Retrofit building energy performance evaluation using an energy signature-based symbolic hierarchical clustering method

Sebin Choi,Hyunwoo Lim,Jongyeon Lim,Sungmin Yoon

DOI: https://doi.org/10.1016/j.buildenv.2024.111206

IF: 7.093

2024-01-28

Building and Environment

Abstract:Retrofitting existing buildings is crucial for significantly reducing energy consumption in the building sector. The continuous monitoring and evaluation of retrofit building energy efficiency is necessary to maintain optimal energy performance. This study proposes a novel method for evaluating building energy performance that combines energy signature analysis and hierarchical clustering. Symbolic data were defined by k-means using differences in gradients and y-intercepts before and after retrofitting extracted from energy signatures. Hierarchical clustering was then performed using the symbolic datasets. This symbolic hierarchical clustering method enhances the utility of open data and facilitates rapid decision-making. Additionally, it allows for a simple assessment of energy performance at the city scale. Through implementing this approach in 49 retrofitted buildings in Gangwon-do, South Korea, five types of symbolic data were identified (Types 0–4). Using hierarchical clustering, these buildings were clustered into six groups (Clusters 1–6). Type 3, representing ideal retrofitting outcomes, was observed in Clusters 2, 3, and 4 (73.47 % of all buildings). Conversely, Type 4 symbols, indicating a rebound effect, were observed in Cluster 1 and 2 (6.12 %). These findings provide meaningful information after retrofitting at the regional level, contributing to effective building management during retrofitting.

engineering, environmental,construction & building technology, civil