Buildings are operating nowhere as smart as they could be, if you ask Salt Lake City–based startup PassiveLogic. According to the company’s recently authored Quantum digital twin standard, building control technologies largely remain stuck at a level 1 autonomy, meaning they can adaptively adjust set points of static manual sequences. In fact, level zero technologies, which operate on fixed set points or PID (proportional, integral, and derivative) controls, remain prevalent.
PassiveLogic says its novel platform can take buildings to full autonomy—level 5—and beyond. Its development of physics-based “deep digital twins” can factor in a theoretical n number of systems to determine building operations across n dimensions by utilizing artificial intelligence.
“It’s not been until relatively recently that we’ve had AI powerful enough to operate at the edge,” says Joseph Riddle, who oversees the company’s brand communications. “Buildings are all unique. Some have hundreds of thousands of I/O points, so it’s a huge robotics challenge. They’re giant robots who don’t move.”
To physically embed an independent, analytical, and responsive brain capable of crafting custom operational game plans into any building, the approximately 100-person company is developing a product portfolio that comprises everything from modeling platforms to AI-powered controllers to IoT sensors operating at the edge. A high-level overview on how the platform works follows.
PassiveLogic’s Quantum Creator software allows users—engineers, manufacturers—to create digital, physics-based twins of building equipment and systems: HVAC, electrical, plumbing, lighting. The company has already partnered with several manufacturers, including Belimo HVAC systems and Bradford White water heaters, to create physics-based digital twins of their equipment for inclusion in a product library, Riddle says.
To create the base building model housing the equipment, users—architects and owners—can import models from existing software platforms, such as Autodesk Revit, or license PassiveLogic’s Autonomy Studio. For retrofit projects, the company debuted a LIDAR-scanning app to expedite the modeling of existing buildings at last month’s AHR Expo. “If you’re a building designer, facilities manager, or commissioning engineer,” Riddle says, “you can go into our Autonomy System software and, with drag-and-drop, assemble a 3D model of your building with the actual components that reside in that building and with their physics understood.”
The digital building model, complete with its assemblies, then syncs with PassiveLogic’s Hive controller, an AI-powered engine installed on-site to ensure autonomy off the cloud. All I/O values come together via snap-in-place Cell modules into the 12-core processor Hive controller; multiple nodes can be daisy chained depending on the number of control points or user accessibility desired. Real-time data on building conditions and occupancy is then collected through sensors. PassiveLogic has partnered Middleton, Wis.–based Automation Components, Inc., to create wired and wireless Swarm and Swarm Nano sensors that monitor eight parameters of human comfort: air temperature, radiant temperature, atmospheric pressure, sound pressure, occupancy, indoor air quality, carbon dioxide concentration, and light intensity.
Digital twins today are largely semantic, offering visual and descriptive building models available on the cloud. Users can manually tag equipment in a digital model—for example, a chiller or boiler—but the objects don’t contain operational information or an awareness of what they are; they might be able to receive and react prescriptively to data from sensors and from external sources, such as weather. At a minimum, owners can use semantic digital twins to locate building components hidden behind finishes and access associated operations and maintenance schedules.
Progressively sophisticated digital twins acquire ontology—or an understanding of what building systems should do—to assess whole-building operations, predict maintenance needs, run performance simulations, all the way to what PassiveLogic says its deep digital twins can achieve: complete, autonomous, on-site control of building systems that adapts in real-time and learns continuously from information received at the edge and from its own understanding of the building in use.
In fact, Riddle says, PassiveLogic is eyeing a future in which buildings can self-guide the installation of their systems and even self-commission—what the company respectively classifies as level 6 and 7 autonomy in its Quantum standard, which it developed in partnership with the U.S. Department of Energy, the Pacific Northwest National Laboratory, and global asset manager Brookfield. “The open-source language will be interoperable with every other language in the building space: Haystack, Brick, BACnet, IFC,” Riddle say. “Any language in which a building’s data is currently held that is proprietary.” With systems speaking a common language—Quantum—buildings can become platforms for developers to create apps, similar to how smartphones have become vehicles for one’s individual communications and informational needs.Going further, Riddle says the PassiveLogic platform enables a building to reach the ultimate level of autonomy—level 8—in which buildings can interact and respond to each other in peer-to-peer interactions.
Early results using PassiveLogic’s AI-powered platform look promising. Compared to the DOE’s EnergyPlus energy modeling tool, Riddle says, “our AI is literally 10 million times faster. We can run the same simulation they do in a week in two minutes because we’re incorporating all the physics into the digital twin.”
Additionally, Riddle says ongoing beta testing of PassiveLogic in about 20 residential buildings in the Salt Lake City area have resulted in an average of “30% energy savings … without any manipulation of the exterior envelope” and 90% savings in time spent installing, programming, and commissioning building systems.
For its commercial beta, expected to begin in the second quarter of this year, PassiveLogic has partnered with Brookfield to install its technology on its properties, Riddle says: “They see PassiveLogic as their only path for meeting their [commitment to be net zero by 2050] that they made at COP26.” PassiveLogic also won a grant from DOE to retrofit municipal buildings in South El Monte, Calif.
One potential slowdown to PassiveLogic’s current momentum: Due to challenges in the current supply chain, the company pushed back expectations for a full market release of its product catalog to year end.