A Knowledge Graph Enhanced Digital Twin Framework for Discrete Industrial Production Systems

With the rapid development of artificial intelligence (AI) technology, the digital twins (DTs) can provide more intelligent and convenient services to industrial systems. Aiming to the design, commission, operation and maintenance of discrete production systems, an overall framework of knowledge graph (KG) enhanced digital twin is proposed in this paper, whose main components include KG of production system, KG based DT models and KG driven DT services. Structure, behavior and performance KGs of production systems are constructed from data and schema layer to describe the specific purposes of DTs. The DT models based on KGs at resource, process, system levels with different disciplines can be generated by using diverse AI methods. The offline and online DT services for different lifecycle of production systems will be provided taking advantage of powerful reasoning and querying abilities based on KG. An example line is given to illustrated the primarily application of the proposed DT framework. The proposed method is expected to achieve the multi-source heterogeneous data fusion, multi-disciplinary model interoperation and multi-lifecycle service integration of discrete production systems DT.