CityTFT: Temporal Fusion Transformer for Urban Building Energy Modeling
This work addresses the need for efficient energy demand modeling in urban environments, offering a faster alternative to traditional methods, though it appears incremental as it adapts an existing TFT framework.
The paper tackled the problem of time-consuming physics-based Urban Building Energy Modeling (UBEM) by proposing CityTFT, a data-driven framework that achieved an F1 score of 99.98% for predicting heating and cooling triggers and an RMSE of 13.57 kWh for loads in unseen climate dynamics.
Urban Building Energy Modeling (UBEM) is an emerging method to investigate urban design and energy systems against the increasing energy demand at urban and neighborhood levels. However, current UBEM methods are mostly physic-based and time-consuming in multiple climate change scenarios. This work proposes CityTFT, a data-driven UBEM framework, to accurately model the energy demands in urban environments. With the empowerment of the underlying TFT framework and an augmented loss function, CityTFT could predict heating and cooling triggers in unseen climate dynamics with an F1 score of 99.98 \% while RMSE of loads of 13.57 kWh.