氢能源站（Hydrogen Fueling Station, HFS）对氢气生产和供应至关重要。然而HFS通常在午夜和凌晨生产氢气，同时由于热负荷处于高峰，热电联产（Combined-Heat-and-Power，CHP）机组的灵活性降低。在可再生能源比例较高的电热综合能源配网系统（Integrated Electricity-Heat Energy System，IEDS）中，CHP机组缺乏灵活性将不可避免地影响IEDS和HFS的经济和安全运行。先前的研究侧重于配电网系统和HFS的协同运行，而没有考虑灵活性问题。论文旨在提出IEDS和HFS的协同分布鲁棒协同调度模型，以实现协同优化运行，从而降低运营成本和安全运营。在该模型中，HFS中产生的废热被回收后注入热网系统。利用历史数据对风电场景进行聚类，构建风电不确定性集。并相应地提出了具有更好收敛速度的改进Benders分解算法，从而进一步以并行和分散式求解所考虑的优化模型。此外，基于综合能源测试系统实例进行了算例仿真，在考虑HFS协同调度的前提下，所提出的协同调度模型能够使运行成本降低7.8%，并分析了所提出的协同调度模型在处理风电不确定性方面的有效性。最后，相比经典Benders算法，论文提出的改进Benders分解算法求解时间降低了70%，迭代次数减少了一半，验证了分散式优化具有良好计算性能。
英文摘要:
The hydrogen fueling station (HFS) is essential for the hydrogen production and supplies. However, the HFS often produces hydrogen at midnight and early morning, while the flexibility of the combined-heat-and-power (CHP) unit is reduced due to the heat load peak at the same time. In the integrated electricity-heat energy distribution system (IEDS) with a high proportion of renewable energy, the lack of flexibility of the CHP units will inevitably affect the economic and secure operation of IEDS and HFSs therein. Previous research focuses on the cooperative operation of power systems and HFSs without considering the flexibility issues. In this paper, it is aimed to propose a cooperative distributionally robust cooperative dispatching model of the IEDS and HFSs to achieve coordinated optimal operation for the operating cost-reducing and safety operation. In this model, the recycled heat power generated in the HFSs is considered to be injected into the thermal system. Moreover, the uncertainty set therein for the wind power is constructed based on the clustering wind power scenarios using the historical data. Furthermore, a modified Benders decomposition algorithm owning a better convergence rate is correspondingly proposed to further decompose and thus solve the considered optimization model in a parallel and decentralized manner. In addition, based on the example of the integrated energy testing system, a numerical simulation is carried out. Under the premise of considering the HFS cooperative scheduling, the proposed cooperative scheduling model can reduce the operating cost by 7.8%, and the effectiveness of the proposed cooperative scheduling model in dealing with the uncertainty of wind power is analyzed. Finally, compared with the classical Benders algorithm, the solution time of the modified Benders decomposition algorithm proposed in this paper is reduced by 70%, and the number of iterations is reduced by half, which verifies that the decentralized optimization has good computational performance.