As the proportion of renewable energy continues to rise, the demand for rapid load balancing and frequency regulation in power systems is increasing. Advanced energy storage stations (ESSs), being highly flexible and adjustable resources, can provide quick and active support to the grid. However, the large number of these resources and their complex characteristics make it challenging to form effective control resources on a large scale. This paper proposes a method for evaluating the active support capability of clustered energy storage stations based on multi-scenario analysis. Firstly, using a combination of structural and functional performance indicators and the affinity propagation algorithm, multiple energy storage stations are grouped into clusters. Secondly, based on the proposed multi-level, multi-dimensional evaluation indicator system, a combined subjective and objective weighting method is employed to establish a grid-support matrix that links energy storage station indices with grid demand scenarios. Finally, to ensure the reliability and usability of the evaluation results, the K-means++ algorithm is employed for hierarchical evaluation of the active support capability of the energy storage clusters. The proposed method is validated on the IEEE 39-bus system.