Credible demand response capacity evaluation for building HVAC systems based on grey-box models

• Published in: Applied energy Vol. 395; p. 126144
• Main Authors: Jiang, Siyu, Hui, Hongxun, Song, Yonghua
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.10.2025
• Subjects: Credible demand response capacity; Demand response; HVAC systems; Probabilistic estimation; Demand response; Credible demand response capacity; HVAC systems; Probabilistic estimation
• ISSN: 0306-2619
• DOI: 10.1016/j.apenergy.2025.126144

Demand response (DR) has been promising in recent years for maintaining the balance between power supply and demand in the power system. Evaluating the DR capacity is significant for the stable operation of the power system and for improving end-user participation. Heating, ventilation, and air conditioning (HVAC) systems account for about 40 % of the total power demand and have enormous potential. However, the power consumption of HVAC systems is subject to various uncertainties, making it difficult to evaluate the range of their DR capacity credibly. To solve this issue, this paper proposes a credible DR capacity evaluation framework based on grey-box models. This framework utilizes a probabilistic model to estimate HVAC consumption baseline intervals and leverages an adaptive equivalent thermal parameter model to derive credible DR capacity intervals. The intervals can reflect the non-linear relationship between multiple uncertainties and the DR capacity. A probabilistic model is proposed by combining a temporal convolutional network and ensemble conformalized quantile regression to estimate the baseline intervals. Additionally, an adaptive equivalent thermal parameter model is adapted to quantify the DR capacity under different regulation levels and different confidence levels. Finally, the effectiveness of the proposed framework in evaluating credible DR capacity is verified using realistic scenarios in Macao. •Proposing a credible DR capacity evaluation framework.•Developing a probabilistic model to estimate HVAC consumption baseline intervals.•Adapting an equivalent thermal parameter model to quantify DR capacity.•Capturing the non-linear relationship between uncertainties and DR capacity.