It is no longer only visionaries like Tony Seba who are mapping the contours of an integrated emissions-free electricity and transportation system, with rooftop solar and other forms of renewables powering not only grids but also electric vehicles (EVs). Instead, this is becoming more and more the mainstream expectation of what the future will look like.
An important component of this vision is the potential synergies between EVs and solar. EVs have the ability to serve as a mobile fleet of batteries, and EV charging will supply additional flexible demand which could soak up a much higher degree of solar production than could otherwise be utilized.
The first part of this vision is already coming. EV sales are increasing and prices are falling in a virtuous cycle, and we may be approaching a tipping point where change happens much more rapidly. But while EVs are technically able to complement solar and wind, this will not happen automatically. Instead, like other aspects of the energy transition, it will take careful planning and supportive policies to develop the synergies between these two technologies. And if such work is not done, EVs could end up as a problem for the entire power system, and could even make renewable energy integration more difficult.
The coming EV boom
As was the case with solar 15 years ago, EV deployment is still in its infancy. According to Bloomberg New Energy Finance (BNEF) EVs represented slightly less than 2% of global automobile sales in the third quarter of 2017, and are a much smaller portion of overall vehicles on the road.
Certain geographies are much further ahead. As of Q3 2017, EVs had captured 40% of automobile sales in Norway and 20% in Iceland. EV sales also represent 4–10% of the automobile market in the cities of Los Angeles, San Francisco, and San Jose, California, as well as 6–8% in Shanghai and Beijing.
Given current growth rates, BNEF is forecasting that 40% of new car sales will be electric in 2030. And while BNEF’s predictions are much more optimistic than those of the International Energy Agency and the U.S. Department of Energy’s Energy Information Administration, they may still be conservative. It is notable that BNEF has changed its long-term forecasts for EVs several times in the past few years – each time predicting more EVs on the road sooner.
The dynamics of demand are relevant here. As has been the case with solar, falling costs and increasing competitiveness with fossil fuel technologies are driving a growing market. With the exception of nations like Norway that provide heavy subsidies, EV ticket prices are still typically much higher. However, because EVs typically have much lower fuel costs, are vastly simpler than internal combustion engines (ICEs), and require far less maintenance, sticker price does not tell the whole story.
A 2017 study by the University of Central Florida’s Electric Vehicle Transportation Center shows that even over a five year time frame the cost of owning an EV is already lower than a comparable ICE vehicle, and this differential gets greater the longer one owns the vehicle.
With EV prices continually falling year after year, the logic of market transformation is beginning to take place. At this point automakers are seeing that the future lies with EVs, not ICE, and an increasing number are setting timetables to shift entirely away from ICE in the future. This is further reinforced by France, the United Kingdom, and China preparing to ban ICE, as well as California Governor Jerry Brown’s goal to have 5 million EVs on the road in 2030.
All of this can lead to much faster market growth. And while these changes are usually incremental and driven by the aggregate of many personal decisions, fleets of vehicles can move more rapidly. This has been demonstrated in the city of Shenzhen, China, which recently switched its fleet of more than 16,000 busses from diesel to electric.
Technical dimensions
A more critical service that EVs can supply is to shift electric loads. While the additional demand that EVs offer is of great interest to utilities which have seen electricity demand growth flatline, this demand will come at specific times that can be hard to predict, and this creates significant technical concerns. But if this demand can be matched to times of wind and/or solar output, this can allow a much greater volume of renewable energy to be deployed.
Utilities and regulators are already experimenting with ways to alter the timing of customer charging. There are several ways to do this, including active control of chargers and management of fleets, but the softest approach and the one that has the most promise for charging systems not controlled by utilities is to implement time-varying rates.
Rocky Mountain Institute (RMI) says that a pilot program by The EV Project and utility San Diego Gas & Electric Company (SDG&E) has shown that rate design can “substantially influence charging behavior,” with a 6:1 price ratio between peak and off-peak rates enough to shift 90% of charging to a “super-off-peak” period.
However, this program, like others, is focused on overnight charging. For EV charging to complement solar, such charging must shift to mid-day, which also means that charging stations must be located at workplaces and other locations where EV owners park their vehicles during the day.
