Ontology-Based Semantic Translation for Interoperability across Dataspaces

: Classic Approach

Abstract

The expansion and development of dataspaces in various domains around the world have introduced a new paradigm of achieving interoperability. Interoperability has various forms, including legal, semantic, organizational, and technical. This thesis focuses on attaining Semantic Interoperability between dataspaces in the Energy Sector. The increase in the adoption of microgrids, smart grids, electric vehicles, home automation, Vehicle-to-Grid(V2G) technologies, coupled with the shift towards renewable energy sources, has created new challenges in maintaining stability between the grid and end consumers. The amount of data collected, managed, and analyzed has increased significantly. Since these data are often transferred between different entities using different ontologies and standards, accurate and meaningful translation is critical. In simple terms, the data intended to be transferred must be understood by the receiving system without alteration, without any loss of clarity, concept, meaning or understanding of the vocabulary used by other systems. This thesis work addresses the need by developing a translation service which facilitates semantic interoperability between two existing ontologies: CIM ESMP and SAREF. This model reads the input XML data, which is structured according to CIM ESMP ontology, and translates it into equivalent SAREF ontology using pre-defined mappings between ontology classes, attributes, and vocabularies. To implement this, the thesis explores the structure of various European dataspaces and examines the role of semantic artifacts and ontologies. The core of the work involves the construction of mappings between ontologies and their application using a rule-based mapping approach. Although this implementation is based on rule-based and manual matching, it lays the foundation for developing fully automated translation systems using advanced machine learning techniques. This work contributes to the field through a scalable, semiautomated approach to achieve Semantic translation, allowing interoperability across dataspaces in the energy sector.

Date of Award	2025
Original language	English
Supervisor	Thomas Ziebermayr (Supervisor) & Shievam Kashyap (Supervisor)