Dynamic sizing of automatic and manual frequency restoration reserves for different product lengths

On last year's EEM conference a method for the dynamic sizing of frequency restoration reserve capacity based on quantile regression was presented. Further research has improved the method and has made it ready for use. It contains the following new features, which will be presented in this paper: an adaptive bias correction function, the allocation of frequency restoration reserves (FRR) to automatic FRR (aFRR) and manual FRR (mFRR) and the calculation of needed reserve capacity for different product lengths. The results will show the advantages of the adaptive bias correction, estimate the needed reserves for aFRR and mFRR, and demonstrate the influence of different product lengths on the needed reserve capacities.