Over the past year, the Department of Energy has been putting an increased focus on technology to integrate renewable energy into everyday grid operations. This work has included opening loan guarantees to distributed renewables, grants to support field tests of distributed energy integration, and future funding plans to support the development of grid modernization technologies.
Now the DOE’s blue-sky research agency, ARPA-E, is getting into the action -- or, as befits its mission, ahead of the action. Last month, ARPA-E announced $33 million in grants for its Network Optimized Distributed Energy Systems (NODES) program, meant to help 12 university, corporate and DOE laboratory projects that are trying to turn grid-edge assets into networked “virtualstorage” systems.
These projects are meant to “enable real-time coordination between distributed generation, such as rooftop and community solar assets, and bulk power generation, while proactively shaping electric load.” That could allow utilities to manage greater than 50 percent renewable penetration on the grid, “by developing transformational grid control methods that optimize use of flexible load and distributed energy resources.”
This is not a unique concept. Distributed energy resource management software, or DERMS, platforms are being developed to tackle this challenge in one way or another, with grid giants like Siemens and Toshiba and startups such as Spirae, Enbala, Integral Analytics and Smarter Grid Solutions providing different pieces of the puzzle.
Beyond that, there’s work being done by consortia such as Duke Energy’s Coalition of the Willing and Pacific Northwest National Laboratory's transactive energy pilot project to allow lots of distributed energy assets to communicate and act in concert to solve local and system-wide grid challenges.
Some of the projects funded by ARPA-E’s NODES program would help support these kinds of ongoing distributed energy resource (DER) integration efforts, while others would go several steps beyond what today’s utility grid control platforms and DERs are built to handle. Here’s a short description of each project and its aims.
- The University of Vermont won $1.54 million to develop and test a new approach for demand-side management, called Packetized Energy Management, based on approaches used to manage data in communication networks that lack centralized control, but also need high levels of privacy. (That’s similar to the architecture being developed by Duke Energy and other partners in support of a new standard for distributed grid communications known as Open Field Message Bus, or OpenFMB, by the way.)
- The University of California, San Diego won $2.34 million to develop coordination algorithms and software meant to allow DERs and end loads to serve frequency regulation services -- fast-responding grid services now performed by power plants and large-scale grid batteries. Utilities and grid operators are testing similar approaches using smart water heaters, pumps, refrigerators and other flexible energy assets -- and UCSD’s technology could help integrate them.
- Arizona State University won $3 million to develop algorithms that could help manage stochastic (i.e., randomly determined) inputs into traditional grid power flow models. Today’s models have a hard time making sense of inherently unpredictable resources like rooftop solar or electric vehicles. ASU’s algorithms are meant to “integrate uncertainty from renewable resources, load, distributed storage, and demand-response technologies” to “provide system operators with real-time guidance to help coordinate between DERs and demand response.”
- Stanford University won $3.5 million to develop its Powernet technology -- an open-source, open-architecture platform that will use a “Home Hub” device to link rooftop solar inverters, appliances and other end loads using smart switches that replace traditional fuses. Those hubs will serve as a central point for monitoring and control to help meet local and system-wide grid needs, while also maintaining the home's power quality -- an important consideration for a technology meant to be embedded as part of a consumer-facing commercial offering.
- General Electric won $3.9 million to work on creating “synthetic reserves” -- aggregations of DERs that can provide the same kind of flexibility now provided by power plants and large-scale demand response. That will involve forecasting the available flexibility from lots of disparate resources for day-ahead markets, as well as the scalable controls architecture to ensure they deliver as promised -- challenges faced by DERMS competitors across the board.
- The DOE’s own National Renewable Energy Laboratory (NREL) won $3.9 million to build out a “comprehensive distribution network management framework that unifies real-time voltage and frequency control at the home/DER controllers’ level with network-wide energy management at the utility/aggregator level.” Loosely translated, that means linking lots of grid-edge devices with utility grid control systems 00 a complex and expensive proposition. Part of NREL’s goal is to find “computationally affordable” -- in other words, cheaper -- ways to go about this, “by decomposing network-level optimization into smaller sub-problems.” This project is likely to tie in with NREL's existing work with parties like Siemens, Omnetric and Smarter Grid Solutions.
- Another DOE lab, Pacific Northwest National Laboratory, won $2.7 million to work on a “hierarchical control framework for coordinating the flexibility of a full range of DERs,” based on two separate speeds of communication. The slower of the two will communicate data on economic incentives for devices to make their flexibility available to the grid, while the faster one will ensure that devices actually respond appropriately, and don’t interact in unexpected ways to worsen the problems they’re trying to solve. (This sounds a lot like a continuation of PNNL's transactive energy work in the stimulus grant-funded Pacific Northwest Demonstration Project.)
- The University of Minnesota won $2.95 million to work on the same stochastic challenge that ASU is taking on, with a combination of “centralized cloud-based and distributed peer-to-peer networks” to provide both fast-acting and centrally controlled flexibility.
- Northwestern University won $2.69 million to develop a “frequency-based load control architecture,” tapping the data inherent in the frequency of grid power itself to guide how DERs adjust themselves to solve grid problems. The “multi-layer control architecture” will respond to grid frequencies at the distribution grid scale to keep customers’ power clean and stable, while also tapping a centralized scheme to ensure stable frequency synchronization.
- DNV GL won $2.15 million to develop an “internet of energy” platform that will take on the “automated scheduling, aggregating, dispatch, and performance validation of network optimized DERs and controllable load” -- a short list of the main features contained in most DERMS platforms. DNV GL is tapping the technology of Geli, a startup with software to manage energy storage systems for multiple industry partners, and will test it out at Group NIRE’s utility-connected microgrid test facility in Lubbock, Texas.
- The National Rural Electric Cooperative Association won $1.34 million to work on the low-cost side of the DER integration equation. Its GridBallast technology will “monitor grid voltage and frequency and control the target load in order to address excursions from grid operating targets” and integrate that functionality into demand response-enabled water heater controllers and smart circuit controllers.
- Eaton Corp. won $3.3 million to support its own version of DER aggregation for grid reserves, using what it calls a “disruptive cloud-computing solution that will provide agile and robust synthetic regulating reserve services to the power grid.” As with most other approaches to this challenge, Eaton will separate tasks into locally managed and centrally controlled portions to reduce cost and complexity.