The coordinated operation of hybrid energy storage systems （HESS) in the power system can make up for the shortcomings of single energy storage in multiple time-scales power and electricity scheduling, but its economic limitations limit its potential to participate in auxiliary grid peak-shaving and enhance new energy consumption in coordination with source load. Therefore, a hybrid energy storage load collaborative optimization scheduling method is proposed based on dynamic time zone partitioning. Firstly, based on the power imbalance between the source and load, a dynamic time zone division strategy is designed by considering the prediction error of new energy output and the demand side response, which guides hybrid energy storage to synergistically enhance the new energy consumption capacity and auxiliary peak shaving effect from the perspectives of electricity quantity and power at different day-ahead-intraday time scales. Secondly, with the goal of minimizing the total daily operating cost of the system, a system day ahead day rolling optimization scheduling model is constructed. Energy based energy storage provides peak shaving power support at the day ahead scale, while power based energy storage smooths out new energy fluctuations at the intraday scale. Rolling optimization is used to reduce system source load prediction errors, and an energy storage cycle life model is embedded to make the system scheduling results closer to actual operating conditions. Finally, the effectiveness and superiority of the proposed method in terms of economy, new energy consumption, and peak shaving effect are verified through numerical examples.