A novel methodology for knowledge discovery through mining associations between building operational data

Nowadays, vast amounts of data on building operation and management have been collected and stored. However, the data is rarely translated into useful knowledge about building energy performance improvement, due mainly to its extreme complexity and a lack of effective data analysis techniques. This paper reports the development of a new methodology for examining all associations and correlations between building operational data, thereby discovering useful knowledge about energy conservation. The method is based on a basic data mining technique (association rule mining). To take full advantage of building operational data, both daily and annual time periods should be mined. Moreover, data from two different years should be mined, and the obtained associations and correlations in the two years should be compared. In order to demonstrate the applicability of the proposed method, the method was applied to the operational data of the air-conditioning system in a building located in Montreal. The results show energy waste in the air-conditioning system as well as equipment faults. A low/no cost strategy for saving energy in the system operation was also proposed. The results obtained could help to better understand building operation and provide opportunities for energy conservation. (C) 2011 Elsevier B.V. All rights reserved.