Researchers at Washington State University use artificial intelligence to search through a very large set of possible 3D-printing configurations for GRCop-42, a heat-resistant metal alloy used in aerospace applications. The study tests around 40 million possible ways to print the alloy out of an estimated 100 million settings, and reports that only a small number of configurations work.
The AI approach is described as finding printable parameters that enable production on more common commercial 3D printers, including using a laser at power levels as low as about 500 watts. According to the accounts, the AI-assisted search identifies feasible settings in roughly three months—faster than what the researchers report human-driven exploration achieved over many months. The outlets also say the method reduces cost and time for printing the alloy and could speed up research in other scientific and engineering areas that rely on optimizing complex processes.
