Development of Day-Ahead Peer-to-Peer Energy Trading with Time-Series Clustering

Abstract

Day-ahead trading of electricity has been applied to ensure the balance between the amount of electricity sold and bought. Even so, due to the intermittent distributed energy resources (DERs), the actual condition can be varied significantly, and forecasting can be costly in order to provide high accuracy to minimize losses. Hence, this paper proposes a novel model-based day-ahead peer-to-peer (P2P) energy trading with regionalized trading prices, which are determined through time-series clustering. To improve the determination of price regions, the data parameter is derived from the day-ahead condition, which is forecasted from network condition, trading capacity, and trading price of the P2P energy trading. The performance of the proposed model of day-ahead P2P energy trading is evaluated with respect to the market operation stability and optimality.