Analysis of Australian EV Charging Behaviour for Network Hosting Capacity Analysis

The growth in the adoption of electric vehicles (EV) poses significant challenges for the power grid, which calls for novel approaches for distribution network hosting capacity analysis. One of the main challenges is the modelling of the charging behaviour. In this context, this paper analyses real-life smart meter data from 109 Australian prosumers. A unique feature of the dataset is that it covers users who also have rooftop solar, which is increasingly common in Australia. The analysis shows that the presence of rooftop solar is the main driver of EV charging patterns, which stands in contrast to data from the UK used as a benchmark. The downside of the dataset is the lack of information about the battery state-of-charge, battery size and EV charger type. Still, the analysis shows that hidden information, including charger types, travel distance, and daily plug-in factor, can be inferred from the data. This shows that the charging data alone provide sufficient information for hosting capacity analysis. To address the challenge of the limited size of the dataset, we also propose a method for generating synthetic EV charging profiles using the Markov Chain model and Gaussian Kernel Density Estimation. The preliminary results demonstrate the effectiveness of the proposed modelling approach.