Postdoctoral scholar, Stanford University
Learning, Modeling and Optimization of the Smart Grid
The electric grid is undergoing profound transformation driven by the growth of renewable and distributed energy resources, deployment of new technologies, and involvement of consumer preferences. This transformation presents new challenges in system operation and opportunities in energy management.
In operating the electric grid, system operators must ensure that supply and demand is continually kept in balance. As more renewable resources such as solar and wind are being integrated into the grid, matching generation to load becomes more challenging due to the intermittency issue. The electric grid, therefore, can no longer be operated in a deterministic way. To this end, a stochastic paradigm is created for power system and market operation. The key idea is to characterize future operating conditions in a probabilistic way and incorporate these uncertainties into operations and decisions. In this paradigm, the first probabilistic forecasting method is developed for real-time operations, and a stochastic optimization framework is proposed for inter-regional interchange scheduling.
On the consumption side, electric meters with enhanced communication capabilities (smart meters) are becoming more prevalent, especially for residential customers. How to process and utilize data at different time-scales and resource resolutions from smart meters is the key to demand side management (DSM). As an essential element for implementing DSM, data-driven models are built for electricity demand of individual homes and even appliances. Customers’ demand and preferences are then learned from these load models.
Yuting Ji is a postdoctoral scholar at Stanford University, hosted by Ram Rajagopal. She received the PhD degree in electrical engineering from Cornell University, advised by Lang Tong, in 2017, and the bachelor degree in computer science from Tsinghua University, Beijing, China, in 2011. Her research interests are in data analysis, optimization, and statistical learning, with application to energy systems and the smart grid.