For all the theorizing about what the utility of the future will look like, real world examples of how to adapt current power sector business models to the new world of renewables and distributed resources can seem few and far between.
While utilities often trumpet their new smart grid technologies, microgrid projects and storage pilots, actually working out how to make those solutions scalable and profitable can be a lot harder than it looks from the outside.
But utilities across the nation can learn from each other's experiences, with the aim that the questionable technologies of the day can become the ubiquitous tools of tomorrow.
That was the goal of the emerging technologies panel at the recently-concluded Energy Storage North America 2015 conference in San Diego. There, representatives from four major utilities highlighted the challenges and successes of a diverse set of DER pilots, hoping their struggles could translate into easier adoption of distributed resources and demand side resources at other companies.
Learning to monetize batteries in California
The first lesson Pacific Gas and Electric learned from four years of operating a 2 MW sodium sulfur (NaS) battery on the California grid is that it's not yet making enough money from it, joked Substation Engineering Manager Steven Ng.
In early operations, the Vaca Dixon battery system's 75% round trip efficiency made it unable to compete profitably in the California Independent System Operator (ISO) day-ahead market.
“On a good day, I made $80," Ng said. "That is not enough.”
To shift to participating in the ISO’s frequency regulation market, PG&E turned control of the Vaca Dixon system over to the grid operator last August, but the revenues still weren't satisfactory to make up for the initial investment
“We earned $8,000 for the entire month,” Ng said. “That is not enough.”
While it was losing money, Ng said, PG&E was learning. It discovered glitches in the system caused by its own inadequate understanding of market rules. It discovered others associated with ISO operations. Most of the glitches have been smoothed out, Ng said.
The biggest challenge in the frequency regulation market has been predicting the NaS battery system’s State of Charge (SOC).
“Some days we provide regulation for many hours, some days we don’t provide it at all,” Ng said. “We are stilling learning how to predict and manage the SOC.”
Perhaps the most important thing the utility has learned so far is that the system can earn full “mileage payments” from the ISO because how quickly and accurately the battery can respond to the grid operator's signal.
“This time response capability is extremely fast when compared to any fossil, hydroelectric, or nuclear plant,” a July paper from PG&E and the California Energy Commission reported.
The Vaca Dixon system “can follow the signal in one-tenth of one second,” Ng said. “About 30% of our earnings now come from mileage payments.”
Based on these outcomes, PG&E is in the process of developing contracts for “hundreds of MWs” of battery energy storage, Ng said.
With the automated control system it expects to have in place by the end of 2015, the utility plans to use the almost instantaneous response time of its battery to capitalize on momentary price spikes in the ISO’s real-time electricity market, he added.
And PG&E is pushing the ISO to address market rules for battery storage to provide black start and other electricity market possibilities, he said.
Net zero homes and distributed resources
While PG&E was learning how to make money from storage on the grid, the Sacramento Municipal Utility District (SMUD) was studying how behind-the-meter batteries and other efficiency measures can save it and its customers money.
SMUD's 2500 R Midtown project was launched in 2014 in partnership with Pacific Housing, a not-for-profit Sacramento housing developer, and Sunverge Energy, a provider of integrated solar and storage. It aimed to study how extremely efficient homes outfitted with distributed resources can interact with the grid.
Each of the 34 single family homes in midtown Sacramento is designed to consume zero net energy and zero peak energy. The houses all sold within a year at market competitive prices.
“In addition to being extremely energy efficient, a Sunverge solar plus battery system was included in the home’s price, so the homeowners had no extra lease payment or other fee,” said SMUD Smart Grid Research and Development Senior Project Manager Lupe Jimenez.
To measure the project’s viability, she said, the utility tested a range of use cases, including peak load shifting, an optional time-of-use critical peak pricing (TOU-CPP) rate used by about half SMUD customers, PV smoothing, regulation, uninterruptible power source (UPS), and power quality.
The peak load shifting was essentially renewables shifting and it was successful, Jimenez said. “We saw an average 1.35 kW saving stream daily and an additional 1.31 kW saving on our critical peak days" for each house.
The other tests were equally successful, mostly on a limited scale. Power quality met SMUD standards. UPS was tested on only two customers, one who was a SMUD exec’s relative and one by accident.
In the case of the accident, a customer didn’t understand she still needed utility service and power was turned off for about three days, Jimenez said. As a result, the utility discovered the UPS function works as planned.
“We also realized we need to enhance our communications with the customers,” she added jokingly.
Besides that hiccup, SMUD found that many of its customers in the pilot program were remarkably invested in their energy use — not surprising for consumers who bought zero net energy homes.
One customer even “sent us graphs of his usage and wanted to understand how the system worked and how his participation could be customized,” Jimenez said.
Motivations for customer participation in the program varied, she continued. Some were looking for ways to save, some liked the idea of being involved in something new and exciting, and some wanted to to become as independent as possible.
