A Data Structure for Developing Data-Driven Digital Twins

Abstract

Digital twins have the potential to revolutionize the way we design, build and maintain complex systems. They are high-fidelity representations of physical assets in the digital space and thus allow advanced simulations to further optimize the behaviour of the physical twin in the real world. This topic has received a lot of attention in recent years. However, there is still a lack of a well-defined and sufficiently generic data structure for representing data-driven digital twins in the digital space. Indeed, the development of digital twins is often limited to particular use cases. This research proposes a data structure for developing modular digital twins that maintain the coherence between the digital and physical twins. The data structure is based on a hierarchical representation of the digital twin and its components; the proposed data structure uses concepts from distributed systems and object-oriented programming to enable the integration of data from multiple sources. This enables the development of a digital twin instance of the system and facilitates maintaining the coherence between the digital twin and the physical twin. We demonstrate the effectiveness of our approach through a case study involving the digital twin of an industrial robot arm. Our results show that the proposed data structure enables the efficient development of modular digital twins that maintain a high degree of coherence with the physical system.