According to the problems of output fluctuation and consumption bottleneck caused by the high proportion of renewable energy grid connection

it proposes a capacity optimal configuration method for virtual power plant energy supply units including wind solar and energy storage.Firstly

it constructs a multi-objective optimization model with annual total cost

annual total carbon emissions

and annual total interaction power as objectives

balancing economic efficiency

environmental protection

and system independence.Secondly

it proposes an innovative hybrid solution strategy.It uses the NSGA-Ⅱ algorithm to generate a Pareto optimal solution set

and combines it with the TOPSIS method based on posterior entropy weighting to objectively assign weights and screen the solution set

addressing the conflicts among multiple objectives and the subjectivity of weights.Finally

simulation results based on 8760 hours of meteorological and load data from a region in Europe show that the optimized virtual power plant has an annualized cost 60.11% higher than traditional grid-connected power purchase

but reduces carbon emissions by 99%.The installed capacities of wind and solar power are 123% and 177% of average load

respectively

while the storage capacity reaches 2.5 times the load

enhancing the utilization rate of renewable energy.Further comparisons of configuration effects across different climate regions reveal that Greece(Mediterranean climate) has the best overall performance due to abundant sunlight resources and high load matching

validating the model's universality.Through multi-objective collaborative optimization and adaptation to climate differences

it can provide references for the efficient planning and low-carbon operation of virtual power plants.