Exploratory Study Towards Streamlining the Identification of Sensor Locations Within a Facility

Abstract

Significant efforts are being made in the improvement of building automation and control systems in order to optimize the performance of buildings (e.g. reduction of energy consumption). As sensor networks in buildings increase, the complexity of managing them also increases. For instance, the generation and maintenance of metadata about sensors, such as their location within a building, currently requires significant manual labor. The research described in this paper explores the relationship between different HVAC system sensor measurements and physical characteristics of spaces, and its potential application in streamlining the identification of sensor location within a facility. The energy contained in the conditioned air delivered to each room is presented as a characteristic feature in order to understand the differences between rooms. Being able to understand the relationships between different measurement types and building characteristics is fundamental in achieving an automatic mapping of sensors in buildings, and this paper describes initial observations and results towards this goal.

Unified architecture for data-driven metadata tagging of building automation systems

“…Koc et al [19] examine the effects of data size and whether a nonlinear correlation coefficient is more effective at determining sensor locations. Akinci et al [20] identify the specific rooms in which sensors are located by correlating expected and actual sensor readings with the known HVAC energy input into a room. By identifying a single sensor in a room, this process can be identified with previous work to map all sensor locations in a building.…”