With solar power coming to every new home, the solar sector and grid planners in California try to meet in the middle
When it comes to driving solar capacity, California is basically the overachieving kid in class who does all of the work and lets other slacker states glance at its test for the answers. Remember the state’s Million Solar Roofs program in 2007? It resulted in 3 GW of solar installed over a 10-year period while driving lower system prices to result in a steady, established solar market by the program’s end. Cool, but not good enough for California. The state’s latest curve-wrecking initiative starts in 2020 when new codes take effect mandating that all new homes built in California come with a solar energy component.
Being on the leading edge has its challenges though. Combine that solar mandate with the unknowns of the storage market, the influx of electric vehicle loads, existential cyber security threats and constant wildfire concerns and, well, the state’s investor owned utilities (IOUs) could probably use some CliffsNotes.
“We’ve turned everything on its head essentially, and now utilities are having to adapt in real time, while keeping the lights on. That’s a daunting challenge,” says Jen Szaro, VP of research and education at the Smart Electric Power Alliance (SEPA), a DC-based non-profit that works with utilities, regulators and solution providers to ensure a smart transition to a clean and modern energy future. “I know it’s easy to poke at that because it’s easy for them to make mistakes, or be too overly cautious. That’s why we see sometimes, in the name of risk aversion, some of these utilities taking a very conservative approach, and they’re getting dinged for that as well. And sometimes, rightly so. That’s a balance they each have to make.”
The final two installments of our Countdown to 2020 series will focus on California’s grid concerns, the inherent conflicts between established centralized planning and disruptive distributed generation and some of the cool solutions emerging to help ace this next test.
Before you head into installment V of our Countdown to 2020, here are the previous four installments if you need to catch up:
Installment I: The role solar plays in California’s new building efficiency standards
Installment II: Constructing a new solar pathway
Installment III: Would you like solar with that?
Installment IV: How California is both boosting and undermining its solar + storage future
Macro vs. micro
Let’s start at that new home development that is now mandated to make use of solar energy. With that as fact, what new can be done at the grid level, in planning and operations, that might be more efficient than the old constructs focused primarily on centralized generation? The state’s IOUs are trying to figure that out as part of a new distribution forecasting working group, discussing the best methods for forecasting DER adoption and understanding the impacts on the grid.
“California is the canary in the coal mine,” says Cory Welch, founder and president of Lumidyne Consulting. His company has been working on new-school grid planning and forecasts with San Diego Gas & Electric (SDG&E). In general, Welch says the whole market for DER grid planning is nascent. “Tools are still being developed. Methods are still being developed.”
Lumidyne is big into spatial analysis and not getting bogged down with too much granular data that might lead to short-term decisions that miss the big picture. Welch says the biggest challenge utilities are grappling with right now isn’t accommodating all of the new DER but rather how the rise of electric vehicles (EVs) alters the needs downstream.
“You might think solar installation — a load reduction — will be offset by EVs going in and adding to the grid,” Welch says. “And to some extent that’s true, but they tend to be installed in very different places. You don’t see as much overlap as you might expect, so if you’re not looking at these issues spatially, you might underestimate the impact of one or the other.”
Urban centers and coastal areas, for example, have the highest penetrations of EV, to go along with their high population density. Solar systems, on the other hand, are concentrated more in the less populated suburbs.
In theory, new home developments tied to solar generation from the start will greatly reduce the overall load on the grid at every distribution point, which would likely result in a lower infrastructure investment requirement in those areas. But Welch notes that this is offset to an extent in that area if you’re also adding EVs.
“It’s understanding the combination of those so utilities can get a handle on what are the real infrastructure needs,” Welch says. “It’s not as easy to measure that as it used to be … but I do think an accurate planning forecast will facilitate a greater embrace of DERs.”
Then you have the super granular level data points, which are more abundant and rich than at any point in history (and might be more prevalent in a small-scale distributed future). Pecan Street has been studying the impact of residential renewable energy and energy efficiency on the grid for the last 10 years. Pecan Street is able to gather a data point for reactive power, voltage, current, harmonic distortion and phase angle every second and is on a roadmap to be at 6 billion data points a day by the end of the year.
“When you’re talking about mandated solar, you have to start looking at other changes you’re going to make on the grid,” says Scott Hinson, chief technology officer at Pecan Street. His company is working with power management company Eaton to develop and test a next-generation residential demand response solution that will increase overall efficiency of the electric grid and optimize the use of renewable energy generation resources.
One issue he sees in the granular data that utilities can miss is the cause of low power quality on a feeder in a green-built neighborhood with solar. The general assumption is that solar causes this problem and thus needs to be limited on the line.
“But it’s not solar,” he says. “There’s an interaction with the solar providing real power and the new energy efficient devices in the home, like variable speed drives for fridges and washing machines with direct drive drums. We’re talking compact fluorescent lighting and power supplies. All of this reactive power support and distribution current support was always there but you didn’t see it because the home didn’t have generation. But with generation, suddenly the real power portion going into the home goes away. What’s left is this distortion, current and reactive power. So a power company is like, whoa, the power factor is low.”
To Hinson, the data shows a big opportunity for energy storage and the control of highly variable loads, including those pesky EV loads.
