Heating, Ventilation, and Air Conditioning (HVAC) systems are significant consumers of energy in buildings. Existing HVAC systems often operate on predetermined schedules, with the systems being activated or deactivated based on fixed set points, irrespective of room occupancy. This approach may often prove to be inefficient. By adjusting the HVAC settings based on occupancy, the energy consumption in buildings can be optimized, reducing the overall energy costs and environmental impact. Today, HVAC systems cannot detect the occupancy, and thus cannot minimize unnecessary energy usage in buildings.
The NEPHELE project enables real-time intelligence and observability directly on constrained IoT devices (e.g., actuators, sensors, or thermostats) via Tiny Machine Learning and Complex Event Processing. For example, a thermostat can be extended with the intelligence from NEPHELE. The extension enables the device to learn on-the-fly occupancy patterns in buildings, adapting them throughout the day and week, and optimizing HVAC schedules. This reduces the carbon footprint of buildings and the overall energy costs. Moreover, NEPHELE solution enhances the openness and interoperability of building devices thanks to the W3C Web of Things standard.
Partner involved: SIEMENS