Transfer learning for thermal comfort prediction in multiple cities

The HVAC (Heating, Ventilation and Air Conditioning) system is an important part of a building, which constitutes up to 40% building energy usage. main purpose HVAC, maintaining appropriate thermal comfort, crucial for the best Additionally, comfort also well-being, health, work productivity. Recently, data-driven models have achieved better performance than traditional knowledge-based methods (e.g. predicted mean vote model). An accurate model requires large amount self-reported data from indoor occupants undoubtedly remains challenge researchers. In this research, we aim address data-shortage problem boost prediction. We utilize sensor multiple cities in same climate zone learn patterns. present transfer learning-based multilayer perceptron (TL-MLP-C*) Extensive experimental results on ASHRAE RP-884, Scales Project Medium US Office datasets show that proposed TL-MLP-C* exceeds state-of-the-art accuracy F1-score.