← Blog Home

Getting Off to a Hot Start in 2019 – We’ve Acquired EEme!

We’re on a roll. After closing out 2018 with the news of Rubicon’s strategic investment, Tendril is kicking off 2019 with another big announcement: We have acquired AMI disaggregation pioneer EEme.

EEme (pronounced \'ē-mē\) disaggregates individual appliance energy usage from smart meter data, delivering valuable appliance level insights without the need for expensive hardware in the home. Energy geeks like me call it NILM, short for Non-Intrusive Load Monitoring.

Here’s why I’m excited: Integrating NILM into the Tendril platform allows us to do things like detect newly minted EV owners and engage with them around vehicle charging best practices; run remote diagnostics on HVAC equipment to identify inefficient systems and recommend replacements; even orchestrate HVAC runtime around other large but unconnected loads, such as pool pumps and water heaters, to minimize peak demand; and much more.

How does it work? Artificial intelligence. Machine learning algorithms are trained on “ground truth” sub-metered data to identify appliance specific “signatures” in the aggregate AMI data. I lead a team of data scientists at Tendril, and, as you can imagine, my team and I have been chomping at the bit to develop our very own home-grown NILM algorithms for years. In fact, when we started this diligence process back in 2017, we did the analysis and estimated it would take us roughly 18 months to build AMI disaggregation in house.

Then we took a look out at the market. Wow. Crowded field to say the least. There are literally dozens of AMI disaggregation providers these days. Some specialize in high frequency AMI data, others in low frequency AMI data, some have apps, others have APIs, some are based in the US, others are located internationally, across Canada, Europe, and Australia. Given a market teeming with providers, it behooved Tendril to pursue acquisition and partner strategies before building the functionality ourselves. So, we conducted a comprehensive, year-long assessment, vetting more than 10 NILM providers across 15 diligence categories—and we ended up finding a needle in the NILM haystack.

EEme was the only NILM provider that ticked all the boxes: First, its algorithms—custom built hidden Markov models and other machine learning algorithms—are state of the art, developed by PhDs from Carnegie Mellon’s highly ranked College of Engineering. Second, having coined the term “disaggregation-as-a-service,” EEme’s cloud-based API easily plugs into Tendril’s microservices architecture, and can scale to terabytes of AMI data. Third, EEme is the only one out there with a publically available third party evaluation of its technology as it works in the wild, conducted by Pecan Street in 2015. Fourth and most importantly, EEme’s technology has the unique capability to use hints about the physical structure of the home to improve the accuracy of its results.

At Tendril, physics is in our DNA. TrueHome, our physics-based home simulation model, validated by the National Renewable Energy Laboratory, delivers personalized energy insights and precise orchestration for any home, anywhere, based on its physical description—its geometry, mechanical equipment, appliances, thermodynamics, occupants, and local weather conditions. By integrating TrueHome and NILM—physics and machine learning—we improve the accuracy of NILM while unlocking valuable new use cases. It’s one plus one equals three.

The integration is already underway. Just as behavioral science spread quickly through our products following our previous acquisition of Grounded Power in 2010, appliance-level insights will quickly infuse our solutions moving forward. Look for that and a lot more from us this year—it’s going to be a good one.




  • Continuous Demand Management
  • Customer Ops & CSAT
  • DERs
  • DSM
  • Data Analytics
  • Demand Response
  • Disruption
  • Energy Efficiency
  • HERs
  • High Bill Alerts
  • Privacy & Security
  • Smart Home
  • Solar

    Invalid Email Address

    Subscribe to This Feed

    Thank you for subscribing to Behind the Meter Blog.