Electrification Research
About
Our work connects EV transportation demand with smart-grid constraints using data-driven models to create feasible infrastructure recommendations that meet driver needs. We are also building an End-to-End Decision Support System (E2E-DSS) that jointly tracks the transportation system state, the smart-grid state, and charging availability to match EV users to open chargers in real time.
Projects
Data-Driven EV Charging Electrification Planning for Tennessee
Our electrification work provides a data-driven view of how EV charging is actually used across Tennessee, translating real-world charging sessions into actionable insights for infrastructure planning and operations. By analyzing statewide charging behavior and linking station performance to surrounding land uses and local conditions, this effort helps utilities, cities, and charging providers make better decisions about where to expand, how to prioritize upgrades, and how to improve reliability and user experience as EV adoption grows.
Machine Learning–Based EV Charging Detection from Smart-Meter Data
This project develops a practical machine-learning capability to detect EV charging activity from standard, low-frequency smart-meter data—without needing dedicated sensors on the charger—so utilities and grid operators can better understand and anticipate charging demand at scale. By reliably identifying when EV charging is occurring in real-world household electricity profiles, the approach supports applications such as load forecasting, infrastructure planning, and operational strategies that reduce stress on the grid as EV adoption accelerates.
E-Transit-Bench: Integrated Electric Bus–Grid Simulation for Chattanooga
This project (E-Transit-Bench) provides an integrated simulation platform that evaluates electric bus transit operations together with the local power distribution grid, enabling planners to see how day-to-day transit service and charging decisions jointly shape mobility performance and grid impacts. Using Chattanooga as a real-world testbed, the platform supports applications such as planning charging infrastructure and depot/en-route charging strategies, assessing operational scenarios (including special-event “peak day” conditions), and informing grid-aware transit policies that reduce system stress while maintaining reliable service.