Live BACnet data for a data center digital twin, from CRAC units, power systems and sensors

The challenge

A digital twin is only as good as the live data behind it. In a data center, the cooling units, power systems and environmental sensors already report over BACnet, from many manufacturers.

The software company needed its platform to act as a BACnet client: discover what is on the site’s network, read the values it needs, and keep reading them, without writing a protocol implementation of its own.

Why the CAS BACnet Stack

• The CAS BACnet Stack is a client as well as a server. A platform discovers devices with Who-Is, reads with ReadProperty and ReadPropertyMultiple, and subscribes to changes of value.
• It decodes every message to JSON or XML, which feeds a software platform’s data pipeline directly.
• Adapters for C#, Java, Python, Node.js, Go and Rust mean the platform calls the stack from the language it is already written in.

The solution

The platform embeds the CAS BACnet Stack as a BACnet client. It finds the devices on the site’s network, reads the live values of the CRAC units, power systems and sensors, and hands them to the digital twin.

The twin compares the live data with its simulated data, and the difference drives the platform’s predictive analytics.

Results

• The digital twin collects live data from CRAC units, power systems and sensors over BACnet.
• Live and simulated data are compared for predictive analytics.