RMI Electricity Practice Manager Chris Nelder, the lead author of multiple reports on EVs, notes that there is no real technical or economic difference between the fast-charging stations that are implemented at homes versus workplaces, but there is still the matter of ensuring via policy and/or market signals that daytime charging stations get built.
A moving target
When discussing technical aspects of EV charging, it is important to note that utilities, grid operators, and even participants in the EV space are attempting to plan for several changes, the time lines and details of which are far from clear. Utilities are expecting more EVs, but are not sure when. Meanwhile, both charging technology and EVs themselves are rapidly evolving.
The first EVs relied on Level 1 charging at the homes of customers, but as time goes on more are moving to faster Level 2 charging, and an increasing number of DC fast charging (DCFC) stations are being deployed. The more rapid nature of these systems means that if a vehicle is left connected for an extended period, such as an eight hour workday, there is more choice as to when to charge. But such charging also draws much more power over a shorter time period, making alignment of demand and supply more critical.
EV batteries are also getting bigger, with many EV models already sporting 40, 50, or even 80 kWh batteries. This further reinforces the demand for faster chargers.
Many fast-charging networks will not be located at workplaces, but like gas stations will be built along transportation routes. Customers will want to use these on an on-demand basis, not over an eight hour period. Because of this, RMI’s Nelder says that the only way to manage on demand charging with DCFC is to incorporate redundant battery storage, so that these batteries can draw power from the grid at the most optimal time, then use this stored power to charge EV batteries.
But even as redundant batteries can solve this problem, a further issue for utilities is the lack of spatial control. Whether at workplaces or along transportation routes, DCFC will be located according to other considerations than what is most convenient for utilities.
Ownership and policy barriers
The need to balance supply of electricity with charging demand, both temporally and spatially, creates a new set of considerations. If third-party charging systems – and their batteries – are to be located at more optimal locations for the grid, developers will need to be able to access distribution-level grid data. Such data are also important for other distributed energy resources, but grid modernization efforts such as New York’s Reforming the Energy Vision have shown that utilities are often slow in parting with their precious information.
“There is a major barrier in terms of the information that utilities need to provide.” Energy Innovation’s Senior Fellow Eric Gimon tells pv magazine. “Until the utilities complete hosting analyses to the distribution grid, it will be difficult to identify the best places for charging stations.”
One solution to the multiple technical challenges here is to let utilities build their own charging networks, and California regulators have recently switched their position and concluded that this can be in the public interest. Nelder says that utility partnerships are especially important for daytime workplace charging. This does not solve all problems. Energy Innovation’s Electricity Policy Manager Mike O’Boyle notes that while issues with managed charging can be addressed in order to provide load-shifting, EV batteries will likely not be allowed to discharge to supply grid support services, as this will void warranties around cycle times.
Catching up in time
Both RMI and Energy Innovation note that catching policy up to real-world developments represents a potent challenge. “Vehicle electrification isn’t an if or a when question anymore; it’s only a question of how fast and can we be ready in time,” notes RMI’s 2017 report From Gas to Grid. The report notes that even 1.5 million electric vehicles could add 10 GW of load in California, which could increase the peak by around 20%.
From Gas to Grid notes that there is great potential in shaping and controlling EV charging, however the report also states that if utilities respond to EVs “late and reactively,” they could make things worse, including inhibiting integration of renewables and increasing curtailment.
Energy Innovation’s Gimon warns of “rate inertia,” but it is not only rates which pose a problem here. “The biggest institutional barrier to getting the most out of EVs is the time frame of utility decisions,” notes Mike O’Boyle.
And if rates make things difficult for EV charging, this could also affect deployment. “If people lose money installing chargers, they won’t install more,” argues Gimon.
There is a lot for many parties, including utilities and third-party developers, to gain from the move to EVs, but it is up to utilities, grid operators, and regulators to put the best policy frameworks in place to prepare for a future which may come sooner than many expected. “Utilities need to improve distribution system planning, and making sure that the distribution grid is ready for a whole new kind of resource that is going to be drawing a lot of power,” states O’Boyle. “There is no reason any longer to delay building out the charging infrastructure.”
