Digital Twin data eXchange Standard

Open multi-domain standard describing principles of digital twin data exchange.

Motivation

Missing standardization of data exchange protocols for data managed by digital twins was the gap identified during the EU co-funded research projects PLEIADES and DORADO. Therefore, the initiative to develop this DTXS standard has been started.

This standard is developed in an international cooperation of universities, research institutions, large enterprises and SMEs, as well. For the full list of partners involved in the development of this standard, see this list of partners.

Originally, DTXS has been developed as a single-purpose data exchange standard for development of a digital platform to support nuclear decommissioning planning. However, the generic architecture of the standard enables its extension to cover any other industry domain.

For more information, go to section About.

Main goal

Simply put, a digital twin, as a virtual representation of the physical asset, needs to collect, process, manage, analyse and monitor data related to the physical asset. This data can have different types (e.g., 'as is' data or historical data), different formats (e.g., structured or non-structured), different priorities (e.g., critical or non-critical), different nature (e.g., static or real-time) or different sources (e.g., IoT sensor or engineering process).

Additionally, the data is almost always created by technologies from different vendors which makes the diversity even more complex.

Diversity of these aspects makes it very challenging and time-consuming to centralize the management of the data and to connect a new technology to already running digital twin.

To overcome this complexity, the goal of this initiative is to describe a generic standard for storing any kind of digital twin data (DTD) and providing it in a standardized way to any connected technology (an 'edge' or 'node').

The architecture and basic concepts

The architecture of the standard is based around the central data storage with a secure API, providing rich set of features to store, retrieve and find relevant data. As illustrated on the diagram below, the architecture works with following concepts:

• common domain-related ontology described using JSON-LD, OWL, SKOS or RDF,
• records for the management of structured data,
• documents and folders for the management of non-structured data,
• streams for the management of real-time data,
• classification of all data based on the specified ontology,
• unique identification for each record, document or stream,
• versioning and confidentiality setup for each record, document or stream,
• REST API backed up with a secure authentication,
• support for 3D visualisation,
• data centralization using client-server architecture](api),
• edge computing represented by connected nodes, and

Figure: Illustration of the DTXS architecture concept

Ontology

Very important aspect of the DTXS standard is the domain-specific ontology which describes relationships among terms and verbal concepts used in the certain domain (or, to be more specific, industry domain). The ontology of the specific domain is the base for development of the glossary in that domain, which leads to the development of domain-specific language (or jargon). This further helps with the development of classes used for the classification of records, documents or streams and their properties.

Following these steps, the classification of records stored at the central server can be easily understood by technology experts in the domain which simplifies development of clients (a.k.a. edges or nodes) connected to the central server.

More details about utilization of the ontology in DTXS can be found here.

Multi-domain

Originally, the concept of the DTXS has been developed as the data exchange standard during the development of the digital platform to support nuclear decommissioning planning. However, it is open enough to cover any industry domain.