The term “smart building” has been in use for decades, and since its origin, buildings keep getting “smarter”. This is largely driven by the fact that buildings are continuously generating more and more data. Advancement in IoT is transforming the way we interact with our environment. From lighting, to occupancy, to HVAC, buildings are becoming a source of critical information. Information that can help improve efficiency, energy conservation, and occupant comfort.
However, the sometimes over abundance of this raw data means that the value is often lost. The key to unlocking value is contextualized data. Contextualized data refers to information that is enriched with contextual details providing a greater understanding of the building and environment in which it was created.
Contextualized Data at Work
As an example, in a building HVAC system a chiller may produce 3 values from integrated sensors: 53.6, 42.8, and 42.
A trained eye may be able to determine what each value corresponds to but the average person, and the average smart building system cannot. With a little bit of context, if we see that the unit of these values is ° F we can decipher that the three values are temperatures. Context provides that bit of information that allows users and computers alike to understand more about the data.
In the smart building space, contextualized data models (such as Project Haystack) exist with pre-defined tags to help provide a standard for sharing this contextual information. For example, if in addition to units, the tags ‘water', ‘temp’, ‘leaving’, and ‘sensor were applied to the value of 42.8 we could automatically determine that this value is the temperature of the chilled water leaving the unit. Similarly, 'water temp leaving sp' and 'water temp entering sensor' could tell us that 53.6 and 42 were the return temperature and the supply temperature setpoints respectively.
On top of this, contextualized data can also allow us to “see” through data that this chiller is part of a larger chiller plant, connected to specific cooling towers, pumps, and air handling units.
Benefits of Contextualized Data
Contextualized data paves the way for implementing value added functionality to your buildings such as:
Enhanced Energy Efficiency
Improved User Experience
Predictive Maintenance
Optimized Space Utilization
Optimized Mechanical Control
These benefits sound amazing. What building owner and operator doesn’t want to reduce energy consumption? What Systems Integrator doesn’t want to have a greater understanding of when critical equipment is going to fail? If there is value, and the value is unlocked by tagging and adding context to your data, why aren’t more people doing it? The answer is, it’s complicated.
The Hidden Challenges of Tagging Your Data
I don’t mean the answer itself is complicated. I mean that for most people, the act of going through your BMS and applying tags or codes one line at a time, one point at a time, one piece of equipment at a time is extremely laborious time consuming. Time consuming in an industry where free time is already not abundant.
The Elipsa platform, like many intelligent building systems, is driven off of contextualized data. So, we see the struggle of building owners/operators and SIs first hand wanting to get the value of a product but not having the time to unlock that value.
Elipsa uses contextualized data to feed AI algorithms geared towards providing automated fault detection and diagnostics and energy optimization. So, why not reverse that, and use AI algorithms to automate the process of contextualizing your building data in order to enable the advanced AI value?
Automated Data Tagging with AI
Elipsa ingests raw point names from your BMS either through direct connection or by uploading information that was exported from your BMS. Elipsa's simple interface allows users to provide information about the intertwining systems that exist within their buildings.
As information is read in, the Elipsa AI engine works to analyze point names and make suggestions in the configuration process while automatically applying haystack tags to your point data.
For example, if Elipsa's AI recognizes point CT1 as a Cooling Tower while configuring the equipment ChilledWaterPlant, Elipsa automatically applies tags “coolingTower equip hvac” as well as the “equipRef” of ChilledWaterPlant to CT1 in the data model.
The result is a contextualized data model of your building with no coding or manual tagging required. A data model that can then be used to enable enhanced functionality within Elipsa, utilized to feed your building’s telemetry data to third-party system via MQTT, or exported to import back into your BMS to update the tagging dictionaries.
Elipsa’s AI turns building data into actionable information. Contextualized data is critical to reducing energy consumption and increasing building efficiency, unlock the value with Elipsa.
For more information contact us at info@elipsa.ai or schedule a call