This content is protected by copyright and may not be reused. If you want to cooperate with us and would like to reuse some of our content, please contact: editors@pv-magazine.com.
Interesting piece. I see some challenges for EV + PV, but mostly I see great opportunity, opportunity to power the transportation sector via clean, green energy. This opportunity is NOT there at all for traditional combustion vehicles. And this is a fundamental — and crucial — distinction/difference.
I agree with that assessment.
Although the theoretical problems could be real under certain circumstances the actual information available to date suggests a smooth transition. Most people who own EV’s today are charging them at night at home. That might be due to a lack of stations, but it is also due to the convenience of not having to stop at a station which sells nothing else most people want.
Total conversion of the U.S. passenger fleet to electricity requires only 20% more electricity than we presently use. That much growth is already expected. Theoretically, if everyone charged at the same time every night or day, a 100% EV fleet could raise peak demand (quite different from total energy) by more than 100%. If the current practice remains the mode, that will never happen, although we will want time of use rates. Limitations on fast charging are already in place. Utilities will expand access to such facilities as it makes sense, in order to make money.
There is indeed a lot to be worked out. But there is an army of experts at work already. Google PUCO Power Forward and find the presentations for March 6, 7 and 8 to see a good example of how the industry is already on top of this, and presenting to regulators in one state with only a handful of EV’s already on the road.
The cost burden – and cost shift burden – of this type of activity needs to be front and center. As UC Berkeley Haas School recently demonstrated, each electricity customer in California now donates $65/yr in cost shift burden to solar PV customers who use the grid as a battery for free. That’s a significant payment by non-participants to participants. To the extent that EV customers wind up with free public infrastructure and marginal cost pricing that fails to contribute to grid fixed charges, that $65/yr burden on non-participants could easily double or triple. The burden to non-participants, and especially those in disadvantaged and low income areas is currently ignored for the most part. High electricity prices will lead to grid defection and a reduced ability for regulators and policy makers to use utility balance sheets for GHG reductions.
I’m sorry, but the op-ed from the Haas School professor that you cite is an uninformed opinion by someone who either is not familiar with technical aspects of the electric grid and the various studies about the impact on rates from distributed solar, or is deliberately ignoring them.
More than a dozen state-level studies have reached the conclusion that at low penetrations, net-metered solar is a net benefit: https://www.brookings.edu/research/rooftop-solar-net-metering-is-a-net-benefit/
I’m afraid that the back-of-the-napkin estimates from an academic writing outside of his field (and who did not bother to do his homework) simply do not stand up to the weight of research done on this subject.
Let’s do the math for BTM PV with NEM.
The revenue for NEM in California is system average rate, since California utilities use variable rates to maximize conservation through the use of the highest possible average rate. Rounding down, CA res rates are 15 cents per kwh or $150/MWH. Therefore, residential BTM PV revenue for NEM is $150/MWH. Average rates are probably more like $180-200/MWH.
Contracts for PV in RPS scale are running about $30-35/MWH these days, and I’m sure that some developer would be happy to tell me that they’re really in the $20s. In order to give rounding to the BTM PV, let’s say RPS scale PV is $40/MWH.
That means that Residential PV has a $110/MWH subsidy over large scale RPS. I would defy ANY VALID STUDY to demonstrate that the avoided Transmission plus transmission loss is $110/MWH or that the avoided Distribution is $110/MWH – and one or the other is needed to close the gap and eliminate the $100+/MWH subsidy paid by non-NEM customers for BTM NEM PV.
I suppose if you compare BTM PV to coal with a damage based adder for criteria pollution and GHGs, you can get a case where BTM PV doesn’t need a subsidy, but that’s not the case in CA. In CA, head-to-head would mean comparing large scale PV to BTM PV – there’s no carbon difference and the load shape/capacity factor for large scale is generally better due to tracking.
Yes, there’s the subsidy, and we can argue over the level of it, but the gap to close is $100+ per MWH between RPS scale and NEM BTM PV.
