Identification of machine parameters in battery cell production: A comparison of different methods and approaches for mapping parameters

Optimization of battery cell production processes is increasingly based on digital use cases that can reduce costs and improve quality. However, these requires accurate and semantically understandable data that is human- and machine-readable and stored in standardized formats. Machine manufacturers often use unique or proprietary parameter standards and naming conventions, making data integration difficult. Linking machine data to semantic database standards requires expert knowledge and significant manual effort. Automating this process could reduce the workload enormously. This paper evaluates approaches for automating the mapping and comparison of machine parameter sets within the context of battery cell production. Using real production data and semantic modelling from multiple partners, different methods were applied to link individual parameters. The outcomes were assessed with specific metrics, revealing that Large Language Models (LLMs) show significant promise due to their flexibility and effectiveness in handling complex semantic comparisons.