Graduate Researcher
Current
Advanced Process Control And Applied Mathematics Lab, Ucla, California, Us
Developed first-principles and data-based models for capturing dynamics of nonlinear chemical processes simulated in Aspen Plus, to implement model predictive control (MPC)Incorporated encryption algorithms and ML-based cyberattack detection to enhance cybersecurity across centralized, decentralized, and distributed MPC systemsImplemented dynamic optimization of process economics in an encrypted multi-layer control framework and improved economic closed-loop performance by 47%Developed an encrypted decentralized MPC with a state-predictor to eliminate 99% of the time-delay-induced output oscillations, and reduced computation time by over 70%Created encrypted distributed MPCs with state-estimators that use only 40% of the original state feedback, and have a 60% lower computation time