Automated battery disassembly: PEM study provides data model

The Chair of Production Engineering of E-Mobility Components (PEM) at RWTH Aachen University has developed a generally applicable data model as the basis for automated battery disassembly. The study summarises product and process definitions that are necessary to enable the disassembly of different battery systems, which has largely been carried out by humans to date, to be carried out by machines in the future.

‘Europe needs a scalable, automated process approach due to the sharp increase in traction battery returns in the coming years – but this requires a lot of product and process-related data,’ says PEM Director Professor Achim Kampker. According to the study, the depth of existing data varies greatly and is therefore not sufficient as a basis. Furthermore, although the European digital battery passport is suitable in principle as a source, the young document in its current form still provides too little data for the concrete implementation of automated disassembly steps.

Analysis with the help of an in-house demonstrator plant: A product and process data structure was therefore created as part of the analysis by deriving the disassembly sequence and depth for a battery pack of an Audi ‘e-tron’. With the data, structures, relationships and definitions identified in this way, a generalised process model was designed and then transferred into a holistic approach. The study utilised a demonstrator system set up as part of the ‘DemoSens’ research project, which consists of two robots, an RGBD camera, a power computer and several end effectors.

As the majority of the data required to automate the dismantling of used batteries is not yet available, further efforts must be made towards overall validation in addition to the RWTH institution’s model approach, which is based on individual automated steps, emphasises PEM expert and study author Domenic Klohs. In addition, the battery passport should be further expanded as a source of information and data, especially as an important step towards establishing homogeneous material flows for recycling. ‘For the further development of the modelling approach, special attention must also be paid to data deviations in the case of damaged batteries, deformations, external influences or changes due to repairs,’ says Klohs. The study has been published on the open access portal MDPI as part of the DemoSens and DemoRec PEM projects funded by the German Federal Ministry of Education and Research.

[DISPLAY_ULTIMATE_SOCIAL_ICONS]