Abstract：In order to improve the power coordination ability among microgrids in the microgrid group，it is

necessary to consider the correlation between renewable energy generation output and the difference in load time

distribution in different regions in the microgrid group planning. A scenario generation method that considers the

spatio-temporal correlation of renewable energy power generation was proposed. Based on the spatial distance

between microgrids in microgrids and the temporal characteristics of renewable energy power generation output，

Nataf transformation and temporal reconstruction methods were used to generate scene-solar power output scenes

that meet the temporal and spatial correlation. Then，the capacity allocation model of multi-objective microgrid

group with minimum annual total cost and cumulative source-load difference was established. Finally，Jaya

algorithm with non-dominated sorting and congestion improvement was adopted to solve the problem. Fuzzy

membership function was used to evaluate the Pareto solution and select the optimal allocation scheme of wind

storage in microgrid group. The results of the example show that the total investment cost can be effectively

reduced by the precise allocation of the source storage capacity of the microgrid cluster，while the carbon emission

level of the microgrid cluster can be reduced and the absorption capacity of renewable energy can be improved.

      Key words：microgrid group capacity allocation；spatiotemporal correlation；Nataf transformation；multiobjective Jaya algorithm（MOJaya）