Evaluating the Effectiveness of VIIRS Day-Night Band Data for Power Outage Impact Assessment

Authors: Bandana Kar*, Oak Ridge National Laboratory, Olufemi A. Omitaomu, Oak Ridge National Laboratory, Susan Kotikot, Oak Ridge National Laboratory, Caleb Robinson, Georgia Tech University, Fernando Runte, Sam Houston State University
Topics: Geographic Information Science and Systems, Hazards, Risks, and Disasters
Keywords: Impact Assessment, Power Outage, VIIRS, Quality and Accuracy
Session Type: Paper
Day: 4/3/2019
Start / End Time: 4:30 PM / 6:10 PM
Room: Forum Room, Omni, West
Presentation File: No File Uploaded


Although weather related events cause power interruptions and complete power failure, power outage could also occur due to damage to transmission lines, substations, cyberattack of a grid system through hackers, or failure of a simple software system that controls the load re-distribution system as was the case for 2003 Northeast Blackout. Whatever is the cause of a power outage, it is crucial to forecast spatio-temporal distribution of impacted population to help utility companies undertake restoration efforts in a timely manner. Following hurricane Maria, VIIRS Day and Night Band data was used as proxy to estimate impacted customers in Puerto Rico in near real-time. In this project, the derived customer data from VIIRS was compared with a Power Outage Model and power outage data provided by utility companies that is being used in DOE’s EAGLE-I for situational Awareness to determine the suitability of VIIRS data for customer assessment and the quality of the estimated customer data. Given that the failure on the part of utility companies to report power outage data would impact the reliability of EAGLE-I, the framework developed in this project will generate complementary power outage data to help with response and restoration efforts and decision-making activities.

Abstract Information

This abstract is already part of a session. View the session here.

To access contact information login