Utilities can get almost 9% of their electricity from solar without significant costs or compromises in reliability, according to new research.

With most U.S. utilities now getting less than a percent of their power from solar, this important finding from national energy laboratory researchers should significantly postpone concerns about the impact of solar on the grid.

“They can keep their balancing performance in the same range as it has been without PV for a cost of less than $2 per megawatt-hour,” explained Lawrence Berkeley National Labs Senior Researcher Andrew Mills, co-author of Integrating Solar PV in Utility System Operations from Argonne National Labs, the National Renewable Energy Labs, and LBNL.

The study modeled day-ahead, hour-ahead, and real-time operations and resources at Arizona Public Service (APS) and added first 8.8% solar PV and then even higher levels of penetration. At the lower PV level, “nothing is very different than the way the system is operated today and the costs appear to be modest,” Mills said. “They could add almost 1,700 megawatts of PV with uncompromised reliability and only a small increase in the need for balancing reserves.”

The tipping point: 17%

With PV supplying 17% of the APS system’s electricity, there would be challenges if flexibility is not increased, Mills said. With the high PV and low load typical of spring and winter days, curtailment and the need for balancing reserves would increase. The result would be increased integration costs.

That high PV-low flexibility scenario would also cause a decrease in the standard reliability metric, called the Control Performance Standard 2 (CPS2). The CPS2 measures the monthly percentage of ten minute periods during which supply and demand are balanced to within 50 megawatts.

The North American Electric Reliability Corporation (NERC) requires the CPS2 score to be above 90% and the APS internal goal is 99%, Mills said. In a recent western region study, he added, two-thirds of the participating utilities kept their CPS2 scores between 96% and 98%.

To keep their CPS2 scores up, utilities can meet higher PV penetrations with more system flexibility or with a way for the operator to sell the excess power at times of low load, the researchers concluded. When those options were modeled in the study’s sensitivity analyses, Mills said, “the cost of integration came down and curtailment went down to more reasonable levels.”

The big picture

The study is important in a broader context, Mills explained, because it compliments findings in papers like NREL’s landmark, interconnection-wide Western Wind and Solar Integration Study (WWSIS).

”They both show a lot of renewables can be integrated reliably in the overall system,” Mills said. “We focus on a particular utility’s current operating practices and ask what it would look like to that utility to add a lot of PV. At the low PV penetration level, there aren’t things that stand out from this study that contradict or conflict with what NREL found.”

Two things have changed since the researchers concluded their work that advance the study’s conclusions, Mills noted.

The reliability standard

First, NERC’s proposed change in the reliability metric seems to be nearing FERC approval and implementation. “The CPS2 makes the accounting of performance easier but only measures overall and average reliability for an individual utility,” Mills explained. “It doesn’t ask if the utility is responding when the system is stressed.”

The new NERC metric will reflect how individual utilities perform when the overall system begins to deviate away from the 60 hertz frequency curve toward 59.9 Hz or 60.1 Hz. “That encourages helping keep the overall system in balance,” Mills said.

Concern with overall system balance is also driving the second change.

An energy imbalance market

“An Energy Imbalance Market was talked about when we wrote the paper but with the new CAISO-PacifiCorp-NV Energy EIM being put in place, the concept has gained momentum,” Mills said.

To demonstrate the key factors that minimize integration costs and sustain reliability, the researchers did “a worst-case scenario” with high PV penetration, dramatically limited system flexibility, and significantly increased balancing reserve requirements. The result was an integration cost of $9.60 per megawatt-hour and a 10% curtailment of renewables, despite penalty charges to the utility and reserve shortfalls.

It is an “unrealistic” scenario designed to highlight, the researchers wrote, “the importance of finding buyers for excess power during times with high PV production or the need to increase flexibility from existing thermal power plants or other resources.”

With an EIM now being implemented on the western grid, and other balancing authorities, including APS, actively studying how they can participate, Mills said, “it diminishes the need to study cases where there is no outside market. If you were doing the study now, you would want to include a scenario where the utility was part of an EIM.”

The study shows the advantages of an EIM. “It raises the question of finding the opportunities that the studies with bigger scopes find,” Mills said. “APS is definitely following the development of EIMs and working to understand them.”

The paper’s findings, Mills said, are “what utilities need to be thinking about now so when they get to higher PV penetrations they know what they need to do to meet the challenges.”