Battery swapping vs. fast charging study outline
I'm looking for collaborators (and early critique!) on a battery swapping vs. fast charging study.
I want to at least try to come up with quantitive assessments of battery swapping vs. fast charging, circa 2050, in Europe. One of the main arguments for fast charging is that it is compatible with the so-called "structural battery" design, so I will consider that, too.
Assumptions
I assume a stable population, stable mobility demand (both cargo tonne- and passenger-kilometres), 100% electric vehicle penetration (actually, it won't be 100% by 2050 and therefore will not be stable yet, so this assumption is a simplification).
I assume 100% of cargo rides and 90% of passenger rides to be autonomous. I assume road accidents to be down 95% from the current level, thanks to autonomous driving, and battery accidents (fires) that lead to injuries to be down 90%, thanks to better anomaly detection and alerting.
However, I assume that fast charging will still shorten battery life significantly (how much exactly I haven’t quantified yet), that is, we won’t find a “magic” cell chemistry whose ageing speed doesn’t depend on the charging rate at all. Likewise, I assume that cells will still degrade when they are inactive (calendar ageing). (However, is this valid with solid-state, Li-metal anodes? I’m not sure.)
I assume that autonomous heavy trucks and robotic warehousing operations could happen 24/7, much more so than today. I assume light-duty vehicle and passenger ride patterns to not change much.
Metrics
Here’re the metrics that I came up with:
Trucks, light-duty vehicles:
Amortised cradle-to-grave energy usage per cargo tonne-kilometre, Wh/tkm
Amount of city land needed to support a certain amount of cargo tonne-kilometres, m2/tkm
Amount of cobalt, lithium, or phosphorous needed to support a certain amount of cargo tonne-kilometres, kg/tkm
Private cars, shared mobility:
Amortised cradle-to-grave energy usage per “useful” passenger-kilometre, Wh/pkm
Proportion of “useful” passenger-hours (spent covering “useful” passenger-kilometres), dimensionless. “Useless” passenger-hours are spent driving towards/away from charging or battery swapping stations, waiting, and charging or swapping the battery. Applies only to private cars and carsharing, but not autonomous car-sharing or taxis.
Amount of city land needed to support a certain amount of “useful” passenger-kilometres, m2/pkm
Amount of cobalt, lithium, or phosphorous needed to support a certain amount of “useful” passenger-kilometres, kg/pkm
Injuries (due to car crashes and fires) per “useful” passenger-kilometre, 1/pkm
I don’t include any economic metrics because I think it’s both practically impossible to estimate prices of anything so far in the future and that the economics are going to heavily depend on policy (tax incentives, subsidies, etc.), (geo)politics (especially affecting raw material prices), and capitalistic capture (e. g., all automakers are apparently planning to charge a lot for autonomous driving and extract huge profits from it, despite purely energetic and even development cost of it is going to be minuscule). All of this is impossible to predict.
I also don't consider the supposed positive effect of battery swapping on grid balancing because I already think the effect will be very minor (to help to grid balancing meaningfully through battery swapping, batteries should reside at swapping stations for at least 12 hours, that is totally unrealistic, both per the Chinese experience, and from common sense), and also because it's hard to predict how elastic the electricity demand will be in 2050. Also, the energy generation patterns will be drastically different in different countries (e. g., nuclear France vs. solar in Spain vs. wind in the UK), but I want to make the analysis less country-specific.
Resources
Some papers that I’m going to use in the analysis:
Calendar Aging of Lithium-Ion Batteries (2016)
Sweden-China Bridge - Battery-Swapping 1.0 (2021)
Sweden-China Bridge - Exploring Battery Swapping for Heavy Trucks in China 1.0 (2021)
Why Did Better Place Fail?: Range anxiety, interpretive flexibility, and electric vehicle promotion in Denmark and Israel (2016)
Enabling fast charging - Battery thermal considerations (2017)
Why we need battery swapping technology (2021)
Shared autonomous electric vehicle service performance: Assessing the impact of charging infrastructure (2020)
Method and publishing
I want to avoid doing agentic simulations myself, maybe taking some insights from existing simulations (if needed), and otherwise consider more of a static picture, accounting for daily mobility and recharging patterns.
I want to publish a Jupyter notebook, but if you want to publish a paper I'm open to it.
Collaboration and critique
If you want to collaborate on this, please write to me on LinkedIn or at leventov.ru@gmail.com.
If you have any feedback, suggestions, or critique on the outline above, including pointing me to existing resources or maybe existing studies of this kind that make it pointless, please share it, too!