【目的】 针对微电网中可再生能源接入带来的功率波动与用户用能需求多样化导致的负荷峰谷调节难以协同优化的问题，提出一种考虑源-荷双侧差异化需求的混合储能系统（hybrid energy storage system，HESS）协同优化配置方法。【方法】 首先，提出改进经验模态分解（empirical mode decomposition，EMD）方法将分布式电源（distributed generation，DG）出力分为高、中、低三个频段，结合不同储能类型的经济技术特性，实现多时频域下的HESS功率分配。其次，构建考虑电源侧最大波动量指标和新能源消纳率指标以及负荷侧峰谷差降低率指标的微电网源-荷双侧HESS协同优化配置模型。最后，提出基于改进交替方向乘子法（alternating direction method of multipliers，ADMM）的协同优化求解算法实现HESS协同优化配置模型的分布式优化求解。【结果】 仿真结果表明，所提方法在保障较好经济性的前提下，能有效平抑DG出力波动并实现负荷的削峰填谷，源-荷双侧协同优化后实现了新能源消纳率提高到99%，负荷峰谷差降低率达到了84.94%。【结论】 所提协同配置方法有效提升了微电网对新能源的消纳能力与对负荷峰谷差的调节能力，为微电网HESS的合理配置提供理论依据和技术参考。

[Objective] To address the challenge of achieving coordinated optimization in peak-valley load regulation， which is caused by power fluctuations from renewable energy integration and diverse user energy demands in a microgrid， this paper proposes a coordinated optimal configuration method for a hybrid energy storage system （HESS） considering the differentiated requirements on both source and load sides. [Methods] First， an improved empirical mode decomposition （EMD） method is presented to divide the distributed generation （DG） output power into high-， medium-， and low-frequency bands. By considering the economic and technical characteristics of different energy storage types， power allocation of the HESS is realized across multiple time‑frequency domains. Second， a coordinated optimal configuration model for source-load sides HESS in a microgrid is constructed. This model considers source side indicators， specifically the maximum power fluctuation indicator and the renewable energy integration rate indicator， as well as the load side indicator of the peak-to-valley difference reduction rate. Finally， a coordinated optimization solution algorithm based on the improved alternating direction method of multipliers （ADMM） is presented to achieve the distributed optimal solution of the HESS coordinated optimal configuration model. [Results] Simulation results demonstrate that， while ensuring favorable economic performance， the presented method effectively mitigates DG output fluctuations and accomplishes load peak-cutting and valley-filling. Through coordinated source-load sides optimization， a high renewable energy integration rate of 99% is achieved， and the peak-to-valley difference reduction rate reaches 84.94%. [Conclusions] The proposed coordinated optimal configuration method effectively enhances the capability of the microgrid for renewable energy accommodation and its regulation capacity over load peak-to-valley differences. It provides a theoretical foundation and technical reference for the rational configuration of HESS in a microgrid.