Probabilistic Assessment and Interaction for Adjustable Potential of Thermostatically Controlled Loads Considering User Uncertainty

Accurate evaluation on the adjustable potential of air conditioner thermostatically controlled load cluster and its utilization can mitigate fluctuations in new energy such as photovoltaics, and it is of great significance in maintaining the safe and stable operation of power grids. Aiming to address the challenges related to model parameter generalization, physical concept weakening and poor robustness of data-driven methods, a probabilistic assessment method is proposed for the adjustable potential of thermostatically controlled load cluster, taking uncertainties into account. Firstly, the thermostatically controlled load cluster modeling based on hybrid model and data is utilized to track the actual load curve of the thermostatically controlled cluster in a day. Then, the robustness of the proposed method is validated based on KL divergence, considering the unmeasured building parameters and the uncertain user consumption behavior. Finally, the probabilistic assessment and interaction mechanism of the response capacity of the air conditioner load cluster are established based on the hybrid-driven model, taking a 10kV feeder line in Shandong Province as a practical instance. The results demonstrate the effectiveness of the proposed method in mitigating the power flow backflow and realizing the complete local consumption of distributed PV.