The Zentrum für Sonnenenergie- und Wasserstoff-Forschung Baden-Württemberg (ZSW) has been operating a research production line (FPL) for the near-series production of large-format Li-ion cells at its site in Ulm since 2014.
In the “ZellkoBatt” project funded by the German Federal Ministry of Education and Research (BMBF), the ZSW is expanding the existing plant technology. This also requires a new production control and line system that, in addition to automation, above all takes digitization to a new dimension. By assigning all accruing operating, machine and process data as well as product, quality and maintenace data to each individual component of a battery cell to be individually identified, the ZSW will be able to perform evaluations and analyses for process optimization, e.g. by means of AI applications.
The system is completely modular. Thus, it will be possible to react as quickly and flexibly as possible to new requirements in the future. In the case of technical implementations, the shortest possible downtimes are to be expected.
The main challenges here go far beyond a classic division into SCADA and MES with production data acquisition (PDA). In the latter, orders and possibly operations are specified and reported back. Even a classic machine data collection (MDC) would not meet the requirements, since it only reports the status of a machine (in operation, produced, in malfunction, …), a few process parameters and possible malfunctions. Only in combination with a detailed process data acquisition the requirements of the ZSW are fulfilled.
The real production is mapped as a digital twin with the help of ThingWorx and connected to it with Kepware. The process data is recorded as time series and managed in a corresponding database. Process control, alarm management, asset management and key figure monitoring are also implemented on it. The standard functions such as order, recipe, material and warehouse management are mapped with the standard MES S4P.
This future-oriented architecture also makes the integration of AI algorithms and the use of modern human machine interfaces (HMI) based on AR easy for the future.
