General Lab Information

Energy Systems Division

Renewables and Grid Analysis

Advanced Grid Modeling

A Deep Learning Based Online Platform for Critical Anomaly Detection and Emergency Control to Enhance Grid Reliability and Resiliency


Develop a data-driven, deep learning-based solution to prevent the propagation of cascading failures when the grid is challenged by unexpected contingencies or combinational contingencies under uncertain environments.

An end-to-end technology will be developed that will be an online platform capable of evaluating and predicting grid conditions and selecting emergency control actions. The focus will be on load shedding strategies and determination of timing and boundaries for splitting the grid into self-sustained islands, as needed, to mitigate the propagation of cascading failures.


A machine learning method using graph-based recurrent neural network will be developed to predict grid dynamics based on measurement data. Data on generation, wind speed and direction for wind and solar generation sites will be acquired from ERCOT, along with PMU measurement data for transient events that were caused by generator trips.


Y. Zhang, M. Yue, and J. Wang, “Adaptive Load Shedding for Grid Emergency Control via Deep Reinforcement Learning,” accepted by IEEE PES GM, 2021.

S. Zhang and G. Lin, “Robust subsampling-based threshold sparse Bayesian regression to tackle high noise and outliers for data-driven discovery of differential equations,” Journal of Computational Physics, in press, 2020.

Y. Wang and G. Lin, “MFPC-Net: Multi-fidelity Physical Constrained Neural Process,” CSIAM Transactions on Applied Mathematics, in press, 2020.