Data-driven Modelling of Representative Rural Distribution Networks for Reliability Studies

A distribution system is expected to supply energy to its customers with acceptable levels of reliability and affordability. The integration of distributed energy resources (DERs) into rural Australia presents several unique opportunities and challenges due to the sparsely populated areas and long distribution feeders. There is a need to develop distribution network models for reliability study in rural distribution feeders with various DERs. This paper presents a data-driven approach to model rural distribution networks for reliability analysis. At first, erroneous and missing values from the customer's smart meter data are identified and processed. Network data, such as the data of distribution lines, transformers, and circuit breakers, are collected from relevant network service providers and other publicly available sources. Next, representative distribution networks of six Western Victorian Region (Australia) rural areas are constructed. Reliability models of feeders, transformers, solar photovoltaics (PVs), and battery storages are developed using publicly available model parameters collected from relevant service providers. Finally, the reliability indices are evaluated considering a one-year time frame. The percentage of energy not served (ENS) with respect to the total energy required by the grid is more than 0.5% in all selected areas.