The large-scale integration of distributed photovoltaic (PV) systems and their inherent uncertainties have significantly increased the operational risks to distribution networks. Moreover

frequent extreme events have led to substantial out-age-related losses

highlighting an urgent need to improve the networks resilience against disasters and enhance their load-supply capabilities. To address these issues

an energy-storage allocation method that balances economic performance and comprehensive vulnerability is proposed. First

a two-stage vulnerability evaluation framework for distribution networks is established. In the pre-event stage

weak lines and nodes within the distribution network are identified based on improved electrical betweenness

line voltage-deviation rate

and network equilibrium indices. In the post-event stage

based on the identified vulnerabilities

random-fault and worst-case fault scenarios are constructed to simulate extreme disaster conditions. Additionally

an improved network-supply-efficiency index is developed to quantify the recovery capability of grid under such conditions.Subsequently

a multi-objective energy-storage allocation model is established

using economic performance and comprehensive vulnerability as objective functions. A multi-objective particle swarm optimization algorithm enhanced by chaotic mapping and simulated annealing (SA-MPSO) is then employed to determine the optimal siting and sizing of energy storage units. Finally

simulations are performed on a modified IEEE 33-bus test system under random and deliberate attack scenarios

representing realistic random and worst-case conditions. Simulation results demonstrate that integrating energy storage effectively enhances the supply capability and disaster resilience of the distribution network

validating the effectiveness of the proposed method.