The Fraunhofer Institute for Laser Technology ILT is set to host its third "AI for Laser Technology Conference" from 23-24 November in Aachen, Germany, where attendees will learn of the latest ways artificial intelligence (AI) can optimise laser processing tasks such as welding, cutting and additive manufacturing.
AI has been increasingly explored in recent years as a means of interpreting process monitoring data to detect deviations from the norm, which can signal a drop in laser processing quality and the need for machine maintenance.
By identifying and analysing such deviations in real time, AI could soon also be used to actively intervene to adjust a laser system’s parameters to implement quality control – so called ‘closed-loop’ processing. This would both increase laser processing efficiency and reduce the amount of defective parts that need to be scrapped.
AI can be used to quickly derive meaning from the multitude of data generated by today’s industrial laser systems and process monitoring cameras/sensors. Everything from the optical elements in the laser and processing head, to the beam itself and the interaction zone on the workpiece, can be monitored to obtain insightful data.
Such data is currently used to monitor and document the quality of individual processing steps, which enables users to evaluate changes in a laser process over time, either for single machines or across an entire series of laser systems. This can signal a drop in process quality and the need for maintenance.
"These AI capabilities are well known, now it's a matter of building new machines that are designed from the outset to make the most of AI's potential," says Professor Carlo Holly Head of Fraunhofer ILT’s Department for Data Science and Measurement Technology. "The sensor technology and the algorithms must fit together and ideally be planned in a coordinated manner from the outset."
AI-based process monitoring solutions are already becoming available, with Trumpf demonstrating its EasyModel AI at the Laser World of Photonics earlier this year, and Coherent intending to equip its HIGHvision system with AI-assisted algorithms in the future.
The holy grail of AI laser processing, however, is ‘closed-loop’ processing, which according to the Fraunhofer ILT is the “next big thing” that will expand the use of AI beyond the quality monitoring and predictive maintenance of machines that it is currently used for.
Such closed loop processing would be achieved by self-learning machines, which Fraunhofer ILT envisions working in four steps:
- Process monitoring sensors capturing data from the laser system and interaction zone
- The captured data being analysed using AI and made understandable through interpretations based on existing data
- The system simulating how the results of the process will develop depending on certain parameters, or by extrapolating a previous trend
- System parameters being adjusted accordingly to control the processes
Such capability would therefore enable laser systems to adjust their parameters on the fly to achieve optimal results at the workpiece, dramatically reducing the occurrence of defects.
In addition to enabling closed loop processing, as digital process data becomes increasingly available, AI will be able to do even more, according to Fraunhofer ILT. For example, past data makes it possible for AI to calculate a suitable parameter range for a new machining task and specific materials, testing the process in simulations using digital twins. This could save time, resources and therefore cost when planning new applications. “AI can find an optimal solution within multi-parameter systems, for which a human would need much more time," confirms Holly. "We're already seeing AI in use, but that's just the beginning".
For more information on the AI for Laser Technology Conference 2023, for which Microsoft is lined up to deliver a plenary talk on AI software and infrastructure, click here.
Image: Fraunhofer ILT
Explore more AI content: Artificial intelligence augments laser processing | AI-based quality monitoring of auto-body weld joints in real time | How can AI benefit industrial laser systems users? | How is artificial intelligence used in laser processing? | Towards in-situ flaw detection in LPBF through layerwise imagery and machine learning | New project to enhance LPBF using artificial intelligence | Machine learning-powered method detects LPBF pore formation in real time | Research project integrating AI into laser cutting & welding