Abstract: With the increasing penetration rate of the uncertain renewable energy in the microgrids, the challenges of the new energy accommodation in the microgrids has been growing, so the collaborative operation of the shared energy storage and the  multi-microgrid is an effective means to improve their consumption ability. To solve the coordinated dispatch problem of the multi-microgrid and the shared energy storage in the uncertain conditions, an optimal dispatching model considering the uncertainty of the new energy generation is proposed. Firstly, the typical scenarios of the weekly net generation power are generated by a daily state transition process collection generation based on the Markov Chain Monte Carlo algorithm and a daily scenario generation method based on the conditional generative adversarial networks. Subsequently, using the generated typical scenarios as the operational simulation scenarios, an optimization dispatching model with single-leader and multi-follower game is constructed, where the shared energy storage operator is the leader and the microgrids are the followers. In this model, the leader adjusts the leasing prices of the shared energy storage capacity to pursuit the maximum profit, while the followers adjust the leasing capacity schemes and operation plans in response to the leasing prices in order to minimize their own electricity consumption costs. Furthermore, to solve the proposed model, a distributed iterative solution method based on a heuristic algorithm is designed. Finally, the effectiveness of the proposed model is verified by the application experiment.