Two announcements caught the eye this week, both featuring dominant web titan Google. The common thread was their involvement in the energy and utility world, but cover two distinct areas – in-home utility customer relationships, and Distributed Energy Resources (DERs).
The former is not as exciting as the latter, so we’ll get it out of the way first. Along with analytics firm Tendril, Google’s Home devices were used to let Indiana Michigan Power (I&M) customers ask Google Assistant for details about their bills and consumption. I&M is the first utility to have worked with Tendril to offer this feature, which can also provide customers with details on how to lower their energy usage.
Tendril is taking advantage of the 123mn homes that it says it has tapped, in order to harvest analytics-based insights from. These form the basis for the Tendril Platform, which in turn can then provide information to any device using Google Assistant – provided you’ve logged in and authenticated appropriately.
The data gathered from the utility, as well as in-home equipment if it is supported, can be processed by the Tendril Platform. These include the ecobee and Nest smart thermostats, the latter of which is of course a Google venture. Tendril claims it managed to reduce summertime HVAC load by 85% during demand response events, where its users are encouraged to allow Tendril to reach into their homes and control their appliances, in exchange for some form of recompense.
Which brings us on to Google’s second utility-focused announcement – its collaboration with Stanford University and SunTech as part of the US Department of Energy’s (DoE) ARPA-E (Advanced Research Projects Agency-Energy) Networks of Distributed Energy Systems (NODES) project. Again, as with Tendril, Google is taking an increased interest in how DERs can be incorporated into energy grids.
DERs include solar and wind, and require battery storage to accommodate the variable outputs of these renewable sources. In tandem, demand-response (DR) can better facilitate DERs as it would allow a utility to avoid spooling up fossil-fuel reserve capacity, or having to buy more expensive energy on the markets, by dynamically adjusting the demand-side consumption. In essence, turning a neighborhood’s HVAC systems’ draw down by a few % might avoid the need for firing up a gas peaker plant.
In the NODES project, the goal is to ‘coordinate and control consumer loads and DERs out to the absolute edge of the grid.’ SunTech is providing its Pico portfolio of variable speed motor controllers, which in this case are used to control those DER devices.
These effectively provide ‘smart’ functionality to any DER they are attached to, and also act as sensor hubs that can collect and transport data from additional sensors and devices at the edge. For most, this feature will be used for remote diagnostics and preventative maintenance, but feasibly, this feature could evolve into the Pico family becoming much more advanced controllers, and potentially network-edge processing nodes.
“I am glad that SunTech Drive joined Stanford University and Google in this DoE-funded initiative to advance the state of the art in the area of distributed energy systems. The envisioned technology could revolutionize the IoT control in the domain of electric motors, by enabling a bi-directional, cloud-based communication and control for a wide variety of residential and industrial loads,” said Ana Radovanovic, Research Scientist at Google.
SunTech says that the universal motor controllers have also been deployed in livestock cooling, water pumping, and oil and gas applications. It says that the NODES project will see it work on creating mechanisms that let electric utilities interact with individual nodes, using metadata schemas for both demand reduction and diversion load control – essentially spotting distinct patterns in the surround metadata that can be used to target more specific issues.
The NODES project, outlined here, is looking to enable 50% renewables penetration, hoping to facilitate this via new grid management and control processes. NODES is planning on creating a virtual energy storage system based on use of flexible load and DERs.
As ARPA-E puts it, “the challenge is to cost-effectively and reliably manage dynamic changes in the grid by leveraging these additional grid resources, while maintaining customer quality of service. The expected benefits include reduced periods of costly peak demand, reduced energy waste and increased penetration of renewable energy production.”
The energy grid throws up a lot of anxiety among politicians in this day and age. While reports of nation-state capabilities of inflicting outages on rivals seem overblown, the fact remains that the grids are largely old, centralized, and fairly dumb. Projects like NODES are an attempt to improve both efficiency and security, as well as improve environmental impact.
To that end, ARPA-E thinks that the target of 50% renewable generation could offset around 3.3 quads of thermal generation, and displace 290mn tons of CO2 emissions. For scale, a quad is roughly equivalent to 293 tWh of electrical consumption, 25.2mn tons of oil, 36mn tons of coal, or 8bn gallons of gasoline. ARPA-E also thinks the project could replace around 4.5 GW of generation reserves, saving around $3.3bn annually.
On a related not, Oracle says its Opower wing has saved 20 tWh of energy globally, since launching in 2008. Oracle bought Opower for $532mn in 2016, looking to get into the energy analytics market. Opower’s software and services have enabled energy efficiency programs that saved 9 tWh in the past two years, according to Oracle, which boasts that it is saving over seven times that of second-place Tendril.