Accelerated distributed scheduling of integrated community energy systems considering electricity-heat sharing

Integrated community energy systems (ICES) provide an efficient and sustainable approach to local energy management through multiple energy carriers including electricity, heating, and cooling while enhancing end-use energy management. Electrical and thermal connections between ICESs can further improve energy utilization and economic efficiency. However, energy sharing among ICESs still faces challenges, including complex centralized scheduling constraints, privacy protection concerns, and slow convergence and reduced efficiency of distributed algorithm convergence. To address these challenges, this paper develops a stochastic optimization model for day-ahead energy sharing scheduling, accounting for uncertainties in renewable energy sources and losses in inter-community heat networks. The key idea is to establish a multi-community energy management framework in which the ICES operator provides dynamic electricity and heat trading prices based on the initial supply-demand information. Each ICES optimizes power output and energy interactions to minimize operational costs, carbon emissions, and renewable curtailment, thereby improving overall system efficiency. A distributed linearized surrogate augmented Lagrangian relaxation (LSALR) algorithm is developed to solve the optimal scheduling problem for interconnected ICESs in a privacy-preserving and computationally efficient manner. Testing results demonstrate the model significantly improves operational efficiency and promotes renewable energy utilization within ICESs. Compared with benchmark distributed algorithms, the proposed method achieves superior performance in terms of convergence rate and solution quality.