Packetized Energy employees discuss results from recent software testing at the company’s lab in Burlington. The company’s Mello device (foreground) transformed water heaters into smart devices that could balance the variability of renewable energy on the grid and enabled them to demonstrate the packetized algorithms with hundreds of devices. (University of Vermont photo: Bailey Beltramo)
George Harvey
Packetized Energy, a company based in Burlington, Vermont, was sold in early March to a New York company, EnergyHub, which will continue the operation of the Burlington office.
Packetized Energy is a spinoff of the Electrical Engineering Department of the University of Vermont (UVM). UVM received $2 million as a grant from the U.S. Department of Energy, and then $3 million more in federal research funding, which led to the founding of the company. The company had nine employees at the time of acquisition, all of whom are now EnergyHub employees. Packetized’s former CEO and now EnergyHub’s vice president of Power Systems, Paul Hines, expects the Burlington office to double in size in the coming year.
To understand the importance of this, it is necessary to understand a bit about how the electric grid operates. Today, most of the grid operates on a baseload power paradigm that was developed in the days of steam locomotives, when the Model A Ford was a great new car. It depended on centralized power plants, transmission lines, and use of power that was only regulated by the size of the power lines to a given building. Many people are comfortable with the concept of baseload power, because that is what they have always had. They complain that renewables will never be sufficient for a power grid because, “the sun doesn’t always shine and the wind doesn’t always blow.”
What they are missing is that baseload power generation is never sufficient to power a grid, because it has no ability to adjust to grid demand, and the demand is always changing. Because of that combination, baseload power is utterly incapable of powering the grid by itself and needs all sorts of load-following resources to do the job.
If that makes you think of renewables needing batteries, then good. It should.
Baseload power and the centralized grid it supports are not the only way to get electricity to users. In recent years, a decentralized power paradigm has been under development, powered by renewables. It has somewhat different requirements, and that is where Packetized Energy comes in.
I will simplify things. In the old days, when the demand for power changed, the grid operators would see the change on their dials. He would lift the phone receiver, and as soon as the operator said “number please,” he would ask to be connected to the operator of a load-following plant. If things went well, everything was back on kilter within a few minutes. In the meantime, output might have had to be adjusted so there was enough power, but not too much. Not enough would mean burned-out motors and other such problems. Too much meant transformers melted down, and the grid would fail hopelessly.
Using a new, distributed system, small generating plants, which could include solar and wind power, but could also include lots of other options, such as small hydropower, biomass, micro-turbines, pumped storage, batteries, and so on, down a long list. Also, instead of an almost unregulated demand system, some types of demand can be automatically deferred, and these include water heaters, electric vehicle charging, and certain other systems.
The problem with this is not that the sun doesn’t always shine and the wind doesn’t always blow. That really is no worse than the fact that the baseload power plant can’t easily change output. The problem is that controlling the grid doesn’t have some person noticing that demand changed using a phone to call the person at the load-following gas plant to tell him or her to change output. Instead, it requires a whole huge bunch of monitors calling a whole huge bunch of generators and users and doing that in a well-coordinated manner. And while that cannot be done using equipment made in 1922 by AT&T, it could easily be done by a set of relatively small computers. One hang-up, of course, is that the computer has to be programmed, and that is not easy, considering the complexity of the assignment.
Packetized Energy developed unique algorithms that run on devices, such as their smart water heater thermostat, which allows the aggregate demand from distributed loads to track, in real-time, the needs of the grid. They not only developed the algorithms but also demonstrated their effectiveness with real devices in real homes and showed load coordination at shorter time-scales than competing approaches.
And that is what Packetized Energy did to make itself worth buying. And that, indeed, is why this is a big deal. It may be the future of our electric power system.
The founders of Packetized Energy wrote a very readable paper about their technology that appeared in the IEEE Spectrum. It is “How to Prevent Blackouts by Packetizing the Power Grid,” and it can be read online at www.bit.ly/Packetized. There is also a very well-done YouTube video on the Just Have a Think channel (youtu.be/NU3woCaFSZs).
Many thanks to our sponsor:
Leave a Reply