Going to 100% electric cars would increase electricity demand by around one-quarter. Moving all road transport to electric would increase it by around 40%.
Interesting analysis. I like it. A few years ago when I was planning to install solar in Albuquerque, NM, I oversized the system by 35% to allow for the addition of my plug-in hybrid. After two years of owning the car, it looks like I did a good job of estimating because my total electricity use almost exactly matches my electricity production.
As ever, this is an excellent analysis and very helpful in convincing sceptics. I think you could improve it slightly by considering two factors. The Zoe is not a typical EV and will soon be out of production. Many of the new models are, unfortunately, less efficient. The EV database shows the Zoe to be under the average model which is 306Wh/ mi. Of course it depends on what the average of those on the road are. I wasn’t clear too if you have allowed for charging losses. I think the average is about 10-15% which is my experience on a Niro EV. (The Zoe is the worst with 24% losses I recall.). It won’t make much difference to the main point, but may help defend your case to sceptics if everything is covered.
Great analysis, would be interesting if you could add in capacity factors and actual installed capacity needed in a future post as well. Capacity favors of renewables and wind are lower, so the implied installed GW will be higher (assuming no batteries). Actual installed capacity in GW would give you the capital investment needed to meet that demand and the actual cost. Think the Lazard LCOE report has most of these numbers.
Too much of this analysis is apples to oranges, ignoring the very significant difference that petrol/diesel/CNG are sources of stored energy available on demand, while electricity needs to be generated in something close to real time. The bigger practical bottleneck to an all-EV transition will involve building generating capacity that can match peak demand.
The workaround to this problem, of course, is storage, via chemical battery or otherwise. A more instructive follow-up post would involve estimating the additional storage capacity required, alongside the generation capacity. Looking at total energy quantities makes sense for liquid fossil fuels, which can be treated as a stock, but electricity is an energy source that is better considered as a flow.
I recommend you be more explicit "energy efficiency" in your writing, distinguishing heat energy from work energy. For example, you write "Buses use 14 TWh of diesel. Going electric, they would use just 4 TWh.", but this is misleading. An ICE bus uses 14 TWh of heat energy, and its ICE converts it to work energy with an efficiency ranging about 33% to ~45%. [There are additional losses in the power train.] The electric bus uses ~4 TWh of electric energy, converted to work energy at high efficiency by an electric motor, say 98%. But that 4 TWh was likely sourced from heat energy in a thermal power plant, which might have a work/heat conversion energy of 33% (typical), or 59% (latest CCGT running steadily), or 25% (wood burning Drax). Wind and solar sources generate work energy without such conversion losses. [The "primary energy" concept is ill conceived.]
Nicely laid out, and well explained. Thanks for taking the time to run the numbers. To the peak demand point, EVs have so much charging flexibility that serving them with efficient, low carbon resources would not be too challenging given proper managed charging rates.
Would be interested to see a similar analysis on heating! Heat pumps are also very efficient, but much less flexible...
Great article Hannah. Really hits on the key point that transitioning to EVs will significantly reduce overall energy demand. The peak demand issue is a big challenge potentially, but not insurmountable.
Energy use could be cut much further if journeys under 10 km were by electric bike... However, as this is UK, worth factoring in electric car batteries as as a grid battery backup, and that the intermittency of renewables is helped by the alternation between wind and solar (in the UK, solar limited by latitude, but often windy when not sunny).
I suspect this only matters at the margins, but how significant are household's (i.e. not utility-grade) solar panels?
I have solar panels on my house, which I got just before the feed-in-tariff was abolished, so I get quarterly payments based on a guess that half of the electricity my panels generated ended up fed back to the grid. But now that I have an electric car, I charge it during the day when there's enough excess power, which means that (a) I have no running costs, but (b) I'm feeding less power back into the grid than the policy expected.
How much does this sort of thing affect the overall numbers, both for how much energy we expect cars to consume from the overall budget (good), and how much power the grid will have available (bad)?
I’m interested in the need for charging stations and infrastructure needed for a full turn over to all electric vehicle. And how long it would take realistically? What would be the cost for: power grid, lines, charging stations, even payment systems?
If you have already done this review someone please let me know.
I do realize each country and electric grid system would be different but a UK analysis would be a starting point.
The problem with this kind of analysis is that it’s not about Generation it’s about Transmission and mostly Distribution. Ask the question “how many more chargers are needed”, “how many more substations and transformers are needed” and how long does it take to plan, permit, source and build all of these. We are seeing this problem in the US today. It’s reasonably easy to put a charger in a detached house garage and solar on the roof and electrify an EV. But it is much harder to do that in an apartment building in a dense city and incredibly hard to get the electricity needed to charge a depot of delivery trucks or municipal pickup trucks or a bus depot, let along a long-haul truck stop.
The 40% efficiency for natural gas that you site is for simple-cycle natural gas. It is important to note that new Combined-cycle natural gas turbines are around 65% efficient. Plus they can run 24/7, something that wind /solar can never do.
I wonder how much electricity demand would be reduced as the internal combustion is phased out? How much is currently used in the gas/petroleum/coal distribution infrastructure that would be eliminated in a switch to EVs?
