Non-Intrusive Load Monitoring for Public Buildings Based on ResNet and Transformer

With the increasing severity of global energy issues, public buildings, as a critical component of urban energy consumption, play a significant role in achieving energy management and carbon reduction goals. Non-intrusive load monitoring (NILM) technology, with its advantage of enabling fine-grained energy consumption monitoring without additional hardware retrofitting, has demonstrated broad application prospects in public building energy consumption analysis. However, existing methods face challenges in balancing local feature extraction and global temporal dependency modeling in complex load environments. This paper proposes a load disaggregation model combining ResNet and Transformer, where ResNet is employed to extract local patterns from load signals, and Transformer is used to capture global temporal dependencies. Experimental results show that the proposed model achieves high accuracy and robustness in public building load disaggregation tasks and outperforms existing mainstream models in comparative experiments, providing theoretical support for optimizing public building energy consumption.