Wenting Li, Rensselaer Polytechnic Institute, gives CURENT Power and Energy Seminar on Fri., Nov. 1
Wenting Li, Rensselaer Polytechnic Institute (RPI), will give the CURENT Power and Energy Seminar on Fri., Nov. 1. Her talk is entitled Real-time and Agile Data-driven Approaches Enabling Power Grids to be Smart. The seminar will be held in MHK 404 from 12:20 pm - 1:10 PM. Wenting Li will present from New York.
Time: Friday, November 1st, 12:20 PM - 1:10 PM EST
Location: Min H. Kao Building, Room 404
This seminar will be available through ZOOM. See info near bottom of this email.
Presenter: Wenting Li, Rensselaer Polytechnic Institute
Title: Real-time and Agile Data-driven Approaches Enabling Power Grids to be Smart
Abstract: Nowadays, the power system stability and reliability are challenged by the variable renewable energy and the frequent power outages. One promising remedy is to develop data-driven algorithms for power system intelligence, but practical deployment is not universal. One of the main concerns is the vulnerability of pure data-driven algorithms applied to variant power systems conditions. In this talk, I will present three online data-driven approaches based on phase mea-surement units (PMU) data to augment power system intelligence in monitoring. These methods capture power system characteristics and are robust to different power systems situations.
Based on the latest study of the low-rank property of PMU data, the first method is to identify different types of events. The central idea is to characterize events by low-dimensional subspaces, which are highly related to system dynamics and robust to initial conditions. Then we propose to further identify successive events to avoid cascading failures. The chal-lenge is the identification of a subsequent event that occurs when the system is undergoing the disturbance of a previous event. Also, successive events are not sufficient for training a classifier. Our approach consists of three novel techniques to handle these issues: extract dominant features, train a two-path convolutional neural network (CNN) classifier, and present a subtraction-prediction process. Next, we consider locating the faulted line in a large-scale power network with partial measurements. We extract features based on fault current analysis and train a CNN classifier to locate the faulted line. A significant aspect is that under very low observability (7% of buses), the algorithm is still able to localize the faulted line to a small neighborhood with high probability. To further improve the location performance, a joint PMU placement strategy is proposed and validated against other methods.
Bio: Wenting Li received the B.E. degree from Harbin Institute of Technology, Heilongjiang, China, in 2013. Currently, she is a Ph.D. candidate in Electrical Computer and System Engineering at Rensselaer Polytechnic Institute, Troy, NY. Her research interests include high-dimensional data analytics, feature extraction, application of machine/deep learning to power grids concerning event identification, fault location, and load disaggregation.
Zoom Information:
Join from PC, Mac, Linux, iOS or Android: https://tennessee.zoom.us/j/608630490
Or iPhone one-tap (US Toll): +16468769923,608630490# or +16699006833,608630490#
Or Telephone: Dial: +1 646 876 9923 (US Toll) +1 669 900 6833 (US Toll) Meeting ID: 608 630 490
International numbers available: https://zoom.us/u/aFK5Bq5SR
Or an H.323/SIP room system: H.323: 162.255.37.11 (US West) or 162.255.36.11 (US East) Meeting ID: 608 630 490
Upcoming Seminars
Friday, November 8th - No seminar (Site Visit)
Friday, November 15th - Dr. Sid Suryanarayanan, Colorado State University - ToU, or not ToU, that is the question - Maximizing Wind Energy Usage in Regions with Electric Vehicle Fleets
Friday, November 22nd - TBA
Friday, November 29th - No seminar (Thanksgiving)
See the CURENT calendar for more news and seminars
