Commissioning of Intelligent Production Systems

Sustainability and the reduced use of plastics are trends in the packaging industry. The drawing process holds great potential for the manufacturing of fiber based packages in large quantities and thus the substitution of plastic based packaging.

The project aims for a transferable concept of fast commissioning of intelligent production systems for paper and board products. This supports the enterprises to offer modular and configurable systems. Companies from multiple industries along the value chain such as special and packaging machine engineering, control and automation technology, component manufacturers, but also manufacturers of packaging materials will benefit from the results.

The project uses machine learning algorithms based in synthetic data to develop and implement a technology control system. The data is generated with the help of digital twins of the central process stages and their cross-process coupling. Besides process parameters such as blank holder force and tool temperature, the material moisture content is a key for stable drawing of paperboard materials. In this project, the methods for parameterization and the use of learning controllers are to be developed for a production system for the manufacture of sealed cardboard cups. To validate the methods, a humidification module is added to the production system and then commissioned with the aid of learning controllers. This procedure avoids resource-intensive tests for the IBN of complex production systems.