J-Park Simulator: An ontology-based platform for cross-domain scenarios in process industry

Abstract

The J-Park Simulator (JPS) acts as a continuously growing platform for integrating real-time data, knowledge, models, and tools related to process industry. It aims at simulation and optimization in cross-domain and multi-level scenarios and relies heavily on ontologies and semantic technologies. In this paper, we demonstrate the interoperability between different applications in JPS, introduce new domain ontologies into the JPS, and integrate live data. For this, we utilize a knowledge graph to store and link semantically described data and models and create agents wrapping the applications and updating the data in the knowledge graph dynamically. We present a comprehensive industrial air pollution scenario, which has been implemented as part of the JPS, to show how knowledge graphs and modular domain ontologies support the interoperability between agents. We show that the architecture of JPS increases the interoperability and flexibility in cross-domain scenarios and conclude that the potential of ontologies outweighs additional wrapping efforts.