Methodology for Quantifying Building Demand Response Potential for Market Access Judgments

Market access judgment is a critical component of demand response (DR), employed for the early selection of target users and the reliability review of declared information. This necessitates the grid's ability to rapidly quantify the demand response potential (DRP) of different building. However, traditional approaches often develop potential quantification models tailored to individual objects, making it challenging to meet the potential quantification requirements of a large number of diverse buildings. To enhance model applicability, this paper proposes a data-driven method for quantifying the buildings DRP. Firstly, the input and output variables of the model are determined. Subsequently, a toolchain consisting of EnergyPlus, JEPlus, and Eppy is employed to construct a training dataset, and a backpropagation neural network (BPNN) model is trained based on this dataset to obtain a DRP quantification model applicable to a specific class of buildings. The effectiveness of the method is validated using a typical office building in Nanjing, with global temperature adjustment selected as the peak-shaving measure. According to the computational results, the developed potential quantification model achieves an MSE of 0.0096 and an R
²
value of 0.9, demonstrating a good fitting performance. The proposed method considers the influence of different building scales, disturbance levels, meteorological conditions, and response periods on the building DRP.