Assessment of Adjustable Potential for Residential Air-Conditioning Load Clusters Based on Physics-Data Hybrid-Driven Approach

Accurately evaluating the demand response potential of residential air conditioning loads in the power grid is crucial for formulating effective load management strategies. This paper proposes a method for assessing the adjustable potential of residential air conditioning load clusters based on a hybrid physical-data-driven approach. Firstly, a data-driven air conditioning load estimation model is utilized to estimate the summer power consumption of residential air conditioning loads. Secondly, a physical aggregation model for residential air conditioning loads is constructed using the Equivalent Thermal Parameters (ETP) model. By minimizing the error between the aggregated power and the estimated power, the particle swarm optimization algorithm determines the equivalent parameters of the physical aggregation model, including the number of residential air conditioning units, the expected value of temperature settings, and the expected value of equivalent thermal resistance. Finally, taking into account user comfort and willingness, the adjustable potential of the residential air conditioning load cluster is evaluated. Case analysis demonstrates that the proposed method yields a strong correlation between the estimated power and temperature of residential air conditioning loads in summer, and the evaluation results for the adjustable potential of these loads are reasonable.