Objective This study aims to address the economic optimal operation problem of distributed energy storage participating in power grid frequency regulation. A hierarchical optimal operation method between stations and within stations is explored to achieve the optimal allocation of frequency regulation power among energy storage stations and among energy storage units within each energy storage station. Methods First an optimal operation model was constructed based on a comprehensive consideration of the remaining frequency regulation capacity of each energy storage station with the goal of minimizing the energy loss cost and the life-decay cost. This achieved a reasonable allocation of frequency regulation power among energy storage stations. Second using the energy loss cost life-decay cost and regulation performance ratio as indicators the TOPSIS evaluation method with integrated subjective-objective weights was employed to evaluate the frequency regulation capacity of each energy storage unit. Subsequently the optimal allocation of frequency regulation demand among energy storage units was realized. Results The method was verified through comparative simulation experiments. The simulation results indicated that this method not only had good economic efficiency but also could fully mobilize the enthusiasm of each energy storage unit for frequency regulation. Conclusion Distributed energy storage participating in grid frequency regulation resolves economic problems through a two-layer optimization framework inter-station and intra-station levels . The proposed method reduces the cost of energy storage participating in power grid frequency regulation in the regional power grid and fully utilizes the frequency regulation capabilities of each energy storage station and its internal energy storage units.