Scalable Online Learning of Approximate Stackelberg Solutions in Energy Trading Games with Demand Response Aggregators

In this work, a Stackelberg game theoretic framework is proposed for trading energy bidirectionally between the demand-response (DR) aggregator and the prosumers. This formulation allows for flexible energy arbitrage and additional monetary rewards while ensuring that the prosumers' desired daily energy demand is met. Then, a scalable (with the number of prosumers) approach is proposed to find approximate equilibria based on online sampling and learning of the prosumers' cumulative best response. Moreover, bounds are provided on the quality of the approximate equilibrium solution. Last, real-world data from the California day-ahead energy market and the University of California at Davis building energy demands are utilized to demonstrate the efficacy of the proposed framework and the online scalable solution.