Responsibilities include:

• Devise scientific machine learning frameworks for end-to-end learning of optimization problems satisfying hard constraints and providing strong provable bounds.
• Develop algorithms for large-scale problems arising at the integration of planning and operation decisions of highly electrified chemical industries.
• Design frameworks based on large language models to (i) analyze solutions and repair infeasible optimization problems and (ii) detect, identify, and diagnose faults in industrial processes.

Responsibilities include:

• Develop a unit commitment model with AC network constraints based on a second-order conic relaxation of the power flow equations and an algorithm to achieve an AC feasible solution.
• Design an algorithm based on Benders decomposition to address the AC network-constrained unit commitment problem of large-scale power systems.
• Propose a risk-aware network-constrained unit commitment model and an algorithm based on decomposition techniques to address the solution of large-scale power grids.
• Formulate a unit commitment model considering the optimal behavior of the natural-gas market and a solution method for large-scale systems based on an outer approximation algorithm.
• Devise a solution method for the corrective security-constrained unit commitment problem for large-scale power systems using a hybrid decomposition and a Kron-based network reduction technique.

Responsibilities include:

• Develop a tool to visualize and analyze the behavior of distribution networks based on a component-based aggregate load model.
• Investigate the physical feasibility of engaging plug-in electric vehicles to support a power grid with increasing renewable energy.