Automatic Knowledge Graph Construction Based on Relational Data of Power Terminal Equipment

Abstract

In the last years, Knowledge Graph (KG) has been widely applied in domain knowledge modeling, and therefore has received more attention in the Ubiquitous Power Internet of Things (UPIoT) that consists of a large number of power terminal equipment and sensing devices. In this paper, we propose an automatic framework to construct power terminal equipment knowledge graph by extracting ontological information and instantiating objects from the relational data of power terminal equipment. First, the ontological concepts and the hierarchy between concepts are built by analyzing the data schemas of different types of devices and their properties. Then the knowledge graph for different types of device objects is constructed by instantiating every object and their binary relations. In addition, we give the relevant algorithms of automatically constructing the complete knowledge graph of power terminal equipment. We argue that the proposed approach will greatly facilitate the information sharing of multi-source heterogeneous data of power terminal equipment on the environment of UPIoT.