With rising penetrations of variable renewables and customer adoption of distributed energy resources (DER), electricity providers' attention is shifting to distribution systems.
Better distribution system control capabilities are being developed to meet power flow challenges. Switches, sensors and software are allowing power providers to use aggregated residential DER systems like small virtual power plants (VPPs) that can balance renewable generation variability and supply grid services.
"A virtual power plant is aggregated and optimized distributed resources that essentially provide the same services as a traditional power plant," Research Director Peter Asmus at Navigant Research told Utility Dive. "Any resource that can be sensed, touch, controlled and automated can be part of a virtual power plant, which means it can be different every day, depending on grid and market conditions."
Utilities are modernizing distribution systems with hardware and software that turn the challenges of changing power flows into opportunities. But DER management systems (DERMS) and advanced distribution management systems (ADMS) come with price tags. A VPP proposed for the aging Los Angeles Department of Water and Power (LADWP) system that serves Hollywood and much of the city may soon become the first full-scale test of the value proposition.
Defining VPP, DERMS and ADMS
To manage more unpredictable distribution system supply-demand dynamics, electricity providers have increased emphasis on demand side management, VP for Industry Solutions Eric Young of DERMS software provider Enbala told Utility Dive. "Demand side resources can supply and reduce the use of electricity at the distribution level, which also relieves transmission level demand."
With software and automation, VPPs expand DER functions on the distribution system, a paper in the Navigant-Enbala series on VPPs co-authored by Navigant's Asmus added. They "deliver the same services (and more) as a traditional power plant" without concerns about land use, air emissions and waste management.
Initially, VPPs delivered small amounts of load reduction by commercial-industrial customers to wholesale markets for demand response, Enbala's Young said. "Enbala intended to monetize the flexibility of large customers' assets in the power markets." But new distribution system technologies offer "opportunities to harvest the value of all customers' flexible loads for peak load relief."
"Making a VPP work is not easy but it is doable .... The faster it is built, the more costly it is likely to be because it represents a big change from the way things were done historically."
Senior Technical Consultant, Center for Energy Efficiency and Renewable Technologies
Distributed solar, smart inverters, behind-the-meter battery storage, electric vehicles, smart thermostats and smart appliances are being incorporated into VPPs, another paper in the series reported. It is the next step toward managing a "more DER-dominated grid" and "setting the stage" for "the evolution to DERMS" and "optimization of distribution networks."
A DERMS allows a utility to know where each of its DER assets is, "on which specific transformer or feeder," a third paper reported. It can optimize DER performance through "coordinated resource utilization and voltage and power management" and can "mitigate problems by controlling devices" throughout the system.
With this granular control of a VPP, the newest DERMS software "can deliver exactly the load reduction that is needed, when and where it is needed, to protect grid reliability," Enbala's Young said.
But DERMS' "holistic control of grid resources" adds "larger up-front costs" for a VPP, Navigant reported. And utilities often prefer to start with an ADMS because it gives them wider system visibility for integrating DER. With investments in both, power providers have "unprecedented control over the power grid," it added.
While utilities focus on ADMS for DER integration, "solicitations for DERMS solutions are few and far between," Navigant reported. And the market focus remains on monetizing value streams from aggregated DER rather than providing system benefits like capacity, voltage regulation and frequency regulation.
Eventually, "there will be a need for DERMS to solve grid reliability issues in a more surgical way," Navigant predicted. For now, VPPs' DER aggregations "provide a logical first step" and "the ideal foundation for utilities to gain experience with managing and optimizing the increasing amount of DER."
By monetizing DER for peak shaving, VPPs enable DER "financial value" while DERMS maintain "the physics of the power grid," another paper in the Navigant-Enbala series said. Without "adaptive, self-learning, and nimble aggregation and optimization platforms," high levels of DER "could create chaos." But with DERMS, DER become part of a "solution ecosystem."
