Trumpf develops laser welding AI application to improve the quality of electric vehicle products

Trumpf has developed an AI application for laser welding (artificial intelligence) that makes production more efficient - especially when it comes to electric vehicles.

With Trumpf's new AI application, electric vehicle manufacturers can now produce more electric motors in less time, they said. There is also less rework and scrapping. The Trumpf experts explained that they want to take the entire laser, optical, sensor and software system to a new level of performance through AI. It seems to have worked. The AI process for laser processing has already been demonstrated in the mass production practice of electric vehicle components. It is considered flexible enough to be used in a variety of laser welding applications, Trumpf noted.

AI helps support laser sensors for component positioning

At the Laser - World of Photonics trade fair (Munich, June 27-30), Trumpf will demonstrate to the public for the first time how artificial intelligence can support laser welding. Ditzingen experts are demonstrating an AI-based system called Easymodel AI (AI from artificial intelligence) that can create corresponding algorithms. In addition, one can appreciate Visionline Detect's AI filters for applying algorithms, as Trumpf goes on to explain.

This is the reason behind it: To ensure that the weld is always in the right place, the laser sensor must precisely position the weld geometry on the part, otherwise there is a risk of waste. Dirt or scratches on components, poor lighting in the work area, or highly reflective materials such as copper make it difficult for sensors to locate. However, the Trumpf AI tool supports image processing, which reduces this interference.

Trumpf AI's targeted training is child's play

According to TrumPF, users only need to train the AI in high-quality laser welding before using it. The machine setter marks the weld-related parts area in the corresponding image material. As Trumpf promises, no programming knowledge is required.

Using the system is as simple and intuitive as using the well-known painting program on your smartphone or computer. The AI model created through this simple training is then used in the AI filter for Visionline Detect image processing.

Source: Laser Network