3D printing is not a simple process, especially metal 3D printing.It involves a lot of complex mathematical modeling, and even the most basic part of the calculation takes weeks. But scientists from Peter the Great St. Petersburg Polytechnic University have developed a neural network for metal 3D printing that is trained with a large number of parameters, which allows for the faster production of parts as well as the ability to use discovered dependencies to manufacture new parts.
Neural network is a computing system used to process big data input.Researchers at the university used this method to obtain the process parameters of 3D printing, ensuring the stability of the process.
The neural network developed by researchers in Russia is another step towards the automation of the entire additive manufacturing process, which not only has the potential to accelerate this process and improve the quality of parts, but reduces the risk of human error, which is high when it comes to complex mathematics. There is still a lot of waste of time, money and materials in the manufacture of metal additives due to construction failures, but with improvements like this one, these problems are likely to be greatly reduced in the future.