“Are these customers ready for a transactive grid?” she asked rhetorically. “No. But they are definitely interested and willing to have someone else manage it for them.”
“They are still accepting of utilities because we are just starting to change," Jimenez concluded. "But there will a point where they will expect their interface with the utility and the organizations beyond the utility to be autonomous and ubiquitous and easy for them to use. We will have to be able to respond to that.”
Leveraging demand management instead of building a substation
New York’s Consolidated Edison was faced in 2013 with the real possibility of power shortages by 2018 in a Brooklyn-Queens residential neighborhood that was undergoing unprecedented growth, said Con Ed Director of Energy Efficiency and Demand Management Rebecca Craft.
The system capacity was 763 MW, the load was approaching 800 MW, and the area’s substation feeders were at their design criteria limits for hot weather. A roughly $1.2-1.3 billion substation was the obvious solution.
But Con Ed had previously deferred a capital spend of “about $250 million” with an approximately $107 million investment in its commercial networks so it decided to consider a Brooklyn-Queens demand management fix.
“We never did a project this large,” Craft said. “It is a need for customer-sided resources of about 41 MW and non-traditional utility resources of another 11 MW, in addition to some substantial load transfers.”
Some of the 78 responses to its call for proposals seemed workable. The New York Public Service Commission, just at the beginning of what became its landmark Reforming the Energy Vision (REV) proceedings, approved a $200 million spend.
“We had to get the 41 MW off the grid within three years,” Craft said. “We are looking at an average price of about $3.7 million per MW. That is not cheap but it is not of control, especially considering how much it costs to build in New York.”
Of the 11 MW of non-traditional utility solutions, most are energy efficiency and demand response. Despite strong incentives, Craft said, “we have not seen on the customer side a race toward storage.”
Con Ed sees the Brooklyn Queens Demand Management project as a “precursor” to the changes coming through the REV proceedings, Craft said. Track 2 will include the distribution system platform and distribution system implementation plans, due in June of 2016.
The utility is pleased the PSC has decided utilities should be given the first opportunity to provide a distribution system platform, she said. “If REV is going to succeed, it has to offer all players, including utilities, an opportunity to earn adequately for the risk on their investments.”
In New York’s deregulated power market, utilities don’t own generation. The REV initiative won’t change that. If Con Ed is not the distribution system platform provider, its role and profit potential would be sharply circumscribed.
Con Ed’s service territory and electrical system are “extremely unique and we think we are in the best position to be the system operator,” Craft said. “We also need to maintain operational control since we remain responsible for reliability.”
Some have argued that giving utilities control of the distribution system in an increasingly competitive power market would be like putting an airline in charge of air traffic control.
“We are not allowed to own generation, we are not allowed to own DG, we are not allowed to own PV, and we are only allowed to own batteries because they are not generation. It is hard to see under those conditions how the conflict of interest arises,” Craft responded. “And there is no federal restriction against air traffic controllers owning an airline.”
Planning the grid around DERs
Not many years ago, Southern California Edison (SCE) did planning at the transmission level using what it called loads and resources, said Integrated Planning & Strategy Director Mark Nelson.
Today, he said, SCE’s distribution resource planning process perceives the distributed energy resources (DERs) on the distribution system as a potential source of both increased and decreased load.
Now, the utility is studying its distribution system in the planning process to identify where enhanced local value opportunities for DERs are, “based on the bidirectional flow, the flow to the customer and the flow from the customer,” Nelson said.
The planning now also includes a consideration of storage and its potential to move solar over-generation to where it can be effectively used. It also includes the potential of electric transportation as a source of load for some circuits and generation for others.
“The entire situation has gotten exponentially more complicated,” Nelson said. “The distribution grid used to be designed for maximum through-put of central generation at the time of peak demand. Now it is possible to instead use the many small resources, DERs, on the circuits so that increased load on a circuit does not require expanding the circuit.”
SCE President Pedro Pizarro told Utility Dive in August that determining the locational value of storage and other DERs on the distribution system is still an “evolving area,” one that will “continue to get a lot of focus over the years ahead.”
SCE’s Distribution Resource Plan — a framework for how the company plans to modernize its grid — that the utility filed with regulators in July was a big step forward in that process, he said.
“[The DRP] includes a lot of commentary around how we can provide greater transparency in terms of the value of distributed resources at different points in the grid,” Pizarro said.
For instance, SCE has begun putting up online maps that show the available capacity for distributed resources on different points of the grid, Pizarro said, “and there are more tools being developed as a part of the DRP process in the years ahead.”
The distribution system’s expanded possibilities gives planners “a more robust tool box for optimization,” Nelson said. “But it is a lot more complicated optimization problem and we all will be spending a lot of time in the near term in the DRP process and the upcoming integrated resource planning process at the CPUC to figure it out.”
"We have some big tasks ahead," he said, "but we have wonderful opportunities ahead, too.”