Abstract: Considering the allocation of demand-side flexible resources（referred to as “resources”）, the joint planning of grid-load-storage is a critical way to improve the flexibility of the distribution network to adapt to the fluctuation of new energy output. Given this,we propose a research model for distribution network planning combining scenario selection for lack of flexibility and improvement of uncertainty analysis method. Firstly,the distribution network’s flexible supply and demand relationship is modeled from the perspective of resources. Secondly, based on shadow price theory, a scenario selection strategy for distribution network flexibility lackness is proposed, and screening indicators are established to describe the closed-loop relationship between planning resources, operation constraints and scenario screening. Then, the strategy is embedded in the distribution network-multi-resource two-stage joint planning model, the first stage considers the optimal investment cost, and the second stage coordinates operation and flexibility comprehensive cost expectations. The planning model is reconstructed and solved under the framework of improved distributionally robust optimization based on scenario probability driven（ISPD-DRO）, which has the advantage of realizing the organic unity of probability optimization solution and scene dynamic update. Finally, combined with the case analysis, the advantages of the proposed model in improving decision-making economy and distribution network flexibility are verified.