This report makes an initial investigation into the potential for combining very high penetration levels of electric vehicles with similarly very high penetration of variable renewable electricity (VRE) in California. A literature review is performed regarding the potential for high levels of EV sales and VRE penetration at both the U.S. and California level. Such scenarios have been developed by a number of researchers, such as U.S. national laboratories for the White House (under the Obama Administration), and by Energy and Environmental Economics, Inc. (E3) for the California Energy Commission. Such studies indicate that both of these “extreme” futures are entirely plausible and have the potential to coexist. However, none of the reviewed studies has undertaken detailed analysis of how large numbers of EVs could interact with and support a VRE-dominated system, and how these might interact in a useful way. This could include grid-to-vehicle (G2V) and vehicle- to-grid (V2G) movement of electricity, with vehicle batteries providing large scale electricity storage.
We undertake our own preliminary simulation for a 2030 and 2050 scenario for California, using an 8760 hours (full year) electricity demand profile and VRE generation example. We assume a ramp- up of VRE to 60% of all electricity generation by 2030 and 100% by 2050, with a similar increase in the EV share of new LDV sales, creating a significant stock (about 7 million) by 2030 and nearly complete transition (to over 20 million vehicles) by 2050. Using an “averages, peaks and valleys” analysis on the electric side, and a typical spare battery storage potential on the vehicle side, our simulation shows that by 2030 a large share of excess VRE electricity generation could be stored, and a large share of electricity shortfall from VRE could be provided, by electric vehicle batteries throughout the year, though there would be many cases where they cannot provide full coverage of these situations. However by 2050, if nearly 100% of the fleet were EVs, only about half of their available, spare capacity is needed to store the excess electricity from a full VRE system on the highest generation day and only about 40% would be needed to store and supply the shortage from lack of VRE generation on the highest shortfall day.
While these results are encouraging, a deeper simulation is needed to provide a true hour-by-hour assessment of battery use and the incidence of storage need compared to driving need. Management of charging times that could not be assessed here may also play a critical role. In addition, our initial assessment only covers a single day shortfall. Shortfalls could occur for longer periods, particularly if the VRE electricity system were sized to take better advantage of seasonal storage options. Vehicle batteries are best suited to very short duration storage and may not be adequate to keep the electricity reliable for many consecutive days of shortfall. Hydrogen (H2) has the potential to be a longer-term energy storage option and could be stored in fuel cell vehicle tanks (and the H2 system associated with generating, storing and distributing H2 to those tanks). The next stage of our research will involve running a full simulation using our (ITS-Davis) California ZEV power model (“CALZEV”), a version of the larger Message model, applied to consider both electricity and hydrogen (with large numbers of both of these types of vehicles) in order to: 1) gauge the relative storage potential and cost over a range of time frames and VRE scenarios, and 2) estimate the relative value and possible synergies in a system with both types of vehicles and fuels.