A pricing strategy based on bi-level stochastic optimization for virtual power plant trading in multi-market: Energy, ancillary services and carbon trading market

Virtual power plants (VPPs) has been considered as an effective approach to manage internal prosumers participating in market transactions with a jointly clearing of energy markets (EM), reserve ancillary services markets (ASM), and carbon trading markets (CTM). Thus, how to address the problem of energy management for VPP in coupling multiple market trading mechanisms has determined as a technical point. In this paper, a bi-level Stackelberg game pricing strategy for VPP trading on multi-market is devised to incentivize its internal units, especially focusing on electric vehicles (EVs), through distinct price signals. An accurate dispatchable power model for EVs is established based on travel patterns and responsiveness. With the goal of maximizing profits, VPP determines trading prices for heterogeneous prosumers, while prosumers respond consistently with price signals to minimize respective operating costs. The hierarchical model is converted to a unified mixed-integer linear programming problem via duality theory. To overcome the uncertainty of renewable energy during solving, a stochastic programming approach with Conditional-Value-at-Risk (CVaR) is employed to estimate the expected losses. Adopting the proposed pricing method in VPPs increases expected profit by 5.51%, purchase prices being 2.08% higher and selling prices 4.85% lower than unified pricing, benefiting both producers and consumers.