Prof. Dan Bienstock from the Department of Industrial Engineering and Operations Research and Prof. Gil Zussman take part in a project awarded to Los Alamos National Lab (LANL) within the framework of the DOE Grid Modernization Laboratory Consortium (GMLC) (see news item). The $3M project titled “Advanced Machine Learning for Synchrophasor Technology” is being led by Dr. Michael Chertkov and includes co-PIs from Pacific Northwest National Laboratory (PNNL) and Lawrence Berkeley National Laboratory (LBNL).
The main goal of this project is to develop a suite of new machine learning tools to monitor the stochastic and dynamic state of the power transmission grid during its normal operations and to localize significant frequency events in seconds after they occur. To achieve this goal, the interdisciplinary team will focus on combining a number of technical ideas to design a Grid-Modeling aware machine learning toolbox that utilizes and develops: (a) advanced optimization and computational methods and algorithms for machine learning and data analytics; (b) state-of-the-art, industry-grade frequency monitoring software; (c) phasor measurement unit (PMU) measurements at the transmission level; (d) aggregated micro-synchrophasors (uPMU) measurements at the distribution level; and (e) modern map-visualization tools and approaches.
The contribution of Zussman’s group to the project will build on the recent work in the area of resilience of power grids, performed over the past few years mostly by Ph.D. student Saleh Soltan.
