Answers are grounded in this paper and cite their sources.
Proposed model reduces default behaviors in demand response, suggesting improved trading reliability.
The participation of distributed energy resources in demand response enriches the manners of power grid dispatching. However, due to the profit‐seeking behaviors of some demand response suppliers, the cleared trading volumes may not be executed as expected. Therefore, this paper proposes a deep learning‐based credit scoring method for reducing profit‐seeking default behaviors in demand response. First, a modified demand response model is constructed considering the default probability of DRSs to simulate the interaction between the distribution network trading coordinator and demand response suppliers. Then, to evaluate the default probability, a transaction credit and market value‐based default probability evaluation model is proposed. Because of the complexity of data acquisition, the credit evaluation may be biased. Thus, a deep learning‐based credit scoring method is proposed for objectively quantifying credit of the transaction stakeholders, which can be efficiently applied in a variety of complex data types. Finally, simulations on an urban area with an industrial zone, two commercial zones, and a residential zone are shown to illustrate the validity of the proposed method.
No takes yet. Share an insight, caveat, or question.
Ge Liu (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: