AI-based set-up assistance system for transfer presses

The initial set-up and re-set-up of multi-stage moulds is a considerable challenge for set-up engineers and, depending on the complexity of the tool set, a lengthy process. If changes to the process parameters occur in one stage, which lead to changes in process forces, for example, this influences the process sequence in other stages. To recover good part production, the trained personnel utilise implicit knowledge, which in particular also includes the properties of the machine used. Changes in process conditions during production, for example due to wear or temperature changes during production start-up, lead to experience-based readjustments without the cause always being known or able to be rectified.

The research question to be answered as part of the project is whether process data combined with domain-specific knowledge can be processed and used in such a way that the gap between existing modelling approaches and reality can be closed with measurement data-driven modelling of implicit process relationships using AI-based methods.

The specific objectives are as follows

• Identification of significant influencing variables on component quality in step processes
• In-depth understanding of the relationships between process influencing variables in step processes
• Quantitative assessment of the effects of varying the influencing variables
• Utilisation of the knowledge to shorten set-up processes
• Optimisation of the design process for die-casting surfaces

The expected result is the provision of an AI-based set-up assistance system for optimal machine set-up by deriving recommendations for action for the initial and also recovery of good part production after expected or unexpected changes in the boundary conditions of multi-stage forming processes.