Categories:Innovation

An introduction to Digital Twin Definition Language

PostedWillow Team

In the past year, digital twins have gained mainstream traction as a fundamental component of a smart building or infrastructure strategy. However, if they are to truly deliver on the hype and expectations that have built up around them, there must be a consistent and structured approach for the language that underpins their development.

Benefits of Digital Twins Definition Language (DTDL)

  1. DTDL is a flexible language that can cover any vertical or type of model.
  2. It’s the solution for modelling complex relationships in the built world.
  3. It creates a “neural network” data structure.
  4. It enables digital twins to leverage higher order analytics, machine learning and conduct complex computations.
  5. It’s a platform language for B2B integration, for other business functions to integrate with Microsoft services via a DTDL model contract.

The solution for complex relationships: Digital Twins Definition Language

A fundamental challenge with digital twin technology is understanding how to model industry-specific domain knowledge while ensuring different platforms can communicate with one another to make sense of the complex relationships that exist within the physical world.

In the building industry, there have been a number of different initiatives that seek to help move towards a common language such as Brick SchemaRealEstateCore or Project Haystack. However, as industry technology evolves, so too must the languages that support their development.

The team at Willow have been working closely with Microsoft to apply a new language, Digital Twins Definition Language (DTDL), as the fundamental modelling language for WillowTwinTM.

The digital twin language takes human knowledge of the physical world and translates it into a common data ontology and governance solution. For example, in building operations, we need Willow’s software to understand that when we say ‘AC Unit’ we mean a specific classification of  HVAC equipment and all the relationships this unit has with other parts of a building.

In any digital twin, you need to be able to describe the elements in the building and the relationship that they have with each other. You want to be able to define the digital entities that represent the people, places and things in your physical environment,” said Scott.

DTDL will enable the complex relationships between equipment, physical context and live data to be mapped and effectively translated into Willow’s software. The work being done by Willow aims to move the digital twin industry forward at a rapid pace and is gaining adoption by other solution providers. Our end goal for digital twin technology is simple — achieving interoperability and scalability.

A new neural network

We’re solving the problem of how to map complex relationships within the physical world and accurately translate those into the digital world,” said Willow’s Chief Product Officer, Dale Brett who, in conjunction with Willow’s VP of Engineering Rani Adam, has led the effort to integrate DTDL into WillowTwinTM software. The team have been working on what they call a “knowledge graph” model, pictured below, to visually represent this language.

Comparatively, DTDL offers a “neural network” of sorts to map the complex relationships between buildings or infrastructure, their equipment, space and action.

It’s similar to a neural network, a term used in the field of AI which mimics the connections in the human brain,” Dale explains. “Up until now, we’ve had a standard hierarchical tree structure with a parent/child relationship, where for example an instance of an HVAC unit has a parent-child relationship with a specific pump. But what if instead, we could look at it like a complex system of interconnecting relationships, and through these data connections we can identify the specific root cause of the problem. Applying AI predictions, we can fix things before they break. We are experimenting with a new way to navigate the knowledge graph of DTDL, which feels like flying through space or the human mind,” he said.

Rani is excited by the DTDL implementation, “as it has opened many new ideas we want to explore and implement into the WillowTwinTM platform, for example the ability to search entire digital twin graphs rather than a restrictive traversal search.”

How does this impact our customers?

The development on DTDL at Willow is a huge leap for the 1) scalability and 2) flexibility of our software. In a traditional data structure, one of these is typically sacrificed at the expense of improving the other. The adoption of this universal language will make the technology more adaptable and compatible with other platforms and IoT devices that are already in a building or piece of infrastructure as well as opening up the technology to future innovation.

Without a universal language, the digital twins are facing a challenge to reach Industry 4.0 initiatives such as machine learning. These technologies all require a common metadata model which is enabled by DTDL. We’ve seen this with many different ‘dialects’ in other initiatives to standardise the industry,” said Rick Szcodronski, Willow’s Marketplace Manager, who in the past two months alone, has led the development of more than 300 DTDL models and 1200 entities to extend the work of previous schemas.

The team at Willow are at the cutting edge of technology. Adopting the new language is the first step towards machine learning, AI and enhancing building’s operational performance.

We don’t know what we don’t know yet,” said Rick. “The possibilities are beyond our imagination right now. But we are heading in the right direction.

More articles you might like

  1. Join our mailing list

    Stay up to date with all the latest news and updates from Willow.