Accelerating Transmission and Distribution (T&D) Inspections with Drones and Artificial Intelligence

Posted to EPRI in the Transmission Professionals Group
image credit: This image is owned by EPRI.
Jeremy Renshaw's picture
Senior Program Manager Electric Power Research Institute (EPRI)

Dr. Jeremy Renshaw is the Sr. Program Manager for Artificial Intelligence at the Electric Power Research Institute (EPRI) and has been with EPRI since 2012. Dr. Renshaw manages the AI.EPRI...

  • Member since 2021
  • 13 items added with 11,112 views
  • Jul 7, 2021

Utilities collect millions of images of their infrastructure to assess equipment health and support reliability investments, but manually reviewing inspection imagery and data is time consuming, subjective, and costly. With the widespread adoption and use of unmanned aerial systems (UAS) to collect images of T&D infrastructure, the volume of electric power utility imagery has increased exponentially and rapidly outpaced staff’s ability to analyze and evaluate it. 

There are AI solutions, however, that utilities can use to automate both collection and review of inspection imagery. While images of healthy assets are significantly more common, certain components – or types of damage to those components – are much less common and, therefore, more challenging to acquire and label, but are some of the most important images for training AI models. Another important part of the process is image classification, and machine learning (ML) models typically need thousands of images per damaged image class. With hundreds – and potentially thousands – of damaged image classes, this could require one million or more images for training ML models, some of which are the rarest and most challenging images to collect.

To efficiently recognize and label T&D components, an extensive database of labeled images is required. EPRI is constructing a curated, labeled industry database of T&D assets that will help mitigate the time-consuming and costly task of reviewing and labeling the imagery.  Similarly, generative adversarial networks (GANs) can also be used to generate synthetic images to augment available datasets for training.

Improved inspection and review efficiency will accelerate problem identification, which can facilitate the replacement of damaged assets, providing regulators better documentation, and could improve communications among utility stakeholders – including customers. And all of this will lead to a safer, more reliable, and more affordable electric grid.

EPRI has been working on a concerted effort throughout 2021 to engage the AI community to work with the electric power industry on multiple projects, including this effort to improve T&D asset inspection.

Learn more about the transformative solutions generated by the convergence of artificial intelligence and the electric power industry at EPRI’s AI and Electric Power Summit on September 28-29, link here.


Jeremy Renshaw's picture
Thank Jeremy for the Post!
Energy Central contributors share their experience and insights for the benefit of other Members (like you). Please show them your appreciation by leaving a comment, 'liking' this post, or following this Member.
More posts from this member


Spell checking: Press the CTRL or COMMAND key then click on the underlined misspelled word.
Matt Chester's picture
Matt Chester on Jul 7, 2021

It's really cool to see what drones can do to empower AI, along with appropriate sensors. Do you foresee a future where we level up even further with satellites looking at infrastructure? 

Jeremy Renshaw's picture
Jeremy Renshaw on Jul 14, 2021

Matt, great question - the answer is a resounding YES! We may not yet have the resolution to evaluate individual assets like we do with close up images from drones, but we do plan to use satellite imagery for applications like vegetation management, evaluation of wildfire risk, and several more. EPRI is in the process now of gathering satellite data that we plan to openly and freely share with the industry for these (and other) use cases. Standby for a potential future post on this topic!

Ronald Burton's picture
Ronald Burton on Aug 7, 2021

Hacking the Transmission Grid is a major concern of the electric utility, but the use of Drones to destroy the electric grid is catastrophic in my opinion. Here is an article that exposes this threat.
I think one solution would be to set up an IoT (Internet of Things) sensor shields around Electric power substations and switching yards detecting the drone's rotor and motor frequencies detecting their location around these facilities. Data from the shield would alert electric utility of a drone in an unauthorized area. I think this is a major concern.

Jeremy Renshaw's picture
Jeremy Renshaw on Aug 12, 2021

Ronald, thanks for the comment.  This is certainly an item of concern for both intentional and accidental interference from drones with T&D equipment.

Get Published - Build a Following

The Energy Central Power Industry Network is based on one core idea - power industry professionals helping each other and advancing the industry by sharing and learning from each other.

If you have an experience or insight to share or have learned something from a conference or seminar, your peers and colleagues on Energy Central want to hear about it. It's also easy to share a link to an article you've liked or an industry resource that you think would be helpful.

                 Learn more about posting on Energy Central »