Undergraduate Research and Innovation Scholar
Ensemble Machine Learning for Adaptive Electric Vehicle Charging Under User Behaviour Uncertainty
- Machine Learning
My project focuses on managing the electricity demand from a highly concentrated number of Electric Vehicles (EVs), in order to reduce the stress on electricity distribution networks. The first phase involves developing a number of statistical and deep learning models to estimate the charging parameters of EV owners using limited information such as past arrival time and consumption patterns. Ensemble methods will then be applied to obtain better predictive performance. In the second phase, we will build on the existing predictive models by developing pricing and incentive algorithms to shape the aggregate response from the fleet of EVs, to reduce the stress on the grid and maximize consumer welfare.
I am participating in SuperUROP because I am very interested in applying my machine learning and reinforcement learning knowledge (from 6.3900 and 6.7940 respectively) and mathematical background to a longer-term research project. I am excited to expand my knowledge and skills in these areas through a project with real-life practical applications. I hope to have meaningful results to display by the end of the SuperUROP and publish a paper.