Residential storage can undercut benefits of rooftop solar, says new study
Storing solar power increases residential power consumption and emissions, researchers say
The paper, “The impacts of storing solar energy in the home to reduce reliance on the utility,” by Robert Fares and Michael Webber, argues that residential storage can increase power consumption and undercut the environmental benefits of rooftop solar.
The study addresses a business model that has become increasingly invoked as a remedy to the reduction or elimination of net metering benefits. But some criticize the report, saying that it looks too narrowly at the residential use-case for storage.
The report’s focus on transactional round trip efficiency “completely ignores the grid that the home is plugged in to,” said Matt Roberts, executive director of the Energy Supply Association.
In the report, Fares and Webber found that even though a battery storage system could reduce peak power demand by 8% to 32% and reduce peak power injections by anywhere from 5% to 42%, storage inefficiencies increase annual energy consumption on average by 324 kWh to 591 kWh per household.
As a result, storage indirectly increases emissions of carbon dioxide by 153 to 303 kilograms, of sulfur dioxide by 0.03 to 0.20 kg, and of nitrogen oxide by 0.04 to 0.26 kg per household.
The range of results reflect the two scenarios the authors modeled in the paper using full-year 2014 energy usage data from 99 homes in Austin, Texas.
In one scenario, “target zero,” the battery system is used to make the house as energy self sufficient as possible by seeking to keep injections and withdrawals of electricity from the grid at zero at all times.
In the “minimize power” scenario, the battery system has information about the level of electricity demand and solar generation over the day and operation of the storage system is planned to minimize the use of grid power.
In addition, in the zero target scenario the batteries are charged only with solar power while in the minimize approach they are charged with grid or solar electricity so as to minimize demand.
For their base case, the authors use a 3.3-kW, 7-kWh system with a roundtrip efficiency of 85%. In their sensitivity analysis, they consider systems ranging from 1 kW, 1-kWh to 7-kW, 7-kWh with a roundtrip efficiencies in the range of 70% to 100%.
In terms of installed costs for a lithium-ion battery system, the authors used $700/kWh to $1,800/kWh. They found that “under no scenario considered here” could storage provide a sufficient economic benefit to offset a customer’s upfront costs, unless those costs were to fall to $100/kWh.
Battery inefficiencies key
The crux of the matter is that batteries consume energy. So a homeowner who uses batteries to store solar power for later use – when utility rates are higher, for instance – would actually be producing less energy because about 15% of the energy goes is lost in battery system inefficiencies.
Those losses are reflected in roundtrip efficiency numbers of li-ion batteries. The authors cite reported efficiencies of 80% to 93%, but note that those figures could be optimistic because they a not field tested.
More to the point, they say efficiencies greater than 90% would be “difficult to achieve due to losses in the battery pack itself, additional losses in the power conditioning system that converts the battery's d.c. electricity into a.c. electricity suitable for the grid, and supplemental energy required for thermal controls to maintain an acceptable battery pack temperature.”
One of their conclusions is that battery storage is not economic unless it is specifically needed to install a solar system. And that, they say, is unlikely. They cite a study that found that even at 100% solar penetration, Pacific Gas and Electric could maintain adequate voltage levels by increasing the number of transformer tap changing operations at an annual cost that would represent a small fraction of the utility’s annual operation and maintenance budget.
In general, researchers found that using storage to increase solar self-consumption provides a financial benefit when the consumption tariff is higher than the feed-in tariff. But in their modeling they found no scenario where storage provided sufficient direct economic benefit to the customer to offset upfront costs.
Researchers cited Germany, where high electricity rates and low net metering payments can make storage economically beneficial to homeowners, even if it does result in higher levels of electricity consumption. But those benefits are still not sufficient to offset the upfront costs of storage and, as they point out, in systems where fossil fuel-fired generation dominates that leads to higher emissions, both from higher consumption and from shifting load to night hours when fossil resources are more likely to be utilized.
They also considered Hawaii where net metering has been repealed and replaced by two options: grid-export and self-supply. While there are greater economic benefits of combining solar and storage in a market such as Hawaii, the authors still found that the benefits were not sufficient to overcome the upfront cost of a storage system.
In some circumstances, the authors acknowledge that storage tied to solar systems could lower emissions since that comparison is based on the generation mix of the local grid. For instance, In Hawaii under the self-supply option storage would enable more customers to adopt solar panels, which would indirectly reduce system wide emissions. But, the authors note that Hawaii’s isolated grid presents a unique case that is not easily replicable.
Caveats and critiques
Roberts, at the ESA, cited Ontario, which derives most of its electric power from either nuclear or hydro power, as another location where increased energy consumption would not necessarily lead to increased emissions.
The authors correctly point out that the grid in generally carbon intensive, said Roberts, but storage can help enable the transition to a cleaner, more flexible and cost-effective grid.
Roberts also sees a more fundamental flaw in the report.
“No storage system is, or would be, used in this manner.” The research is based on two “unrealistic scenarios,” he said, referring to the “minimize” and “zero target” scenarios in the report.
It is “too big a conclusion” from too few cases, 99 houses in Austin, said Roberts. In addition, the software that underpins an energy storage system can be designed to augment or emphasize an aspect of energy consumption, such as emissions reductions, economic savings, system reliability, off-grid consumption, or any combination of the above. For instance, the results could be very different if Austin Energy, the municipal utility, were to modify its rate structure, says Roberts.
In the report, the authors acknowledge that their study looked at a limited set of circumstances. They set out specifically to examine the storing of solar energy to minimize reliance on the utility because that is the primary value proposition offered to residential customers by storage vendors.
But the important consideration, said the authors, is that energy storage is an energy consumer and not a producer. “It would most likely not reduce emissions or primary energy consumption unless it directly enables intermittent renewable energy.”
Among the authors’ conclusions is what would appear to be an unlikely confluence of interests. They found that storage can be a useful tool in reducing the magnitude of power flows in the grid, and that can benefit a utility by reducing the need for distribution equipment — substations and transformers – as well as the need for new generation to meet peak demand.
They also found that those benefits are greatest under the “minimize power” scenario in which customers seek to minimize their reliance on the grid.
Those are important considerations for policymakers, the authors said, and they argue that the impact of storage should be evaluated separately from the impacts of renewable resources.
In that, the authors are more in alignment with storage advocates such as Roberts. A narrow focus on a single use or even a single beneficiary will not necessarily lead to the best outcome for the grid or the rest of the market, Roberts said.
“Really sitting down and figuring out how we value this for the marketplace, that is the next frontier.”
Follow Peter Maloney on Twitter