Our research is on large-scale data analytics and machine learning for networks. Our research seeks real-world applications important to our society. These problems motivate new methodologies on data analytics and learning in a networked setting. Recent work includes
- Resilient energy networks and impact on people,
- Spatial temporal models for learning from micro data at scale.
In collaboration with colleagues, service providers and policy makers, our research uses field data from large geographical regions in the United States. We welcome collaboration.
Recent publication and news
- IEEE PES Task Force, AM Stanković, KL Tomsovic, F De Caro, M Braun, JH Chow, N Äukalevski, I Dobson, J Eto, B Fink, C Hachmann, D Hill, C Ji, JA Kavicky, V Levi, CC Liu, L Mili, R Moreno, M Panteli, FD Petit, G Sansavini, C Singh, AK Srivastava, K Strunz, H Sun, Y Xu, S Zhao, “Methods for Analysis and Quantification of Power System Resilience,” IEEE Transactions on Power Systems, Early Access, P1-14, Oct. 2022 [PDF]
Abstract: This paper summarizes the report prepared by an IEEE PES Task Force. Resilience is a fairly new technical concept for power systems, and it is important to precisely delineate this concept for actual applications. As a critical infrastructure, power systems have to be prepared to survive rare but extreme incidents (natural catastrophes, extreme weather events, physical/cyber-attacks, equipment failure cascades, etc.) to guarantee power supply to the electricity-dependent economy and society…
-
Amir Afsharinejad, Chuanyi Ji and Robert Wilcox, “Large-scale data analytics for resilient recovery services from power failure,” Joule – Cell Press, Volume 5, Issue 9, P2504-2520, Sept. 2021 (featured article) [PDF]
Abstract: Massive power failures are induced frequently by natural disasters. A fundamental challenge is how recovery can be resilient to the increasing severity of disruptions in a changing climate. We conduct a large-scale study on recovery from 169 failure events at two operational distribution grids in the states of New York and Massachusetts. Guided by unsupervised learning from non-stationary data, our analysis finds that under the widely adopted prioritization policy favoring large failures, recovery exhibits a scaling property where a majority (∼90%) of customers recovers in a small fraction (∼10%) of total downtime. However, …
News:
- Warmest and belated congrats to Amir for becoming Dr. Amir Afsharinejad in Dec. 2021.
- A recent feature article by the editors at Nature Energy 2021 selects our paper 2016 as one of the favorites published at Nature Energy in the past five years. (ECE Gatech News.)