VPPs at utilities
No power plant-scale operational VPPs appear to be in service in the U.S. at present, but there are many small-scale efforts, initial software rollouts and rising interest.
About 25% of 2018 DistribuTECH conference survey respondents reported a small-scale "DERMS or VPP platform in place to control or manage DER," up from 17% in Enbala's 2017 survey. Over half of respondents without a platform said they expect to implement one "in the next 36 months." Costs and regulatory barriers were their biggest concerns, Enbala found.
"Making a VPP work is not easy but it is doable," Center for Energy Efficiency and Renewable Technologies (CEERT) Senior Technical Consultant Jim Caldwell, a former LADWP executive, told Utility Dive. "The faster it is built, the more costly it is likely to be because it represents a big change from the way things were done historically."
Utilities are beginning to see the opportunity and test the concept.
National Grid in Massachusetts and Vermont's Green Mountain Power (GMP) are successfully using customer-sited solar-storage aggregations on a small scale to meet system and customer needs. GMP's 11 MW of DER capacity has generated over $800,000 of net value since 2018, utility spokesperson Kristin Kelly emailed Utility Dive.
"The value of virtual power plants in replacing physical power plants will come down to meeting reliability needs" through a resource adequacy contract.
Energy Analyst, Union of Concerned Scientists
Aggregations of DER developed by Stem, AMS, Ice Bear and others as part of a Southern California Edison (SCE) local capacity project are operating successfully at a smaller than power plant scale and showing enormous potential, the companies have reported.
Xcel Energy selected the AutoGrid DERMS to manage DER pilots across eight states and by National Grid in its solar-storage program, Navigant's Asmus reported. AutoGrid is also being used by New Hampshire Electric Cooperative and is being re-evaluated for "higher value" distribution system applications, according to co-op spokesperson Seth Wheeler. These are also not power plant-scale programs.
VPP value in reliability
In California, which leads the U.S. in DER, "the value of virtual power plants in replacing physical power plants will come down to meeting reliability needs" through a resource adequacy (RA) contract, Union of Concerned Scientists (UCS) Energy Analyst Mark Specht told Utility Dive.
An RA contract compensates generators at higher than energy market rates for providing capacity to meet unexpected demand spikes. To win an RA contract, a VPP would have to demonstrate to California's system operator it was as reliable as the natural gas plants that typically serve RA.
California has a "techno-economic potential" of "approximately 9 GW of RA" across investor-owned utility service territories, according to a study commissioned by Stem and residential solar installer Sunrun. That could significantly reduce the 15 GW to 33 GW of natural gas capacity an Energy + Environmental Economics study found to be California's expected 2050 reliability need.
Sunrun announced July 18 it will provide "several MWs of solar" and "over two MWh of batteries" as RA to help East Bay Community Energy replace a retiring California peaker plant. Sunrun has also contracted to provide ISO New England with 20 MW of reliability capacity using aggregated residential solar-storage by 2022. For now, both potential VPPs remain untested.
Phase 3 of Tesla's VPP in the Australian state of South Australia may provide the first power plant-scale proof-of-concept. In July 2018's Phase 1, 100 residential solar+storage systems were deployed. Late last year, in Phase 2 of the state-supported program, 1,000 more residential solar+storage systems were deployed.
In July, former Australia Energy Minister Dan van Holst Pellekaan acknowledged the success of Phase 2, opening the door for the deployment of 50,000 systems and expansion of the Enbala DERMS used to manage the VPP.
Regulatory and cost barriers must still be overcome in the near term, Asmus and Young acknowledged and UCS's Specht raised a longer-term concern.
There is ample demand in California for VPPs to replace energy, grid services and reliability benefits now coming from natural gas generation, he said. But the reliability value for solar was reduced significantly in the May 24 RA proceeding proposed decision. That value is lower because additional increments of solar during periods of over-generation are no longer as valuable as generation available during evening peak demand periods after the sun sets.
