Learning the Reluctance of Demand-Side Resources From Equilibrium in Price-Based Demand Response

The reluctance of demand-side resources (DSRs) in demand response (DR) is not directly accessible, yet, significantly impacts the DR performance. This work aims to estimate DR reluctance from observed DR equilibrium outcomes by inverse variational inequality (VI). First, the definition and properties of DR reluctance are introduced. Then, the equivalent generalized Nash equilibrium condition in DR is derived by strong duality. Based on inverse VI technique, a data-driven linear-programming (LP) for learning DR reluctance is formulated. Finally, the proposed method is validated through a toy example and larger-scale cases, showing its effectiveness and scalability.