3 AI technologies innovation leaders must consider for utility asset management (Part 2: Drones)
Drones have seen a meteoric rise in popularity for distribution asset management. We explore the pros and cons.
Drones have seen a meteoric rise in popularity for distribution asset management. We explore the pros and cons.
In the introduction to this series of articles, we touched on the complexity of managing the power grid and how stricter regulations and compliance standards urge both public utilities and investor-owned utilities to innovate. We then introduced the 4 buckets of emerging technologies to consider for a viable and scalable distribution utility asset management strategy. If you missed the introduction article, you can find it here: “4 AI techs utility leaders must consider for utility asset management (Part 1: Introduction)”
In this first article, we will be discussing drones.
In recent years, drones have experienced a meteoric rise in popularity as an indispensable utility asset management tool. Equipped with advanced cameras, sensors, and data analytics capabilities, drones can swiftly and safely inspect assets in challenging and remote locations, providing real-time insights to maintenance teams. In some situations, drones can dramatically decrease the need for manual inspections and subsequently reduce operational costs and enhance overall efficiency. As the technology continues to evolve, businesses are increasingly integrating drones into key parts of their asset management strategies.
However, are drones the best solution for managing distribution utility assets? Remember, the key here is technologies that will ultimately scale to cover large numbers of assets. To evaluate this, we will consider the following four categories: data quality, cost, ease of use and scalability.
One major upside of drones is their ability to capture imagery from various angles. This capability provides a comprehensive and detailed view of assets. Drones also excel in capturing high-fidelity, "per-asset" data. By this, we mean that the captured data is detailed enough to associate specific components with their corresponding assets. For instance, with the right data analysis tools, they can accurately pinpoint and associate transformers in relation to their respective poles, facilitating more accurate asset tracking and maintenance.
Additionally, drones can be equipped with different sensors to capture information beyond the RGB space. LiDAR and thermal cameras enable the ability to perform advanced inspection and engineering tasks. These sensors allow you to perform more nuanced and detailed assessments which can include 3D reconstruction, geospatial data collection and heat anomaly identification. More detailed data leads to increased situational awareness, all of which further contribute to proactive maintenance and risk mitigation strategies.
When it comes to leveraging drone technology, innovation leaders at utilities have two options: 1) purchasing and operating their own fleet of drones, and/or 2) leveraging third parties that provide drones-as-a-service.
If you opt to start your own fleet, you’ll have to budget for the upfront investment of buying the drone along with accompanying cameras and sensors, training your field operations teams on how to operate drones, obtaining relevant FAA drone pilot certifications, and extracting and processing the data collected on your grid.
For one drone, expect an average per unit price of around $20,000, which does not include the cost of purchasing data analytics software and training your employees. Depending on the scope of utility asset management work desired, costs of buying, operating and maintaining the drone can quickly ramp up beyond $30,000 the first year.
An alternative is to hire a service provider that specializes in drone-mediated utility asset inspection. A mid-tier drone service will generally cost between $500 and $1,500 per hour per drone for basic imagery and video collection. If you need a more advanced drone service with higher image resolution, video capabilities and more advanced features such as 3D mapping, thermal imaging or LiDAR, you’ll be looking at $1,500 to $5,000 per hour per drone. These costs can quickly add up when using drones for something as expansive as the distribution grid, which can cover hundreds of thousands of miles. Additionally, although there are many firms that specialize in drone-mediated data collection, they do not necessarily have the capability to process the data to generate meaningful insights.
Although autonomous drones (“drone-in-a-box”) with programmable, AI-mediated flight paths and end-to-end data extraction and analysis are increasing in popularity, they are best suited for localized site inspection and monitoring. For example, energy generation facilities such as solar farms and power plants are great candidates as the drones are confined to a certain travel radius.
The distribution grid is a different matter. The sheer number of assets that need to be inspected and monitored make it difficult to automate asset management processes via drones. As referenced in the section above, if you go down the route of purchasing your own fleet, you will not have a simple “plug-and-play” system. You will need to train designated drone operators as well as put data extraction and analysis systems in place.
Even with trained drone operators and a fleet of drones at your disposal, the scalability of this technology for managing and inspecting the distribution grid is constrained by current regulations set forth by the Federal Aviation Administration (FAA) and the limitations imposed on Beyond Visual Line of Sight (BVLOS) operations. Currently, the FAA requires most drone flights to be conducted within the operator's line of sight, limiting the range and coverage area for distribution grid inspections. Although it is possible to obtain waivers that allow for drone operation beyond the operator's visual range, these are subject to strict regulatory hurdles due to safety concerns.
Another scalability challenge is the presence of no-fly zones, especially in urban areas. Drone operation near airports, government buildings, hospitals, and crowded public spaces are strictly prohibited due to security and safety concerns. As a result, utilities looking to leverage drones for asset inspections in these areas face significant roadblocks as they must navigate complex airspace restrictions and seek specific permissions for each flight. The strict enforcement of no-fly zones underscores the need for careful planning and adherence to local regulations, making the integration of drones in city environments a challenging proposition for widespread distribution grid management and inspection applications.
Lastly, another consideration is the fact that drones have limited battery life. Continuous data collection with a variety of sensors is very energy intensive and with current battery technologies, most commercial use drones can only last for 45 minutes to an hour before needing to be recharged. This could be difficult when dealing with the distribution grid as overhead inspection routes often cover broad areas.
Drone-mediated distribution asset inspection offers a multitude of safety benefits. Drones can access remote and hard-to-reach locations, eliminating the need for manual inspections that often involve working at heights, in confined spaces, or in proximity to potentially dangerous equipment. By deploying drones, companies can mitigate risks to human inspectors and minimize the likelihood of accidents and injuries.
Pros:
Cons:
Overall, while drone-mediated distribution asset inspection offers valuable safety benefits and detailed data insights, there are challenges in terms of cost, scalability, and regulatory constraints that need to be carefully considered when integrating drones into utility asset management strategies.
Interested to learn more about alternative technologies for distribution asset inspection? Drop us a line here for a free consultation to explore a low cost, low risk AI-powered utility asset management pilot using fleet vehicle-mounted cameras.