Collecting Data from Building Automation Systems

Approximately 40% of the total energy consumption within the EU is due to buildings, and similar numbers can be found in the US. It is also estimated, by the European Commission, that 75% of the building stock are energy inefficient. If the principal inefficiencies were easily identifiable, a facility manager could focus their resources to make the buildings more efficient, which would lead to both cost savings for the facility owners and decrease the building’s ecological footprint.

In building automation systems today, data is already being collected every second, but due to the lack of standardization for describing data, having access to data is not the same as being able to make use of it. Currently, the heterogeneity makes it very costly to gather data from multiple buildings, thus making it difficult to find the big picture.

Facility managers cannot fix what they cannot see; thus it is important to facilitate the visualization of the data collected from all of the different building automation systems. Doing this has the potential for great benefits with regards to both sustainability and economy. In this thesis, the author’s goal is to propose a sustainable, cost and time effective data integration strategy for facility managers who wish to gain greater insight into their buildings’ efficiency. The study begins with a literature study to find previous and on-going attempts to solve this problem. Some initiatives for standardization of semantic models were found. Two of these models, Brick and Project Haystack, were chosen. One building and automation control systems was tested in a pilot case study, to test appropriateness of a solution.

The key results from this thesis project show that data from building automation systems, can be integrated into an analysis platform, and an extract, transform, and load process for accomplishing this is presented. How time efficiently the data can be tagged and transformed into a common format is very dependent upon the current control system’s data storage format and whether information about its structure is adequate. It is also noted that there is no guarantee that facility managers have access to the database, or information about how that is structured, then other techniques can be used such as BACnet/IP, or Open Communication Communications (OPC) Unified Architecture.

Keywords: Building Automation System, Smart buildings, Big data, Building Management System, Project Haystack