【目的】 高渗透率分布式可再生能源接入使配电网面临源荷时空错配与场景概率偏差并存的不确定性挑战，亟需兼顾结构自治与不确定性适配的规划方法。提出一种运行反馈驱动的动态集群划分-规划配置-调度运行的闭环框架，建立风电、光伏等分布式电源（distributed generation，DG）、智能软开关（soft open point，SOP）与分布式储能系统（distributed energy storage system，DESS）的协同配置模型。【方法】 首先，构建关键时段加权归一化的源网荷储匹配度指标，并结合模块度指标形成刻画结构与资源协同的综合性能指标，以指导集群划分；其次，针对风光出力的不确定性，在场景概率存在偏差的条件下建立两阶段分布鲁棒协同规划模型；进一步，从运行调度中提取跨区互济与电压约束压力等运行边际信息，自适应调整分区边界，使分区结果更适配规划与运行需求，并采用外层分区迭代与内层列约束生成相结合的双循环算法进行求解。【结果】 在IEEE 69节点系统中，所提方法可在年综合成本、可再生能源消纳与弱耦合自治运行结构方面取得协同提升。【结论】 研究成果为DG与灵活资源协同配置提供了理论支撑，对配电网的规划运行具有重要意义。

[Objective] The high-penetration integration of distributed renewable energy sources poses dual uncertainties to distribution networks， including spatiotemporal mismatches between generation and load and deviations in scenario probabilities， necessitating planning methods that jointly balance structural autonomy and uncertainty adaptivity. This paper proposes an operation-feedback-driven closed-loop framework integrating dynamic cluster partitioning， planning configuration， and dispatch operation. A collaborative configuration model is developed for distributed generation （DG） （including wind turbines and photovoltaics）， soft open points （SOP）， and distributed energy storage systems （DESS）. [Methods] First， a weighted and normalized source-grid-load-storage matching index based on critical periods is constructed and combined with the modularity index to form a comprehensive performance metric for cluster partitioning. Second， to address renewable output uncertainty under biased scenario probabilities， a two-stage distributionally robust coordinated planning model is established. Furthermore， marginal operational information， such as cross-zone mutual support and voltage-constraint stress， is extracted from dispatch results to adaptively adjust partition boundaries， ensuring the partitioning outcomes are better aligned with planning and operational needs. The resulting model is solved using a dual-loop algorithm that couples outer partition iterations with an inner column-and-constraint generation （C&CG） procedure. [Results] Case studies on the IEEE 69-node system demonstrate that the proposed method achieves synergistic improvements in annual comprehensive cost， renewable energy accommodation， and weakly-coupled autonomous operation. [Conclusions] This research provides theoretical support for the collaborative configuration of DGs and flexible resources， offering significant implications for the planning and operation of modern distribution networks.