102: Exelon Spans Space and Time to Pinpoint Outages with Spatiotemporal Data Analytics
Date & Time
Wednesday, November 1, 2023, 10:00 AM - 10:45 AM
Po-Chen Chen Vladyslav Anderson

Efficient outage management is a high priority for utilities. Faults in the distribution grid system may include multiple causal factors, including equipment failure and structural damage due to weather-related events.

Exelon built a novel tool using geospatial clustering methods that facilitate the evaluation of available outage related data to pinpoint the true fault damage location. Precise fault location data enables reliability engineering teams to gain insights through trending reliability and asset health concerns at a hyperlocal scale. Engineers can use this information to identify correlation of trouble zones in the same geographic area regardless of whether multiple feeders supply them.

Session Takeaways:

  • Presentation of a novel geospatial clustering framework to pinpoint the fault location of an outage based on outage data, asset GIS data, and crew vehicle telematics data.    
  • Demonstration of how data processing and cleansing are performed and how the insights are observed from the correlation of various datasets.
  • Development of DBSCAN and optimization algorithms to identify the fault location based on the clustering results.
  • Deployment of a web application tool for stakeholder to use. 
  • Discuss the challenges on ongoing progress with building and deploying a scalable optimized geospatial clustering tool for various sizes of feeders in BG&E.
Location Name
Ligurian II
Full Address
Loews Portofino Bay Hotel at Universal Orlando
5601 Universal Blvd
Orlando, FL 32819
United States
Focus Area
Data Science, Grid Analytics