Abstract: Firstly, the polyhedral uncertainty set and the uncertainty set based on Wasserstein distance are used to describe the uncertainty of regulation signal and real-time energy-regulation market price respectively. Secondly, a distributionally robust optimization model of independent energy storage power stations participating in the joint market is established to collaboratively optimize the declared capacity of different markets. In the model, the real-time frequency regulation performance index is introduced to reflect the comprehensive influence of frequency regulation response on regulation revenue, charge and discharge cost and degradation cost, so as to better optimize the self-scheduling plan of energy storage. Then, the Lagrangian dual method is used to transform the established distributionally robust optimization model into a classical robust optimization model, and the Gurobi solver is used to solve the problem after linearization by removing the absolute value and constructing the piecewise function. Finally, the correctness and effectiveness of the proposed optimization model are verified by an example.