LENA 3D

Today, digitisation is a crucial element to increase efficiency in industrial production environments. In the ideal case, factories, production lines and ICT1 can be planned from scratch (Greenfield approach). Existing production plants with correspondingly high life-cycles, which represents the opposite to the above, are called “Brownfield”. It is both a task and a chance to utilise software/AI which, compared to machines and plants, has relatively short life-cycles, to continuously make processes more transparent and efficient. Therefore, the focus should lie on imaging process monitoring for the manufacturing technology of metal printing. Metal printing is a new manufacturing technology, experiencing a major growth. Despite its young age, lots of process data and experience is already available. Thus, methods and tools for process optimisation are key to further increase the economic efficiency for both new and existing printing systems. The image displays an overview of examples for a camera-based process monitoring. The image centre shows measured values and evaluations by specialised, additional sensors. In both cases, the upper and the lower row, the process is assessed as “good” by the sensors. However, a look at the simple black & white camera image indicates that the second case, the lower row, is a faulty process.

The goal of this project is: 1) to improve the automated image data analysis regarding accuracy, 2) to reduce the dependency from specific sensors/boundary conditions and 3) to develop methods for the continual improvement of existing machines/systems.