You may think your building is smart, but is it autonomous?
If there a difference?
Johnson Controls believes there is indeed a difference and plans to soon roll out what it calls an equipment performance advisor for building managers that more precisely controls HVAC and other systems in a building complex based on real-time information.
The ultimate goal is to lower energy costs for buildings, which globally create up to nearly 40 percent of carbon emissions when the creation of materials like concrete are included.
Vijay Sankaran, vice president and CTO at Johnson Controls, told Fierce Electronics at Sensors Converge 2024 that the company has provisioned a system for autonomous analytics in buildings that will compare pre-determined settings for equipment settings with weather and other factors. The results will be displayed in a dashboard for building managers who can then approve or override recommended actions.
In one example, air conditioning could be set to a range of 70-80 degrees F, and the analytics engine might also detect a cold front arriving in the region of the affected buildings and suggest it would be better to run air conditioning at a higher value in the range. Also, if a building is expecting a large crowd of people, the real time autonomous recommendation engine could advise keeping the temperature at the lower range to keep the building cooler.
Seemingly insignificant adjustments to a thermostat for a large building complex could prove valuable in overall energy savings in a year, Sankaran said.
In a keynote address at Sensor Converge, Sankaran said net zero compliance goals have forced a reckoning for building owners and businesses of all types. “Most don’t even know their current [energy] consumption,” he said, noting that generative AI and its energy consumption demands have set Microsoft 30% behind its net zero goals.
With energy use predictions and adjustments to HVAC systems, “a customer in the Middle East might see that small variations could make a huge difference in net zero,” Sankaran said.
Some city governments are setting requirements for energy consumption and carbon emissions, which could be translated via text with a large language model for use in predictive AI and machine learning. A building owner using generative AI could quickly learn that a particular building is expected to exceed on an energy consumption goal for a month, with GenAI then recommending conservation measures while also summarizing the penalties that could result if the energy goals are not met, Sankaran said.
With GenAI, a building manager could ask questions via text or even voice such as, “Which office has the highest use?’ in straightforward language, Sankaran said. This capability is “the tip of the iceberg,” he added.
“We’ve building AI as a service platform. Risk management is something we take very seriously…Digitization and AI are the future and everybody should be thinking about AI and generative Ai what problems you are trying to solve,” Sankaran told the audience of engineers and designers.
“It’s a very exciting time for all of us in IoT…It’s going to be a wild 10 years. Buckle up!”
RELATED: Johnson Controls views AI, sensors at center of smart building push