Abstract
The Internet of Things (IoT) is creating a renaissance in data analytics by connecting devices in a wide variety of industries, as well as in everyday life. However, data collected from these new and disparate sources are typically non-structured and non-uniform, which has become a significant barrier to developing scalable and repeatable analytics. The more than 5.6 million of commercial buildings in the United States of America are an illustration of this phenomena. Indeed, in the past decade, commercial, industrial, and government buildings have been outfitted with sensors and controllers, fully embracing the modern age. Successful application of analytics to systems controlling indoor building environment is providing an opportunity to realize substantial and widespread gains in energy efficiency and improved comfort as well as reductions in operating costs long after the building's commissioning. As such, Building Automation Systems (BASs), controlling heating, ventilation and air conditioning (HVAC) equipment, provide access to some level of information, often in the form of single and overlaid trends, meter data, and key performance indicators on a dashboard. The BAS, as well as analogous approaches, enables manual analysis by trained personnel, but it is cumbersome and cannot be easily scaled. In addition, the vast diversity of equipment, controllers, and configurations among buildings as well as the heterogeneity of components within single systems have further hindered traditional automated systematic analysis. In this work, a methodology for structuring and uniformizing data applied to building environment control systems is demonstrated. The result is a system that enables scalable and repeatable advanced analytics on a macro scale, abstracting the focus from a specific building and configuration, that is likely unique, to systems of multiple equipment that is generalizable to any building.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have