SDG&E Expands Drone Program to Detect Damage

Following the devastating 2007 wildfires, San Diego Gas & Electric Co. launched its wildfire mitigation program, making significant investments in a wide range of technologies designed to help significantly reduce the chances of a utility-related wildfire. One such technology is the use of unmanned aerial systems, or drones.

Drones have proven highly valuable, enabling the utility to complete a wide variety of missions aimed at not only reducing the risk of wildfires but also increasing the safety of its employees while reducing costs. One of these missions, and perhaps the most critical, is the quick and efficient assessment of power lines and equipment within areas at risk of wildfire.

Preventive Assessments

Every year, San Diego Gas & Electric (SDG&E) uses drones to conduct thousands of routine preventive assessments on its infrastructure to identify and repair damage before an outage, but these assessments become particularly critical following extreme weather conditions, in which the utility may have had to proactively turn off power as a last resort to help protect homes and communities from utility-related wildfire risk. Infrastructure must be carefully inspected for damages prior to reenergizing the lines and restoring power as quickly and safely as possible.

The advantages of drone inspections over ground inspections were quickly realized, including the top-down and close-up views drones provided.

Typically, utilities send field crews to visually inspect equipment for damage, which can require significant personnel hours, increasing costs, the chance of worker injuries and the possibility of damaging the surrounding environment when driving or hiking through sensitive habitats. Utilities, including SDG&E, also have used helicopters, which also comes at a cost. Therefore, SDG&E sought approval from the Federal Aviation Administration to use drones for multiple use cases, including assessments. It was one of the first utilities to receive approval in 2014.

Since then, the company’s use of drones has expanded and the utility developed its Drone Investigation, Assessment & Repair (DIAR) program. The DIAR program includes the deployment of numerous highly advanced drones to assess tens of thousands of poles, power lines and equipment and capture millions of images, which are then uploaded to a central database for review by qualified linemen. The advantages of drone inspections over ground inspections were quickly realized, including the top-down and close-up views drones provided. This helped the utility to identify additional issues to repair, leading to a more effective inspection program. However, SDG&E also realized drones could be used for more applications.

“I knew they were collecting the images, so the question was, ‘Could we apply computer-vision models to help us be more efficient long term in our maintenance and fire prevention efforts?’” said Gabe Mika, strategic technology and investment manager for SDG&E. “Could we use the drone shots to build machine-learning models, and make this kind of drone program not just real, but something that had real benefits for our customers?”

Intelligent Image Processing

Analyzing millions of images is an arduous task and takes time. By using artificial intelligence, the digital acceleration team in the Information Technology department was brought in to see if the process could be streamlined further. Through intelligent image processing, SDG&E is now using artificial intelligence and machine learning to automatically detect damaged assets that could lead to an ignition.

“First, machine learning models were created to detect unique asset and damage conditions from the drone program imagery, including insulator, transformer, wooden pole and crossarm damage,” Mika explained. “We involved qualified electrical workers to train the machine models to identify specific equipment damage by drawing a bounding box around an example of damaged equipment on an image. Then, that image is used to teach the machine software to automatically identify the same pattern of damaged equipment in future images.”

A feedback loop was created within the platform, enabling qualified workers to validate machine learning results to optimize the model performance.

After teaching the image-recognition models to accurately identify the equipment by tagging and confirming the contents of all 2.3 million photos, SDG&E’s team applied their experience to grading the condition of equipment. By adding the insights of experienced linemen upfront, the data and model became far more accurate when analyzing images. And if the technology is unable to analyze the images, the model sends them to qualified technicians for further review.

This helped the utility to identify additional issues to repair, leading to a more effective inspection program.

This technology enables the utility to quickly identify damage that could lead to ignition and helps it to prioritize and initiate work in the field to repair the equipment.The team built a map-based user interface to visualize and browse for imagery and machine learning results stored in the image repository. To facilitate continuous improvement and refinement of the models, a feedback loop was created within the platform, enabling qualified workers to validate machine learning results to optimize the model performance.

AI & Machine Learning Models

Since officially kicking off in April 2020, the project team has built more than 35 artificial intelligence and machine learning models. Each one of these models is designed to automatically identify an SDG&E asset and identify a specific type of damage or specific type of condition.

By adding the insights of experienced linemen upfront, the data and model became far more accurate when analyzing images.

The multi-departmental team, together with the asset management and operations departments, has changed the utility’s approach to managing the immense electric infrastructure portfolio. Intelligent image processing is truly groundbreaking and promises to be a game-changer in SDG&E’s mission to mitigate wildfire risk, minimize outages for customers and ensure the safety of the communities it serves.

The DIAR program is just one of many wildfire mitigation projects included in SDG&E’s wildfire mitigation plan filed annually with the California Public Utilities Commission. The plan outlines ongoing practices and additional improvements SDG&E will make to mitigate wildfire threats.

Editor’s note: For more information on SDG&E’s 2022 plan, visit sdge.com/2022-wildfire-mitigation-plan.

Tom Fries has been with SDG&E Aviation Services for over 5 years. Prior to coming to SDG&E, Tom worked for General Atomics Aeronautical with program and engineering teams. Tom is a retired Marine with 22 years of active duty as an F/A-18 pilot. Tom has his undergraduate degree in Mechanical Engineering Technology from Temple University, master’s degrees from both Georgetown University and University of San Diego, and Aviation Safety and Security certificate from University of Southern California.