Towards Reproducible Test Annotation for Cyber-Physical Energy Systems using Ontology-driven Dataspaces
Kai Heussen, Jawad Kazmi, Narges Mehran, Artjoms Obushevs, Terence O'Donnell, Thomas I. Strasser
Reproducibility, traceability, and transparency in testing cyber-physical energy systems are crucial for scientific advancement and cross-laboratory collaboration. Current experimentation and test documentation practices lack formal semantics, making it difficult to reproduce experiments, share data, and apply, for example, the artificial intelligence-driven analysis. A dataspace that relies on structured ontologies aims to address these gaps by providing machine-actionable descriptions. In this work, we outline an ontology-driven approach for reproducibility of cyber-physical energy systems testing and illustrate its applicability through representative cross-laboratory use cases, demonstrating feasibility while identifying remaining semantic and metadata gaps that limit reproducibility. Based on these observations, we propose an open three-viewpoint ontology framework to guide future ontology extensions.
Read on ELI