Yufan Zhang is a postdoctoral researcher in the Department of Electrical and Computer Engineering at University of California San Diego. She received her Ph.D. in electrical engineering from Shanghai Jiao Tong University in 2022, and a B.S. degree in Electrical Engineering from Chongqing University in 2017. Her research lies in machine learning and optimization, with applications in sustainable energy and power systems. She is a recipient of multiple rewards, including the PES General Meeting Best Paper Award in 2020.
To achieve carbon neutrality, energy systems are transiting toward a more sustainable future, by replacing fossil-fuel-driven power plants with weather-dependent renewable energy sources like wind and solar. Introducing renewable energy sources requires rethinking forecasts that inform the uncertainty, and how forecasts impact the system operation. My research aims to understand how improved forecasts can be used to operate the system under uncertainty to improve social welfare and decision-making robustness. Accordingly, my research agenda is centered around developing decision-focused forecasting tools. Thrust 1 studies advanced forecasting methods, with optimization in the loop. While the optimization objective in this thrust may not directly align with the decision goal, such as minimizing operation costs, it provides valuable forecasting methods that are applicable in a broader context for subsequent thrusts. Thrust 2 emphasizes the role of forecasts in reducing the expected operation cost of sequential energy dispatch. It seeks to improve renewable energy forecast toward minimizing the expected operation cost of the entire operation, i.e., improving overall economic efficiency. Thrust 2 primarily deals with normal operations, and contributes theoretically by characterizing the impact of forecasts on sequential operation goals through the establishment of an analytical function. To immune systems against all potential realizations of uncertain renewable energy, Thrust 3 enhances forecasts to enable robust system operation while preserving cost efficiency.