Distributed energy storage (DES) has garnered significant attention as a crucial frequency regulation resource. However

existing DES aggregation methods are difficult to compromise the model accuracy with multiple influencing factors such as temperature and communication reliability

and the frequency regulation strategy cannot consistently maintain the schedulable potential of the cluster. To address these issues

this paper proposes a DES aggregation modeling method with considering multiple influencing factors and a two-layer consistency-based frequency regulation strategy. First

a dual Markov chain aggregation model is constructed

incorporating SOC interval double-layer discretization

health state correction

and a temperature auxiliary chain. Second

a two-layer control strategy is designed as follows: the upper layer preliminarily allocates grid frequency regulation commands based on SOC grouping; the lower layer treats DES within the same group as a multi-agent system

a secondary power distribution strategy is proposed based on fixed-time proportion consistency algorithms under dynamic topology

and the algorithm stability is theoretically verified. Finally

DES responds according to the power command to achieve frequency regulation.