“We have the phenomenal variable resource of plug-in EVs. To me, that’s where the ultimate long-term progress is going to be made,” he explains. “I have a feeling charging patterns will be more variable than expected. There’s a fear that everyone comes home and plugs in at the same time, but that’s not what happens. We have a ton of data on this, and there’s a distributed charging pattern throughout the day. All I can say for sure is no one will be charging at home at 7 a.m. because they will want those batteries full before they go to work. Starting at 10 a.m., there is some level of charging for 21 hours.”
Failure to communicate
In between the top down, 10-year utility plan and the potential influence of each solar household is a ton of devices, some of which the utility owns, some of which they don’t but still need the ability to operate (under new communication protocols) — all of which that have to work together as one. No biggie!
Luckily, California was proactive on this front, updating its interconnection rule, Rule 21, in 2017, in part because of lessons learned in Germany after high penetrations of less advanced inverters caused grid instability and required (expensive) retrofits to fix it. This is why Phase 1 of the Rule 21 revisions established the outer bounds of the voltage range and required autonomous inverter functions to widen that voltage operating window and bolster ride through capabilities.
This put the onus on solar inverter manufacturers to engineer more advanced products, a challenge the manufacturers have largely embraced.
“The industry is acknowledging what goes into being a good citizen of the grid — a greater responsibility in terms of providing a stable operating platform for the electrical system,” says Tom Tansy, chairman of the SunSpec Alliance, which has been working at codifying industry-wide communication standards for a decade. “There are certain tradeoffs — like not being able to produce as much power — that the industry seems to accept. In return, we get a more stable and reliable grid, don’t have to deal with nuisance tripping, and I think most people are in the mode of ‘we want to get over these technical hurdles and get these new inverters in place.’”
With grid disturbance problems eliminated so early, Tansy calls Phase 1, active since September 2018, “the single most important factor” for laying the groundwork for 2020’s Title 24 new-build solar mandate.
For the grid operator, it isn’t so much how much DER is replacing centralized resources, it’s how many different types of hardware and software are scattered around. The big virtue of centralized planning that is still eluding DER systems is a common language.
“A big focus for us is on clarifying interoperability standards; they’re kind of still all over the place,” Szaro says. “We’re trying to bring together all the stakeholders, but having a utility try and integrate 72 different products that are all proprietary from the developers because they’re trying to get their edge, I mean, that’s a total nightmare.”
This is where Phases 2 and 3 of the Rule 21 come in. Phase 2 stipulates that inverters must have the capability to communicate with the utility operators and provide an interface to change the voltage settings. Phase 3 then specifies the methods by which you could manipulate those settings. As of today though, there are four communication pathways laid out [See the graphic on pg. 12 for details]. For Tansy and SunSpec, who are working to establish the tests and certifications for this common language, this is a problem.
“The common smart inverter profile — developed in California — stipulated that manufacturers could use any communication protocol to talk to an inverter and unfortunately there is no compliance test for any protocol, just tests for specific ones,” he says. “We can offer testing and certifications for all pathways and interfaces with the exception of the ‘any’ option.” To address this challenge, the California Public Utility Commission is working on a solution that may involve vendor attestation and other compliance methods. Stay tuned for additional details.
The effective date for Phase 2 has been a moving target. Originally set for Feb. 22, 2019, then changed to August 22, and it has now been moved to Jan. 22, 2020. The hope remains that by the end of this regulatory process an elusive standardized communication protocol would be set and products could be tested and certified to meet it, making grid interaction similar to plugging in any router to produce a WiFi signal.
“They’ve got the communication between any aggregated resource and utility well defined but leave it up in the air for the aggregator and their hardware of choice,” Hinson says. “You will still have a bunch of competing standards and some will be manufacturer specific and proprietary, and at end of the day, any kind of residential demand response, any kind of energy storage system, they all need to be easily addressable and secure and available for the utility to make use of.”
A bridge over these concerns could be at the home developer or aggregator level, essentially letting those entities select standard brands and ensure the residential and commercial buildings on one circuit are all compatible and operating as one asset.
“Then you would have an easy-to-aggregate resource in a geographically known location that a system aggregator could bid into the California market,” Hinson says. “If you’re in PG&E territory, you say this is my resource, this is its location, this is the feeder it’s attached to and these are the features I know I have.”
Pecan Street’s data shows there are advantages with aggregated resources being distributed and smaller in scale when it comes to the ramping reserves or fast frequency regulation markets.
Aggregation vs. aggrevation
We’re at 2,000 words, and we haven’t even touched on possibly the three biggest factors looming over everything right now:
- The revision to the Federal Energy Regulatory Commission (FERC) Order 841 that opened up energy storage to wholesale markets.
- The role of California’s Community Choice Aggregators and other third-parties forming virtual power plants.
- Cyber security for all of the above.
“You can make a lot of money there,” Hinson says. “If you can add frequency regulation to that and accommodate multiple markets, you can really make money with batteries and load control systems. Distributing the response can minimize or eliminate the impact to quality of life and totally change the consumer electric model.”
In the next and final installment in the Countdown to 2020, we will make that case.
Chris Crowell is the managing editor of Solar Builder.