On the morning of April 24, 2022, Zhao Junbo, assistant professor with the School of Electrical and Computer Engineering, University of Connecticut, was invited to deliver a speech at the “Online Lecture Hall” on the topic of “Physics-Informed Deep Reinforcement Learning for Power System Optimization and Control”. The series of online academic lectures were jointly organized by the School of Electrical Engineering of Chongqing University and the State Key Laboratory of Power Transmission Equipment and System Security, and was hosted by Associate Professor Ren Zhouyang.

Before the lecture officially started, Professor Ren extended a warm welcome to Professor Zhao’s participation and provided a brief introduction of Professor Zhao’s academic background and research areas.

During the lecture, Professor Zhao categorized three application scenarios according to the different quality and quantity of historical data, and expounded the methods of enhancing machine learning with physical constraints respectively. At the same time, he emphasized that deep learning framework combined with physical constraints has great application potential in power system operation optimization.

During the exchange session, Professor Zhao had in-depth discussions with the participants on the accuracy of the prediction model, the advantages of multi-agent reinforcement learning and the specific embedding method of physical constraints.

Professor Zhao encouraged students to take challenges, break conventional thinking and pursue the heights of scientific research. He also welcomed interested students to join the research group for doctoral or post-doctoral study and research.