Public Research Seminar by Sustainable Energy and Environment Thrust - Power System Security Rules Extraction and Embedding in Operation Optimization under High Renewable Energy Penetration

The strong uncertainty and low inertia that stem from high renewable energy generation have fundamentally changed the power system operation states, making the system more likely to face complex security and stability events frequently, such as the England power cut in 2019 and the Texas power crisis in 2021. Besides, most security and stability rules are difficult to express analytically, difficult to understand, and difficult to be embedded in power system operation optimization. To this end, we propose sparse oblique decision tree and its ensemble method to extract arbitrary power system security rules and embed them in operation optimization to prevent potential events. The extracted rules are sparse, linear, and easy to be embedded in optimization. The method has successfully embedded N-k security, small-signal stability, and voltage stability rules as constraints in power system optimization. The results show it can significantly increase the percentage of secure states, and reduce the solution time of the security-constrained economic dispatch. The work provides key models and effective methods for power system security-constrained decision-making under high renewable energy penetration.