A virtual synchronous generator (VSG) is a promising control strategy in future power systems owing to its inertial support and primary frequency regulation. However

these benefits are achieved at the cost of more transient energy demand (TED) of VSG

which results in a larger energy storage (ES) and hence high cost. Therefore

a cost-efficient VSG parameter optimization procedure aimed at minimizing TED while meeting the requirements of the grid frequency support effect

is proposed in this paper. First

a systematical modelling method and calculation algorithm for the TED and the maximum grid frequency deviation (MGFD) during a frequency regulation transient is presented. Afterwards

by considering MGFD as a constraint and TED as an optimization objective

we propose an optimization approach to find the optimal VSG parameters for cost-efficient frequency support. Based on the proposed calculation and optimization algorithms

we also investigate how main VSG parameters influence MGFD and TED and how VSG penetration rate affects optimization results. Experimental results validate the accuracy of the proposed calculation and optimization algorithms. It is shown that by applying the proposed optimization procedure

the capacity of energy storage can be reduced up to 44%.