Machine Learning Transforms Laser Metal Manufacturing With Real-Time Precision And Efficiency

Empa scientists are harnessing the power of artificial intelligence to automate and optimize laser-based metal production, slashing development time, boosting accuracy, and enabling non-experts to achieve industry-grade results with unprecedented efficiency.

When the laser learns: Laser-based welding processes can be optimized in real time thanks to machine learning. Image: Empa

*Important notice: arXiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as definitive, used to guide development decisions, or treated as established information in the field of artificial intelligence research.

Laser-based processes for metals are considered particularly versatile in the industry. Lasers can be used, for example, to precisely weld components together or produce more complex parts using 3D printing – quickly, precisely, and automatically. This is why laser processes are utilized in various sectors, including the automotive and aviation industries, where maximum precision is essential, as well as in medical technology, for example, for the production of customized titanium implants.

However, despite their efficiency, laser processes are technically challenging. The complex interactions between the laser and the material make the process sensitive to even the most minor deviations, whether in material properties or laser parameter settings. Even minor fluctuations can lead to errors in production.

“To ensure that laser-based processes can be used flexibly and achieve consistent results, we are working on better understanding, monitoring, and control of these processes,” says Elia Iseli, research group leader in Empa's Advanced Materials Processing laboratory in Thun. In line with these principles, Giulio Masinelli and Chang Rajani, two researchers on his team, aim to make laser-based manufacturing techniques more affordable, efficient, and accessible, utilizing machine learning.

Vaporize or melt?

First, the two researchers focused on additive manufacturing, i.e., the 3D printing of metals using lasers. This process, known as powder bed fusion (PBF), works slightly differently from conventional 3D printing. Thin layers of metal powder are melted by the laser in precisely the right spots so that the final component is gradually “welded” out of them.

PBF enables the creation of complex geometries that are virtually impossible with other processes. Before production can begin, however, a complex series of preliminary tests is almost always required. This is because there are basically two modes for laser processing of metal, including PBF: In conduction mode, the metal is simply melted. In keyhole mode, it is even vaporized in some instances. The slower conduction mode is ideal for thin and very precise components. Keyhole mode is slightly less precise but much faster, making it suitable for thicker workpieces.

Where exactly the boundary between these two modes lies depends on a variety of parameters. The right settings are necessary for achieving the best quality of the final product, and these vary significantly depending on the material being processed. “Even a new batch of the same starting powder can require completely different settings,” says Masinelli.

Better quality with fewer experiments

Typically, a series of experiments must be conducted before each batch to determine the optimal settings for parameters such as scanning speed and laser power for the respective component. This requires a lot of material and must be supervised by an expert. “That is why many companies cannot afford PBF in the first place,” says Masinelli.

Masinelli and Rajani have now optimized these experiments using machine learning and data from optical sensors that are already incorporated in the laser machines. The researchers “taught” their algorithm to “see” which welding mode the laser is currently in during a test run using this optical data. Based on this, the algorithm determines the settings for the next test. This reduces the number of preliminary experiments required by approximately two–thirds, while maintaining the quality of the product.

“We hope that our algorithm will enable non-experts to use PBF devices,” summarizes Masinelli. For the algorithm to be used in industry, it would only require integration into the firmware of the laser welding machines by the device manufacturers.

Real-time optimization

PBF is not the only laser process that can be optimized using machine learning. In another project, Rajani and Masinelli focused on laser welding, but went one step further. They not only optimized the preliminary experiments, but also the welding process itself. Even with ideal settings, laser welding can be unpredictable, for example, if the laser beam hits tiny defects on the metal's surface.

“It is currently not possible to influence the welding process in real time,” says Chang Rajani. “This is beyond the capabilities of human experts.” The speed at which data must be evaluated and decisions made is a challenge, even for computers. This is why Rajani and Masinelli used a special type of computer chip for this task, a so-called field-programmable gate array (FPGA). “With FPGAs, we know exactly when they will execute a command and how long the execution will take, which is not the case with a conventional PC,” explains Masinelli.

Nevertheless, the FPGA in their system is also linked to a PC, which serves as a kind of “backup brain”. While the specialized chip is busy observing and controlling the laser parameters, the algorithm on the PC learns from this data. “If we are satisfied with the performance of the algorithm in the virtual environment on the PC, we can 'transfer' it to the FPGA and make the chip more intelligent all at once,” explains Masinelli.

The two Empa researchers are convinced that machine learning and artificial intelligence can make significant contributions in the field of metal laser processing. That is why they continue to develop their algorithms and models, expanding their area of application in collaboration with partners from research and industry.

*Important notice: arXiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as definitive, used to guide development decisions, or treated as established information in the field of artificial intelligence research.