Describe and query semantic building digital twin data in temporal Knowledge Graphs

Abstract

With the emerging application and development of digital twin technologies in the Architecture, Engineering, and Construction (AEC) sector, effective monitoring and prediction of the built environment is an important step toward energy transformation. The temporal properties are always implied in the events, either explicitly or implicitly. However, if the temporal evolution data between multiple stakeholders is not properly documented and structured, this valuable data will not be transformed into meaningful actions. As a result, the AEC collaborative work requires a structured model to integrate heterogeneous information and fully document state changes over the process of multiple iterations. In this paper, we introduce a workflow on a Temporal Knowledge Graph (TKG) in order to link observation data and the building environment context in a generic inductive framework. The temporal properties of events are documented in a graph and can be queried through dedicated SPARQL queries. The main contribution of this research is an approach to track the temporal information of linked building data, in order to derive additional information for digital twin scenarios.