Like utilities across the country, Rappahannock Electric Cooperative (REC) is expending substantial effort in future-proofing transmission and distribution networks amidst ever-increasing Electric Vehicle (EV) adoption. These efforts range from understanding the network effects associated with members' current EV charging loads, to predicting the likely impacts of members' future EV adoption, to evaluating the networks' ability to accommodate those EV charging loads. During this presentation, we will share the background and results from our recent transformer-loading assessment and also provide an overview of our machine-learning approach to detecting previously unidentified EV chargers across our distribution network.

As we prepare for the continued adoption of EVs by our members, distribution service transformers will play a critical role in maintaining reliable service across our service territory. These transformers vary substantially in both size and age, with many of them installed long before the recent emergence of 7-20 kilowatt EV charging loads. This year, we completed a study that allowed us to visualize transformer loading across our network and, using demographic-based EV-adoption-rate projections, predict future transformer loading and the associated individual-service limitations.

We are continuing to explore how machine learning can enable us to identify member-load anomalies that are consistent with EV charging. With the means to identify likely EV owners, we could then offer time-of-use rates to our members, monitor EV adoption rates in near real-time, and improve system planning and scheduling.

Session Takeaways:

• Future-proofing the grid requires identifying both the macro and micro impacts of Electrical Vehicle charging
• Distribution transformers will play a critical role in the EV transition process
• Machine learning can allow a utility to monitor the EV transition in near real-time