You are attempting to conflate two different arguments here. It is much easier to argue that new utility-scale PV contracts are a better deal for ratepayers than behind-the-meter solar, versus net metering representing a net subsidy.
Per the new contracts, utility-scale solar in California is the cheapest form of new generation, and much cheaper than many of the other resources that have been deployed to date. So while it is likely much cheaper for ratepayers than behind-the-meter solar (however this requires actual and not back-of-the-napkin math, which you and the Haas professor seem stubbornly resistant to), that is very different than comparing BTM solar to the entire portfolio of resources that are currently deployed for generation, either in California or the United States as a whole.
Furthermore, it is very convenient of you to use California, where utility-scale solar is particularly cheap, and where the levels of solar penetration mean that the peak-shaving benefits of solar are largely used up. Per a recent study by Lawrence Berkeley National Laboratory, in states like California where more than 10% of electricity comes from solar, net metering is a net subsidy – or at least it was before NM 2.0 and TOU rates – but that level of subsidy is so small as to be basically insignificant compared to other utility actions.
So to clarify, per my previous note about the 12 studies, in almost all locations across the United States net metering is a net benefit.
I also want to note that is is a peculiarity of American internet culture where people feel free to challenge the work of actual scholars working in their fields with back-of-the-napkin math. Please in the future do not waste our time. Thanks.
Dear Rocky Mountain Institute (RMI): This is a very well researched and thoughtful article by PV-Magazine’s Christian Roselund.
Accordingly, my question to RMI is what about hydrogen fuel cell electric vehicles (HFCEV) in RMI’s modeling? Theoretically speaking, its simple Newtonian physics: F=MA. By replacing the heavy lithium ion batteries with lighter than air hydrogen in the EV platform, the dynamics of Newton’s Second Law are obvious:
https://en.wikipedia.org/wiki/Hyundai_Nexo
https://www.energy.gov/eere/articles/us-doe-hydrogen-prize-winner-exports-innovative-small-scale-hydrogen-refueling
http://www.ivysinc.com/simplefuel-main-page
RMI is to complimented for their global leadership into a low-carbon economy!
Warmest regards, Carl L. McWilliams – Glenwood Springs, Colorado
Thank you for the compliment! And I agree, RMI does incredible work. Especially Chris Nelder.
There could be another stacked value for the home battery that soaks up rooftop solar output in the day, uses it in early evening to discharge on a “balance Amps” strategy to keep the home feeder draw under 100A so the main panel (and grid) don’t have to be upsized (for all the new electrification of EV, and heat pumps etc), then discharges the rest into the main panel on a TUO schedule or grid signal while the car charger is providing flexible charging on a grid signal.
The key issue is making sure that EVs are not re-charged with fossilized electricity at charging stations that the electric utilities are pushing to help compensate for their lost customers.
Too much focus has been on batteries , when EV chargers can be as ubiquitous as gas stations and in every motel parking lot . I am an early investor in ENVISION SOLAR http://www.envisionsolar.com Their EV chargers are powered by a solar panel , they are designed to fit into a standard parking space, no permitting , no digging , they just roll of a flatbed truck and set up in minutes. They can charge an EV faster than any other charging station with solar elelctricty!
Amory : please check this company out in San Diego !
Can some one explain how “behind the meter” plus storage is more expensive to the rate payer than “utility scale” pv.
First, the entire cost of a “behind the meter” system is borne by the system installer, not the ratepayer. Second, electric generated and utilised by “behind the meter” system is creating space in the grid for the rate payer to increase their demand at no cost to them. Third, the utility is effectively acquiring generation at absolutely no cost to them in generation or distribution terms.
Yes. Utilities must still maintain the lines and wires to deliver electricity to a customer home or business. Under net metering residential customers who own PV typically export power during the day, and use it in the morning and evening. Even with batteries, there is typically an importing and exporting of power, because few battery systems make the owner entirely independent from the grid.
So while the cost is borne by the ratepayer of installing the system, there are other costs that the system owner does not pay for. Additionally, the utility is typically not able to utilize customer-sited batteries to provide services to the grid.
However, when customer-sited batteries can be utilized by utilities, they can provide substantial benefits to the grid and other ratepayers.