4 hours. This eliminates the human factor and possible errors.
The technology is currently being tested in the laboratory of the Engineering and Technology Center. In the near future, RUSAL plans to adapt it for use at production enterprises, which will automate quality control and increase the efficiency of production processes.
Source: dprom.online
