Analysis
Sponsored
Grid edge
AI

Sense CEO: To get AMI 2.0 right, the devil is in the details

To use the new wave of smart meters to its full potential, Mike Phillips said, utilities need to invest in the right architecture.

Listen to the episode on:
Apple Podcast LogoSpotify Logo

Photo credit: Sense

Photo credit: Sense

After a decade and a half since their initial roll-out, utilities are exploring replacements for their smart meter networks.

  • The top line: Equipped with artificial intelligence and machine learning capabilities, this new wave of smart metering infrastructure — or AMI 2.0 — comes with the promise of bill savings and grid intelligence. But as Mike Phillips, CEO of smart meter company Sense, outlined on a recent episode of The Carbon Copy podcast, utilities need to invest in the right architecture to make sure the potential doesn’t go untapped: high resolution data, enough computation, and the ability to have real-time networking. 
  • The market grounding: Utilities rolled out the first wave of smart meters back in 2009 with high hopes that they would support a digital grid and save customers money. They were mostly a disappointment, in part because they didn’t deliver on their promise of visibility on the grid.
  • The current take: “Most people think of meters just as data collection devices,” Phillips told host Stephen Lacey. “You have to start to change that mindset, and once you start to think of this as a distributed platform — not just a data collection device — this entire world of making use of machine learning at the edge starts to get opened up.”

The key issue with the first generation of smart meters, according to Phillips, was that the architecture didn’t allow for real time visibility on the grid. The smart meter collected low resolution data and sent it to the service provider to be made available at some point in the future.

Phillips compared the technology to a hypothetical Google Maps, with a phone collecting the user’s location data on 15-minute intervals, sending it in batches to the telecom provider, and only passing it along to Google Maps a day later to use.  

“What would [the point of] Google Maps on your phone be?” he said. “It would be like a static map and maybe a monthly historical report of your traffic on your route to work and maybe compare your neighbors how you drive compared to your neighbors. But would you use that app?” 

In theory, AMI 2.0 is a chance to solve those flaws.

Sense is not an impartial observer. The company built an energy monitor that gives real-time information on every device in a home using machine learning techniques. In 2019, Sense partnered with smart grid technology producer Landis+Gyr to put its technology in smart meters. 

National Grid has started installing Sense-enabled meters in New York, with the goal of installing them for all customers by the end of 2027, at a rate of 5,000 installations per week. The company told local outlet Spectrum News 1 this month that it has installed 170,0000 meters so far. 

Another meter maker, Utilidata is also bringing AI to smart meters for real-time data collection and analysis.

“The most basic thing is to get high resolution data of voltage and current in the grid,” Phillips said. “This means continuous sampling of these voltages and current waveforms…There's meters on the market that are doing that at 15,000 times per second. So that's 50 million times more data than AMI 1.0.” 

To make use of that high-resolution data, Phillips said, utilities need enough computational capabilities to run software in the meters, the ability to have real-time networking, and the software to protect their consumers’ privacy. 

“I know these seem like dry technical topics,” he said. “But they’re what matters because if you don’t have the right technical capabilities, you’re kind of stuck.” 

Listen to the full episode of The Carbon Copy:

Listen to the episode on:
Apple Podcast LogoSpotify Logo
No items found.
No items found.
No items found.
No items found.
No items found.