Hybrid market design for decentralized energy trading: multi-k double auction with enhanced stable matching

The rise of distributed energy resources (DERs) and sophisticated digital communication systems has led to the emergence of Peer-to-Peer (P2P) energy trading as a viable method for facilitating flexible, autonomous, and efficient energy exchange within Local Energy Communities (LECs). This paper proposes a novel market model for P2P energy trading by integrating a multi-k double auction framework with the Gale-Shapley stable matching algorithm and metaheuristic optimization. The multi-k double auction offers greater pricing flexibility compared to traditional mechanisms such as uniform, discriminatory, and pay-as-bid auctions. The trading price is computed using a weighted parameter k balancing buyer and seller preferences, where k is evaluated for static values (0, 0.5, 1) and optimized dynamically using Ant Colony Optimization (ACO). Buyer-seller matching is performed using a constraint-aware Gale-Shapley algorithm that respects buyers' budget limits and sellers' capacity constraints. The model is validated using realistic data derived from photovoltaic generation, load forecasts, and prosumer preferences based on the IEEE 13 and IEEE 37-node residential feeders. Simulations were conducted over hourly trading periods divided into bidding and power exchange intervals. In the present work, the proposed hybrid architecture achieves improved social welfare, fairness in price allocation, higher market liquidity, and enhanced satisfaction for participants. This framework establishes a scalable and adaptive decision-making process that supports stable, efficient, and equitable operations in decentralized energy markets.