"California faces an almost incomprehensibly massive transition. It may need 150 GW of solar, and VPPs can be one solution. But there is no single solution."
Executive Director, Large-scale Solar Association
"There are similar limits, which the state is nowhere near, for the solar+storage in VPPs," Specht said. Eventually, system operators will need very long duration reliability which could exceed the capacity of solar stored in batteries before there is adequate solar to recharge them. Other resources, like geothermal or biofuels, will be needed.
VPPs like the one proposed by Sunrun for LADWP "should be part of the state's planning toward its net zero emissions by 2050 mandate," Large-scale Solar Association (LSA) Executive Director Shannon Eddy told Utility Dive. "California faces an almost incomprehensibly massive transition. It may need 150 GW of solar, and VPPs can be one solution. But there is no single solution."
VPPs go Hollywood
The Sunrun proposal came in response to LA Mayor Eric Garcetti's Green New Deal calling for LADWP to shutter its three remaining natural gas plants and get 80% of its power from renewables by 2036, Sunrun VP for Energy Services Audrey Lee told Utility Dive.
One of the retiring natural gas plants could be replaced with an approximately 295 MW VPP composed of "rooftop solar and battery storage," Sunrun proposed. It can be done by 2030, and for almost $60 million less than the cost of new natural gas generation.
LADWP's 182 MW of residential solar on 36,000 homes is only 2.5% of the utility's 1.34 million residential customers, Sunrun reported. By contrast, San Diego Gas & Electric (SDG&E) has 740 MW of installed solar on 139,000 homes, 11% of its 1.25 million residential customers.
Using existing incentives and programs, LADWP could have 403 MW of residential solar by 2024 and 862 MW by 2030, Sunrun spokesperson Andrew Newbold emailed Utility Dive. It could also have at least 168 MWh of residential battery storage by 2024 and approximately 1,180 MWh by 2030.
"If batteries are used to provide electricity to homes and neighborhoods over a peak period lasting four hours, this is roughly equivalent to a 295 MW natural gas power plant," Newbold said. It is "the same order of magnitude of the peak capacity provided by either LADWP's Scattergood or Harbor generating plants."
The DER installations being used by SCE include "hundreds of aggregated commercial-industrial customers' behind-the-meter storage systems being dispatched in place of natural gas generation," Lee said. "Sunrun can operate residential resources as a virtual power plant in the same way."
LADWP would likely need to streamline its interconnection practices, she acknowledged. "An interconnection that takes SDG&E a day or less can take LADWP 60 or even 90 days." LADWP's slower interconnection times are in part a failure to institute the simplified and online protocols used by SDG&E, surveys have shown. But they could indicate technical challenges to the proposed VPP.
As aggregator, Sunrun's software platform would deliver the DER, Lee said. "Sunrun will adapt to LADWP's needs, but the utility may need to procure a DERMS to communicate with Sunrun's platform and manage the distributed assets."
The not inconsiderable cost of an operating platform is only one part of the grid modernization challenge LADWP faces, Sierra Club spokesperson Luis Amezcua told Utility Dive. "It is one of the few utilities in the region still dependent on 4.8 kilovolt lines."
A VPP might streamline that modernizing because "it will require distribution system upgrades as the solar and storage are added," Amezcua said.
But it will add to the cost and the challenge, he and others agreed.
"There is no question LADWP needs to modernize its distribution grid," CEERT's Caldwell said. "It is Korean War-era topology and technology with circuit loads twice what they were designed for. But the people at the utility understand and are already investing in reliability."
The challenge of a VPP could give them a reason to move more quickly and innovatively, Caldwell agreed. "It will be interesting to see if they can make it happen."
As LADWP moves from "a centralized infrastructure to a more distributed, local, and clean one, it will need collaboration across technologies," Sunrun's Lee said. "The people at LADWP have shown the courage to look pragmatically at what they need to do and move ahead. A VPP can be part of a portfolio of solutions."