Speaker

Dr. Masoud Barati, Assistant Professor, University of Pittsburgh

Abstract

Non-convex programming deals with optimization where the objective function or constraints are non-convex. Such problems are common in engineering systems. Although there's extensive research on convex and non-convex quadratic programming, existing methods mainly cater to the convex form, often underperforming for non-convex cases. This presentation delves into a new integrated approach aimed at effectively solving non-convex programming, especially prevalent in sectors like power and energy systems, transportation, and communications.

One pressing issue in power and energy system management is the inherent difficulty of optimization. This talk's primary goal is to tackle this non-convexity by introducing optimized techniques for various nonlinear energy problems, especially the Optimal Power Flow (OPF) problem. Existing solutions for DC-OPF are restrictive, and this discussion will focus on AC-OPF, which is NP-hard and involves quadratically constrained quadratic programming. The proposed solution draws from convex optimization techniques such as DC decomposition, search procedures, and relaxation methods. It also necessitates new theories for algorithmic convergence and global optimality assurance. Given the robustness and speed of linear programming and the familiarity of the power community with convex programs, this algorithm offers a user-friendly solution to AC-OPF. Its efficacy will be assessed through theoretical evaluations on established testbed systems and synthetic datasets. To enhance the robustness of distribution networks post-disasters, we propose a service restoration technique that accounts for the fluctuations and limited availability of generation resources in microgrids. We've developed an optimization model focused on microgrids that prioritizes the restoration of critical loads. Additionally, we've integrated the reserve capacity status into the restoration constraints, ensuring a consistent power supply to the rejuvenated loads.

Biography

Dr. Masoud Barati is an Assistant Professor of Electrical and Computer Engineering at the University of Pittsburgh. He earned his Ph.D. in Electrical Engineering from the Illinois Institute of Technology, Chicago, in 2013. Before joining the University of Pittsburgh, He held a postdoctoral position at the University of Chicago's Booth School of Business. He has also served as an Assistant Professor at both the University of Houston and Louisiana State University. A dedicated member of the "Microgrid Protection Systems" subcommittee under the "Power System Relaying & Control Committee," He has been honored with the IEEE Certificate of Appreciation Award for founding a workshop on Harmonic Power System in collaboration with the S&C Company in the IEEE Chicago section. His research primarily focuses on the application of optimization and applied mathematics in the power system, as well as wide area monitoring and enhancing power system resiliency and recovery.