氢能作为一种清洁无污染、能量密度大的二次能源，是大规模消纳新能源的理想储能载体，耦合氢储能系统和可再生能源的电热氢综合能源系统为消纳新能源提供新的思路和方案。围绕电热氢综合能源系统中如何以经济合理的方式投入氢储能设备展开研究，解决氢储能设备容量的合理配置问题以及考虑源荷不确定性对电热氢综合能源系统运行的影响，提出了一种考虑季节性存储和源荷不确定性的电热氢综合能源系统中氢储能容量优化配置方法。首先针对风电功率预测误差较大，电热气负荷预测精度较高的特点，分别采用不确定集和抽样场景描述源和荷两侧的不确定性。然后建立了考虑源荷不确定性和季节性存储的氢储能容量配置双层鲁棒随机优化模型，其上层问题以年化投资成本和运行成本的总成本最小化为目标确定氢储能系统装置容量，下层问题采用两阶段鲁棒随机优化模型模拟电热氢综合能源系统典型日在风电出力最恶劣场景下的最优运行方案。由于该模型难以直接求解，提出基于粒子群优化算法和列与约束生成算法对该类复杂模型进行求解。最后，通过对某个电热氢综合能源系统算例进行分析，算例分析结果验证了所提方法的有效性，获得的氢储能系统容量优化配置方案能够促进风电消纳和提高系统运行的经济性。

As a clean, pollution-free secondary energy source with high energy density, hydrogen energy is an ideal energy storage carrier for large-scale consumption of new energy. The electric-heat-hydrogen integrated energy system (EHH-IES), which couples hydrogen energy storage system (HESS) and renewable energy, provides new ideas and solutions for the consumption of new energy. Therefore, this paper focused on how to put in hydrogen energy storage equipment in an economically rational way, and aims to solve the problem of reasonable allocation of hydrogen energy storage equipment capacity and consider the impact of source and load uncertainty on the operation of electrothermal hydrogen integrated energy system. This paper proposed a method for optimizing the capacity of HESS in an EHH-IES considering seasonal storage and source-load uncertainty. Aiming at the relatively large prediction error of wind power and high forecasting accuracy of electric, heat and gas loads at first, the uncertain set and sampling scenario were used to elaborate source and load uncertainty, respectively. Then a bi-level robust stochastic optimization model for configuring hydrogen energy storage considering source-load uncertainty and seasonal storage was constructed, where the upper model optimizes the capacities of devices in hydrogen energy storage with the objective of minimizing total cost of annualized investment costs and operating costs, and the lower model was constructed as a two-stage robust stochastic optimization model to simulate the optimal operation scheme of the EHH-IES under the worst scenario of output wind power in typical days. Since the model is difficult to solve directly, particle swarm optimization and column and constraint generation algorithms were used to solve this type of complex model. Finally, through the analysis of case studies of an EHH-IES, the effectiveness of the proposed method was verified. The obtained solution for the optimal configuration of hydrogen energy storage system can promote the consumption of wind power and improve the economics of the system operation.