ComEd launched the Advanced Image Analytics program to build a more effective and proactive maintenance program.
ComEd is combining drone powerline inspection and computer vision AI analytics to build a more effective and proactive maintenance program. The program leverages new and existing data sources, employs advanced analytics and empowers the workforce to more rapidly and effectively identify and repair conditions before they are service impacting. You can download the case study here, which walks through the Advanced Image Analytics program ComEd is implementing to leverage these new technologies to augment and enhance their asset inspection capabilities.
Embracing New Technology, Delivering New Capabilities
The ComEd Advanced Image Analytics program utilizes drones, vehicle-based camera systems, and mobile devices for image capture with AI processing for automated defect detection. These capabilities will improve maintenance efficiencies and provide ComEd with deeper, real-time insights into the state of health of all distribution assets. The program implementation focused on adding capabilities in the following areas:
- Drone data capture
- Visual data management
- Visual data collection program
- AI Algorithm development
- AI development and deployment
- Image analytics & management
- Professional program management
“Optelos delivers a mature, flexible and advanced image storage and inspection platform that is well-designed to pair with artificial intelligence algorithms. ComEd has formed a valuable partnership with Optelos in helping maximize their platform for efficient utility operations.”
Jeff Birkmeier, Senior Manager of Strategic Planning, ComEd
The full implementation of the program also included building out drone data collection capabilities to automatically ingest autonomously gathered inspection imagery and effectively organize and store it for fast and reliable access.
The GUI for the Optelos image analytics platform is mapped to ComEd’s serving territories, making it easier for maintenance crews to adopt and reliably find targeted assets within the system.
Bringing It All Together
The Image Analytics Platform was established and rolled out to the organization in 2023, but there are several additional steps involved in realizing the full potential of this powerful new system:
- Increased platform adoption
- Systems integrations and workflow operationalization
- AI algorithm accuracy and expansion of classifiers
- Expand applications into additional high value use cases
ComEd developed AI algorithms to automatically classify and identify defect conditions on Distribution assets.
Additional Use Cases
ComEd is also exploring new departmental use cases for their centralized visual asset database and AI management platform (Optelos).
- Nameplate Data – gather information, such as manufacturer or serial number, from equipment nameplates to resolve any data gaps that exist.
- Location Discrepancies – identify assets that are recorded in the system of record as being at a certain LAT/LONG, but are observed in the field at a different position
- Environmental – Identifying Leaks – utilize AI to identify leaks before they expand to the surrounding environment and require extensive clean-up costs.
- Vegetation Management – Encroachment Identification – determine the distance between vegetation and conductors; identify areas where vegetation is too close.
- Third Party Attachments – Attachment Audits – verify the location of TPAs and identify attachments that are not currently recorded for billing validation.
- Equipment Model Locating – Finding outdated models of fuse cutouts, lightening arresters, and transformers with PCB labels on them.
An additional use case for power line drone inspection data is third party attachment validation utilizing computer vision AI. This process can utilize previously captured drone inspection data to verify the location of TPAs and identify attachments that are not currently recorded.
What’s Next
ComEd now has all elements of the Grid Analytics advanced image program in place and ready to deliver long term improvements to grid reliability and resiliency. The program can now serve as a foundation for multiple other applications and use cases. While ComEd is still early in its development of the Image Analytics Platform (Optelos), the potential is exciting and gaining traction within the larger organization. We will update this use case as ComEd gains further insights into its expanded applications and capabilities.