Electric vehicle hosting capacity in low-voltage networks considering the probability of concurrent charging

With the rapid development of electric vehicles (EVs) and the growing number of charging stations installed in distribution networks, distribution system operators (DSOs) are facing greater uncertainty and may experience more frequent violations of network operating limits. This work proposes an approach for estimating the maximum EV charging station penetration a low-voltage (LV) distribution network can accommodate without introducing line or transformer overloading, or voltage issues. This is referred to as the EV hosting capacity (HC) of the network.
Unlike photovoltaic (PV) installations that behave almost uniformly in the same region, EV charging station usage depends heavily on customer habits. This makes EV HC estimation significantly more challenging. The HC is estimated using a Monte Carlo approach which allows to address the uncertainties in future EV charging station deployment. Compared to prior work, our contribution lies in introducing a random variable indicating EV stations charging concurrency. This variable quantifies the percentage of customers simultaneously charging their EV during the peak hour, and is estimated by analysing the EV owners charging behaviour through consumption profiles. Tests on a real-world distribution network show that our approach avoids overly pessimistic HC estimates and precipitate investment decisions.