Project Info

Physics-informed machine learning for controlling renewable energy systems

Qiuhua Huang

Project Goals and Description:

Across the globe, the transition to renewable generation is placing legacy energy system control systems under increasing stress, decreasing grid reliability and increasing costs. Advanced control systems are urgently needed to ensure power system reliability by improving the accuracy and speed of critical control tasks such as generation-load balance and preventive control. Simultaneously, exciting theoretical advances are being made in our ability to design optimal and robust controllers in a data-driven fashion, bypassing the costly model-building and validation steps normally required for model-based design. Research in Prof. Qiuhua Huang’s group bridges advanced AI and computing technologies with energy and sustainability applications, developing the former for use in the latter. The research objectives are to build practical and rigorous theoretical frameworks for nonlinear, data-driven control and decision-making for enabling a sustainable energy future, creating transformative change in our ability to manage complex engineered systems.

More Information:

Grand Challenge: Make solar energy economical.
Lab website: Related publications:
  • Huang, Y. Chen, T. Yin, Q. Huang, J. Tan, W. Yu, X. Li, A. Li, Y. Du, "Learning and Fast Adaptation for Grid Emergency Control via Deep Meta Reinforcement Learning," in IEEE Transactions on Power Systems, vol. 37, no. 6, pp. 4168-4178, Nov. 2022
  • Yan, Q. Huang*, R. Huang, T. Yin, J. Tan, W. Yu, X. Li. “Physics-informed Evolutionary Strategy based Control for Mitigating Delayed Voltage Recovery”, IEEE Trans. on Power Systems, vol. 37, no. 5, pp. 3516-3527, Sept. 2022
  • Huang, R. Huang*, W. Hao, J. Tan, R. Fan, Z. Huang. “Adaptive Power System Emergency Control Using Deep Reinforcement Learning,” IEEE Transactions on Smart Grid, vol. 11, no. 2, pp. 1171-1182, March 2020
  • Marot, B. Donnot, K. Chaouache, A. Kelly, Q. Huang, R. R. Hossain, & J. L. Cremer. “Learning to run a power network with trust.” Electric Power Systems Research, 212, 108487, 2022

Primary Contacts:

Qiuhua Huang,

Student Preparation


  • Motivated to solve challenging energy and sustainability problems
  • Background in Electrical Engineering or Computer Science ( background in power systems is a plus)
  • Basic knowledge of machine learning, data science
  • Programming skills in Python or MATLAB


4-5 hours


  • Machine learning, particularly reinforcement learning,  for power and sustainable energy systems
  • How to approach a challenging real-world problem, break it down into manageable subtasks, and translate them into rigorous mathematical formulations
  • Programming skills
  • Paper reading and writing


Provide a project research plan and related references on day 1; have weekly meetings for discussions; will also involve the students in the research group and support the students to work with Ph.D. students on similar topics.


Share This