Decarbonizing the building sector by improving the energy efficiency of the existing building stock through retrofits in a targeted and efficient way remains challenging. This is because, as of now, the energy efficiency of buildings is generally determined by on-site visits of certified energy auditors which makes the process slow, costly, and geographically incomplete. In order to accelerate the identification of promising retrofit targets on a large scale, we propose to estimate building energy efficiency from remotely sensed data sources only. To do so, we collect street view, aerial view, footprint, and satellite-borne land surface temperature (LST) data for almost 40,000 buildings across four diverse geographies in the United Kingdom. After training multiple end-to-end deep learning models on the fused input data in order to classify buildings as energy efficient (EU rating A-D) or inefficient (EU rating E-G), we analyze the best performing models quantitatively as well as qualitatively. Lastly, we extend our analysis by studying the predictive power of each data source in an ablation study. We find that the best end-to-end deep learning model achieves a macro-averaged F1-score of 62.06% and outperforms the k-NN and SVM-based baseline models by 5.62 to 11.47 percentage points, respectively. As such, this work shows the potential and complementary nature of remotely sensed data in predicting energy efficiency and opens up new opportunities for future work to integrate additional data sources.
翻译:通过有针对性和高效率的改造提高现有建筑库存的能源效率,从而实现建筑部门脱碳,这仍然是个挑战。这是因为,到目前为止,建筑物的能源效率一般是由经认证的能源审计员的现场视察决定的,这导致这一过程缓慢、昂贵和地理上不完全。为了加快大规模确定有希望的改造目标,我们提议仅从遥感数据源中估算能源效率的建设;为了这样做,我们收集联合王国四个不同地理分布的近40,000座建筑物的街道视图、航空视图、足迹和卫星载地表温度(LST)数据。在培训了多个端对端的深层次学习模型之后,将建筑物的能效数据数据数据数据数据数据数据数据数据数据数据数据数据数据数据数据数据数据数据进行了宏观平均、62.06%和遥感数据数据数据数据数据预测,从而分别将基于F1和SM47%的模型数据数据数据数据预测,将这一数据数据数据数据数据基数和数据基数(以62.06%和5M-47 %的数据基数据基值为基础,将数据基数、5.06和遥感数据基数点的今后数据基数数据基数定位数据基数定位数据基数和数据基数定位数据基数和数据基数(分别为62.0数据基数)分别将数据基数、5.06-5-5-0和遥感数据基数的计算出一个潜在数据基数点和遥感数据基数。