How AI can enable the push to bury powerlines 

Yogev Shifman, Exodigo

While the perils of above-ground powerlines are well documented – sparking countless disastrous (and all-too-often deadly) wildfires, adding unsightly additions to formerly picturesque hillsides and causing major wildlife and biodiversity shifts – there has not been as much action to overcome these obstacles. Maybe that is because there has not been an effective way to both strengthen our utility infrastructure and manage all of the associated environmental, societal and operational headaches associated with trying to make major changes in how power is delivered … until now. 

But before we dive into what can be changed and where we are headed, let’s dive deeper into why the current infrastructure is broken. 

Power infrastructure – A story of unloved lines

While our need for reliable utilities remains undisputed, power lines are often the cause of heated debates in any community they transverse. To support millions of miles of distribution lines across the United States, the U.S electric transmission network consists of approximately 700,000 circuit miles of lines and 240,000 miles of high-voltage transmission lines. These lines transfer electricity from central generating stations to local distribution networks, where a patchwork network of regional monopolies oversee energy delivery. While that may seem like a lot of lines (and a lot of powerful monopolies), the current infrastructure still is not enough and it is certainly not immune to service interruptions and issues. 

Existing lines simply can’t support surging energy needs or the next phase of clean energy – think wind, solar, hydroelectric, electric vehicles, etc. Adding to the distribution challenge, power production centers – be they hydro, wind, solar, etc. – are not usually close to the urban centers needing the most power. Moving massive amounts of solar-generated electricity from the desert or wind farms set high in the hills or offshore to urban centers and rural communities is no small task! 

Unfortunately, adding more lines also means dealing with more local battles as rural and suburban communities generally cringe at the thought of more towers, more lines and more eyesores. And perhaps their pushback is justified, as adding more lines also creates more risk. 

 Power lines have sparked devastating wild fires for years – one of the most well-known being the Dixie Fire that burned hundreds of thousands of acres across five Californian counties. While not all line fires create raging forest fires, they are concerningly common and are estimated to occur hundreds of times each year. Unfortunately, as climate change causes drought conditions to continue to escalate, so does the risk of power line fires jumping to dry forest areas and causing massive fires – more lost lives, homes and acres of forest are at risk than ever before. 

So if we need more power, and more ways to safely distribute it to more locations, isn’t that just calling for even more risk? If we don’t rethink how utilities move power across the country, then the answer is a resounding yes. 

The next power play – What needs to change

The need for expansive lines covering hundreds of thousands of miles certainly won’t change anytime soon, but there are steps we can take as an industry to reduce wildfire risks, while also preparing for future infrastructure needs. Placing lines underground is a highly effective wildfire mitigation strategy. Once buried, the risk of power lines starting massive fires plummets. There is also a clear call to improve the line transmission network across the United States to accommodate continued power needs and prepare for renewable power capacity, which could also be supported by buried lines. However, despite declaring the need to make major changes, including burying key lines, utility companies have been slow to actually act.

While burying lines seems like an obvious solution, it has traditionally been an extremely costly endeavor. Burying lines comes with its own set of cost and environmental challenges. In fact, it can cost as much as ten times more than installing aboveground lines. Blasting through granite, digging trenches and burying lines across unmapped rural areas or even through established urban centers is certainly no small feat. It requires an immense amount of planning before even one small shovel breaks ground to limit the risks that start once the project is in motion – cut lines, spills, explosions, emissions from heavy equipment … they all have the potential to create massive safety shockwaves and environmental concerns.

Navigating regulatory barriers, land restrictions and diverse terrain to bury lines would be incredibly tough, but possibly the hardest part? Knowing what they are going to hit when they dig! The underground is full of unknowns – including variables that could completely devalue the entire project like unexpectedly high water tables near coastal areas or contaminated soil/water pockets that would make the land unusable, or an unknown natural gas pipe that could cause immediate construction dangers. 

Understanding what you will encounter when you break ground and where to dig can limit the number of unexpected events, keeping costs and safety in check. Unfortunately, most companies can’t access reliable maps of the underground. Traditional underground discovery techniques are costly, outdated and astoundingly inaccurate. While most would think the underground might be quiet, it is noisiest, and therefore one of the most complicated, frontiers for imaging. All signals that are affected by the underground suffer from highly noisy environments, a major dynamic range, and completely different normalizations.

AI and advanced sensors

Luckily, many of the elements that have made burying lines cost prohibitive in the past can now be overcome thanks to critical technology advances. Artificial intelligence (AI) breakthroughs are empowering the utility industry to better plan projects and decrease risks and costs by providing access to accurate, easy-to-understand maps of what anomalies or potential hazards lie below the surface.

Expertise that was previously limited to the geophysical arms of major oil & gas and mineral exploration companies is now within reach of companies of all sizes throughout the utility and construction communities. While industry giants have been able to afford large, technologically advanced teams to create detailed underground maps, AI has unlocked that power for the masses at scale – adding data analysis capabilities that surpass what human minds can handle. With new AI solutions that make sense of the massive amount of data generated from a collection of extremely sensitive sensors, utilities can accurately map every type of underground asset and layer across all terrains. 

Combining all imaging signals (seismic vibrations, magnetics, radar, electromagnetic, etc.) into a single normalized matrix, AI has finally cracked the code for low-SNR signals in noisy environments. With artificial intelligence, utilities can rely on one easy-to-read, comprehensive map from all the disparate data signals. And that means that utilities no longer have an excuse for not exploring the potential of buried lines, which would improve the safety (and strength) of our power infrastructure. 


About the Author

An active advocate for more creative, accurate and sustainable underground vision systems built on advanced artificial intelligence, Yogev Shifman currently serves as co-founder and chief product officer at Exodigo. Prior to Exodigo, he led software engineering and technical product teams for Buildots, an AI company serving the construction industry. A highly decorated former Israeli intelligence officer, Shifman previously served as R&D Section Head in one of Israel’s most elite deep tech units, where he led numerous award-winning R&D and product development programs for sophisticated real-time embedded systems, involving both hardware and software. A graduate of Technion – Israel Institute of Technology with a degree in computer software engineering, he also earned an MBA in entrepreneurship and technology from Tel Aviv University after starting his career at Intel Corporation.

Trump’s ‘unpredictability’ shakes investors – This Week in Cleantech

This Week in Cleantech is a weekly podcast covering the most impactful stories in cleantech and climate in 15 minutes or less.
wind turbines in front of an orange sunset

Renewables permitting has been ‘paralyzed’ by Trump – This Week in Cleantech

This Week in Cleantech is a weekly podcast covering the most impactful stories in cleantech and climate in 15 minutes or less.