Title: Improved Monitoring of the State and Topology of Distribution Systems
Abstract: As the solar photovoltaic (PV) units connected to distribution systems rapidly increase, their dispatch, control, and protection require timely and accurate monitoring. Unlike transmission systems, distribution systems historically lack enough sensor measurements, making their real-time monitoring almost impossible. Recent deployment of diverse types of devices such as phasor measurement units (PMUs), smart meters, solar inverters and weather information sensors opens up new ways of monitoring these systems, with the assistance of customized machine learning (ML) tools.
In this talk, we will describe an approach which will render distribution systems fully observable, such that the hosting capacity for solar generation can be accurately estimated, and unnecessary solar curtailments can be avoided. The talk will present the developed grid-model-informed ML tool which integrates heterogeneous data streams and creates synchronous measureme
nt snapshots for the state estimator(SE) which is designed as a hybrid robust SE which provides not only accurate state estimates but also real-time feedback for ML model refinement.
One of the shortcomings of machine learning applications to power systems is that they either completely disregard network models or simply assume a fixed network topology. As a result they fail to produce fully satisfactory results under network model changes which are frequent in feeder operation. In addition, most of them do not receive proper supervision from power system domain knowledge, leading to reduced efficiency and reliability. This deficiency will be addressed by using a combination of machine learning methods which account for topology changes and feedback on the accuracy of the predicted measurements from a robust state estimator. As a separate tool, a sparse estimator which can successfully detect power system events based on a limited number of phasor measurement units will also be described. Developed tools will provide comprehensive situational awareness (both steady-state operation and during dynamic events) to grid operators by the help of three modules: graph-learning measurement synchronizer; hybrid robust state estimator and sparse event identifier. The talk will present examples showing application of these modules under simulated scenarios.
Bio: Ali Abur is currently a Professor in the Department of Electrical and Computer Engineering at Northeastern University, Boston. He obtained his B.S. degree from Orta Dogu Teknik Universitesi, Ankara, Turkey and his M.S. and Ph.D. degrees from The Ohio State University all in Electrical Engineering. After receiving his PhD degree, he joined Texas A&M University and worked as a Professor in the Electrical Engineering Department until November 2005. He then moved to Boston and served as the Chair of the Electrical and Computer Engineering Department at Northeastern University, Boston until 2013. His research and educational activities have been in the areas of electric power systems modeling, state estimation, detection and identification of errors in the network model and measurements, fault location and electromagnetic transients modeling and simulations.
Join from a PC, Mac, iPad, iPhone or Android device:
Please click this URL to start or join. https://iastate.zoom.us/j/94087524568?pwd=WXc4eWtLaWJJKzNpcHZydlptOUZIZz09
Or, go to https://iastate.zoom.us/join and enter meeting ID: 940 8752 4568 and password: 324328
Join from dial-in phone line:
Dial: +1 312 626 6799 or +1 646 876 9923
Meeting ID: 940 8752 4568