*********************************
There is now a CONTENT FREEZE for Mercury while we switch to a new platform. It began on Friday, March 10 at 6pm and will end on Wednesday, March 15 at noon. No new content can be created during this time, but all material in the system as of the beginning of the freeze will be migrated to the new platform, including users and groups. Functionally the new site is identical to the old one. webteam@gatech.edu
*********************************
Title: Large Scale Data Analytics for Resilience of Energy Networks
Committee:
Dr. Chuanyi Ji, ECE, Chair, Advisor
Dr. Scott Ganz, School of Business, George Town University
Dr. Mark Davenport, ECE
Dr. Deepak Divan, ECE
Dr. Valerie Thomas, ISyE
Abstract: Massive power failures are induced frequently by natural disasters in a changing climate. Two fundamental challenges arise in face of such failures: First, how recovery can be resilient to the increasing severity of disruptions and their impact on service users in a changing climate. Second, how can we measure the impact of failures and recovery and its heterogeneity on customers with different characteristics. 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, recovery degrades with the severity of disruptions: large failures that cannot recover rapidly increase by 30% from the moderate to extreme events. Prolonged small failures dominate entire recovery processes. Further, our analysis demonstrates the promise of mitigating the degradation by enhancing recovery of a small fraction of large failures through distributed generation and storage. Next, a dynamic resilience metric is developed using spatiotemporal failure and recovery processes incorporating the cost imposed on customers. The resilience metric is then combined with inference to design a framework on how to study the dynamic cost and its heterogeneity on customers with different characteristics. Our framework is validated on large scale data from multiple states in the US.
