Learning to Stabilize: ML-Enhanced Frequency Control in Renewable Power Systems

Proposed Faculty Supervisors: Ashu Verma, Priyank Srivastava

The global shift towards sustainable energy solutions has led to a significant increase in the penetration of renewable energy sources and distributed energy resources (DERs) within power grids. While this transition brings numerous environmental and economic benefits, it also introduces challenges to grid stability, particularly due to the variable and intermittent nature of these resources. One critical consequence is the time-varying system inertia, which undermines the effectiveness of conventional frequency control strategies that were designed under the assumption of constant, high inertia.

This PhD project proposes the development of a machine learning-based optimal frequency control framework that remains robust and effective across a wide range of system inertia conditions. The primary objectives are to design a controller that:

• Achieves optimal frequency stabilization under dynamic and uncertain inertia conditions,
• Incorporates formal mathematical analysis to guarantee critical properties such as stability, robustness, and optimality, and
• Operates independently of real-time measurements of system inertia, enhancing its practical applicability in future power systems.

The research will involve theoretical development, rigorous mathematical analysis, and validation through simulations and hardware-in-the-loop implementation.

Background required: Bachelor�s degree in electrical engineering. A strong interest in mathematical control theory and power systems is desirable.