A Comparison of The Renewable Distributed Generation Models used in Reliability Assessment

Abstract

The increased penetration of renewable Distributed Generation (DG) creates room for further improvement of the reliability of supply in distribution networks. Several modelling approaches have been proposed to address the variability of renewable sources in reliability assessment. In this paper the renewable DG models used in reliability assessment of distribution networks are discussed and three representative models (capacity probabilistic table, hourly profiles and representative time-segments) are critically compared. The capacity of the methods to model the chronological variability of demand and generation is also addressed. The results of reliability indices and the computational times were used to compare the three modelling approaches when applied to a distribution network. It will be shown that hourly profiles yield the most accurate modelling of variable renewable generation but require intensive computational resources. Probabilistic table is not recommended when accurate results are required, while representative time-segments can be used as an alternative to hourly profiles when computational efficiency is needed.