Abstract: Energy storage device is key to support the construction of new power system. For a single entity, investment in energy storage under a specific application scenario is difficult to guide the cost and low return on investment. Through the business model of "crowdfunding co-construction and cluster sharing", energy storage utilization and benefits can be improved, and the cost allocation and benefit sharing of energy storage resources can be realized among multiple entities. Therefore, based on co-construction and sharing business model, this paper proposes an energy storage planning framework that considers multi-scenario demands of multiple investment entities, an energy storage planning framework is established as a data-driven distributionally robust optimization model, and the uncertainty of new energy output is depicted via comprehensive norm constraints. At the same time, dynamic frequency safety constraint is established in the optimization model to ensure the safety of system frequency, and the complex non-convex nadir of frequency constraint is transformed into linear constraint by linearization technology. Secondly, the CCG algorithm is used to solve the programming model. Finally, the cost allocation mechanism based on the Shapley value method reasonably apportion the investment cost of energy storage under co-construction and sharing model. The validity of model is verified on the IEEE39-node system. The results show that the proposed energy storage planning method can compromise the demands of multi-entity and multi-scenario, and reduce the investment cost of each entity.