BuildSTG: A Multi-building Energy Load Forecasting Method using Spatio-Temporal Graph Neural Network
This work addresses energy load prediction for buildings with spatial dependencies, offering an incremental improvement through graph-based modeling.
The paper tackled multi-building energy load forecasting by proposing a spatio-temporal graph neural network method, which achieved superior performance over baselines like XGBoost and GRU on the Building Data Genome Project 2 dataset, demonstrating robustness and interpretability.
Due to the extensive availability of operation data, data-driven methods show strong capabilities in predicting building energy loads. Buildings with similar features often share energy patterns, reflected by spatial dependencies in their operational data, which conventional prediction methods struggle to capture. To overcome this, we propose a multi-building prediction approach using spatio-temporal graph neural networks, comprising graph representation, graph learning, and interpretation. First, a graph is built based on building characteristics and environmental factors. Next, a multi-level graph convolutional architecture with attention is developed for energy prediction. Lastly, a method interpreting the optimized graph structure is introduced. Experiments on the Building Data Genome Project 2 dataset confirm superior performance over baselines such as XGBoost, SVR, FCNN, GRU, and Naive, highlighting the method's robustness, generalization, and interpretability in capturing meaningful building similarities and spatial relationships.