Hierarchical distributionally robust scheduling strategy for distributed energy systems in the energy-sharing environment

Energy sharing in distributed energy systems constitutes the pivotal strategy development for enhancing clean energy utilization and low-carbon emission achievement. However, as market mechanisms continue to improve, energy trading in distributed energy systems will shift from the traditional system-to-system model to the multi-stakeholder model. Therefore, this study constructs a two-layer energy-sharing framework that contains different stakeholders. Firstly, the energy system operator guides the energy-sharing behavior among distributed energy systems through energy transaction pricing to maximize its revenue. Then, the distributed energy systems obtain optimal energy sharing and internal operation strategies based on the energy system operator's price signals to minimize their energy costs. Additionally, this study addresses the uncertainty of renewable energy generation in distributed energy systems using the Wasserstein metric ambiguity set, and combines it with the energy sharing issue to form a distributionally robust energy trading optimization model. Finally, to solve the two-layer multi-agent distributionally robust energy sharing problem, we employ strong duality theory to transform the problem into a more solvable form. An adaptive genetic algorithm-analytical target cascading method is proposed to achieve optimal transaction pricing and energy scheduling. The case analysis results demonstrate that the proposed strategy can achieve economic benefits of 5.51 % and environmental benefits of 5.73 %, effectively balancing economic efficiency and robustness.