Abstract: Objectives In order to improve the power quality of ships under complex working conditions and solve the problem of power distribution and capacity configuration of composite energy storage system, this paper proposes a collaborative optimization method for hybrid energy storage systems in hydrogen-powered vessels based on an improved filtering strategy. Methods Firstly, the filtering strategy of wavelet transform and low-pass filtering is used to allocate the load power of ships. Considering the coupling characteristics of energy management strategy and capacity configuration of composite energy storage system, a collaborative optimization model considering ship power quality and economy is established. Then, the exponential distribution algorithm ( EDO ) is used to solve the relevant parameters of the optimal power allocation strategy and the optimal capacity allocation strategy. Results Simulation results show that the proposed method improves power fluctuation suppression by 13.01% and 55.35% compared to the two-layer optimization method and the rule-based method, respectively. In terms of economy, the system investment cost is reduced by 3.41% and 49.75%, respectively. Furthermore, this method confines the hydrogen fuel cell power fluctuation within 5 kW and reduces the power change rate to 0.85 kW/s, significantly enhancing system stability. Conclusions The collaborative optimization method effectively suppresses hydrogen fuel cell power fluctuations and improves the economic performance of the energy storage system, providing a feasible technical reference for the design and energy management of hybrid energy storage systems in hydrogen-powered ships.