Leader-Follower Dynamics in P2P Energy Markets: A Bilevel Stochastic Optimization Approach

Peer-to-peer (P2P) energy trading has emerged as a solution for facilitating direct energy transactions between peers within a local energy community. The role of an aggregator in this context is crucial, as it facilitates these transactions and maximizes economic benefits according to current market conditions and future uncertainties. However, there is a gap in effectively coordinating aggregator-peer dynamics, as existing models often fail to consider both market conditions and individual peer preferences simultaneously. To address this, we use a leader-follower bilevel stochastic approach, where the aggregator acts as the leader and the peers as followers. This work implements a bilevel optimization framework, where the aggregator aims to maximize social welfare by optimizing profits from energy transactions at the upper level, and the peers minimize their net energy expenditure while maximizing revenue from selling excess energy, responding to the aggregator’s pricing strategy at the lower level. The stochastic nature of renewable energy is handled by considering different scenarios in the model, capturing uncertainties in solar PV generation, energy demand profiles, and market prices. Furthermore, we compare this bilevel optimization approach with a single-level optimization function to demonstrate the advantages of the bilevel leader-follower dynamic in achieving more efficient and economically viable outcomes. A case study is conducted to demonstrate the model’s applicability and effectiveness in optimizing P2P energy transactions in the real world.