Building energy performance assessment using volatility change based symbolic transformation and hierarchical clustering
RIS ID
122064
Abstract
This paper presents the development of a symbolic transformation based strategy with interpretability and visualisation for building energy performance assessment. The strategy was developed using shape definition language based symbolic transformation and hierarchical clustering. Advanced visualisation techniques including dendrogram, heatmap and calendar view were used to assist in understanding building energy usage behaviours. A comparison of this proposed strategy with a Symbolic Aggregate approXimation (SAX) based strategy was also performed. The performance of the proposed strategy was tested and evaluated using the three-year hourly heating energy and electricity usage data of a higher education building. The result demonstrated that the proposed strategy can identify distinct building energy usage behaviours. The visualisation techniques used also assisted the information discovery process. The discovered information helped to understand building energy usage patterns. The comparison of the proposed strategy with the SAX based strategy showed that the proposed strategy outperformed the SAX based strategy for the case building tested in terms of the variations in building energy usage. This proposed strategy can also be potentially used to evaluate the operational performance of building heating, ventilation and air-conditioning (HVAC) systems.
Publication Details
Ma, Z., Yan, R., Li, K. & Nord, N. (2018). Building energy performance assessment using volatility change based symbolic transformation and hierarchical clustering. Energy and Buildings, 166 284-295.