In the precision-driven domain of laser welding, achieving consistent high-quality welds while maintaining process efficiency poses a significant challenge. This paper introduces a state-of-the-art AI system designed to optimize the laser welding process in real-time, addressing this challenge through advanced monitoring and control techniques. The system comprises two key components: an AI-based monitoring system and an AI-based controller. Utilizing data from a coaxial high-speed camera and a high-speed microphone, the monitoring system integrates multiple AI models and data fusion to predict five critical quality measures. These include surface quality parameters and subsurface characteristics, such as weld depth and binding width, with weld depth predictions achieving less than 10% error.

The control system leverages these real-time quality assessments to compute optimal process control parameters, ensuring continuous process optimization. Featuring seven control parameters—including beam oscillation in the x and y directions—the AI controller employs an AI-based system model and an optimizer. Operating with a takt-time of 3 ms and a reaction time of 25 ms, this low-latency system swiftly corrects unforeseen deviations, reducing weld defects by 30% in the critical use case of welding copper electrodes for batteries. A digital twin of the welding process, acting as a high-fidelity simulator, facilitated system testing and tuning before putting the system into real operation. The novelty of this work lies in its low-latency operation and high complexity, managing five quality measures and seven process parameters.