Comparison of models and tools for distribution planning

The integration of no-network planning alternatives into distribution system planning tools requires a proper modelling of load and generation behaviours within network calculation. To capture the operational aspects that can affect the planning stage, the time variability of demand and generation has to be explicitly represented in the planning calculations. Moreover, to reduce the computational burden and take account of the uncertain production and consumption of electricity, stochastic models have to be implemented. Two approaches are commonly proposed in the scientific literature: the clustering of demand and generation variability in a year taking account of the occurrence probability of each cluster, and the identification of typical daily profiles that characterise the weekly and seasonal customers’ behaviour during a year modelling their uncertainties through suitable probability density functions. In the paper, the comparison of these two approaches have been shown, underlining qualities and faults (e.g., accuracy, computational time burden, their applicability). The comparison has been performed considering a realistic distribution network, in presence of distributed generation (DG).