Doctor of Philosophy
Sustainable Buildings Research Centre
The operation of buildings and building Heating, Ventilation and Air-conditioning (HVAC) systems may experience faults such as equipment malfunctions, sensor reading faults, inappropriate operating procedures, incorrectly configured control systems, and the degradation of equipment, which may deteriorate building performance. Assessing building performance is therefore essential to understand building performance levels and identify potential operational issues that may influence building energy efficiency and indoor thermal comfort. Thanks to the wide deployment of building automation systems and smart meters, a massive amount of high resolution building operational data can now be easily accessed. This data provides a great opportunity to better understand the characteristics of building energy usage and operational performance. However, without advanced data analytics, the valuable information underneath this massive amount of operational data cannot be fully extracted and used to assess and improve building energy performance.
Data mining is a proven method for extracting valuable and actionable information from the massive data. The aim of this thesis sought to develop data mining based strategies to evaluate the performance of building HVAC systems, individual buildings, and multiple buildings to facilitate the initiatives for building performance enhancement.
Yan, Rui, Data mining based strategies for building performance assessment, Doctor of Philosophy thesis, Sustainable Buildings Research Centre, University of Wollongong, 2017. https://ro.uow.edu.au/theses1/79
Unless otherwise indicated, the views expressed in this thesis are those of the author and do not necessarily represent the views of the University of Wollongong.