EV Charger Placement

At the University of Texas at Austin, I worked with a team of other undergraduates to implement a reinforcement learning agent to try optimizing EV charging station placements. Check out the preprint here!

Problem & Motivation

Electric vehicles (EVs) are an important tool for mitigating climate change, as they reduce carbon emissions and harmful pollutants that are often emitted from conventional vehicles. However, as the number of EVs increases, there are concerns about the accessibility and convenience of charging stations for consumers. Thus, placing charging stations in areas with developing EV infrastructure is critical to the future succcess of EVs. We present a supervised machine learning model that predicts charging demand in a city based on factors such as traffic density and points of interest (POIs), as well as a Deep Q-Network that learns to sequentially place charging stations in optimal locations.

Methods

Results

Poster Presentation: I presented this research at the American Geophysical Union Fall Meeting in Chicago, IL in 2022. The full poster presentation is shown below.