Interesting analysis. I like it. A few years ago when I was planning to install solar in Albuquerque, NM, I oversized the system by 35% to allow for the addition of my plug-in hybrid. After two years of owning the car, it looks like I did a good job of estimating because my total electricity use almost exactly matches my electricity production.
As ever, this is an excellent analysis and very helpful in convincing sceptics. I think you could improve it slightly by considering two factors. The Zoe is not a typical EV and will soon be out of production. Many of the new models are, unfortunately, less efficient. The EV database shows the Zoe to be under the average model which is 306Wh/ mi. Of course it depends on what the average of those on the road are. I wasn’t clear too if you have allowed for charging losses. I think the average is about 10-15% which is my experience on a Niro EV. (The Zoe is the worst with 24% losses I recall.). It won’t make much difference to the main point, but may help defend your case to sceptics if everything is covered.
Great analysis, would be interesting if you could add in capacity factors and actual installed capacity needed in a future post as well. Capacity favors of renewables and wind are lower, so the implied installed GW will be higher (assuming no batteries). Actual installed capacity in GW would give you the capital investment needed to meet that demand and the actual cost. Think the Lazard LCOE report has most of these numbers.
Too much of this analysis is apples to oranges, ignoring the very significant difference that petrol/diesel/CNG are sources of stored energy available on demand, while electricity needs to be generated in something close to real time. The bigger practical bottleneck to an all-EV transition will involve building generating capacity that can match peak demand.
The workaround to this problem, of course, is storage, via chemical battery or otherwise. A more instructive follow-up post would involve estimating the additional storage capacity required, alongside the generation capacity. Looking at total energy quantities makes sense for liquid fossil fuels, which can be treated as a stock, but electricity is an energy source that is better considered as a flow.
I recommend you be more explicit "energy efficiency" in your writing, distinguishing heat energy from work energy. For example, you write "Buses use 14 TWh of diesel. Going electric, they would use just 4 TWh.", but this is misleading. An ICE bus uses 14 TWh of heat energy, and its ICE converts it to work energy with an efficiency ranging about 33% to ~45%. [There are additional losses in the power train.] The electric bus uses ~4 TWh of electric energy, converted to work energy at high efficiency by an electric motor, say 98%. But that 4 TWh was likely sourced from heat energy in a thermal power plant, which might have a work/heat conversion energy of 33% (typical), or 59% (latest CCGT running steadily), or 25% (wood burning Drax). Wind and solar sources generate work energy without such conversion losses. [The "primary energy" concept is ill conceived.]
Nicely laid out, and well explained. Thanks for taking the time to run the numbers. To the peak demand point, EVs have so much charging flexibility that serving them with efficient, low carbon resources would not be too challenging given proper managed charging rates.
Would be interested to see a similar analysis on heating! Heat pumps are also very efficient, but much less flexible...
Great article Hannah. Really hits on the key point that transitioning to EVs will significantly reduce overall energy demand. The peak demand issue is a big challenge potentially, but not insurmountable.
Energy use could be cut much further if journeys under 10 km were by electric bike... However, as this is UK, worth factoring in electric car batteries as as a grid battery backup, and that the intermittency of renewables is helped by the alternation between wind and solar (in the UK, solar limited by latitude, but often windy when not sunny).
I suspect this only matters at the margins, but how significant are household's (i.e. not utility-grade) solar panels?
I have solar panels on my house, which I got just before the feed-in-tariff was abolished, so I get quarterly payments based on a guess that half of the electricity my panels generated ended up fed back to the grid. But now that I have an electric car, I charge it during the day when there's enough excess power, which means that (a) I have no running costs, but (b) I'm feeding less power back into the grid than the policy expected.
How much does this sort of thing affect the overall numbers, both for how much energy we expect cars to consume from the overall budget (good), and how much power the grid will have available (bad)?
Like the article. Not preaching just facts.
I’m interested in the need for charging stations and infrastructure needed for a full turn over to all electric vehicle. And how long it would take realistically? What would be the cost for: power grid, lines, charging stations, even payment systems?
If you have already done this review someone please let me know.
I do realize each country and electric grid system would be different but a UK analysis would be a starting point.
The problem with this kind of analysis is that it’s not about Generation it’s about Transmission and mostly Distribution. Ask the question “how many more chargers are needed”, “how many more substations and transformers are needed” and how long does it take to plan, permit, source and build all of these. We are seeing this problem in the US today. It’s reasonably easy to put a charger in a detached house garage and solar on the roof and electrify an EV. But it is much harder to do that in an apartment building in a dense city and incredibly hard to get the electricity needed to charge a depot of delivery trucks or municipal pickup trucks or a bus depot, let along a long-haul truck stop.
The 40% efficiency for natural gas that you site is for simple-cycle natural gas. It is important to note that new Combined-cycle natural gas turbines are around 65% efficient. Plus they can run 24/7, something that wind /solar can never do.
Are you assuming 1:1 replacement for all UK ICE transport? That seems very unlikely.
TaaS would drastically reduce the ownership of EV's and # of EV's on the road. e.g.
https://www.clubofrome.org/publication/earth4all-ahmed/
I wonder how much electricity demand would be reduced as the internal combustion is phased out? How much is currently used in the gas/petroleum/coal distribution infrastructure that would be eliminated in a switch to EVs?
Good analysis. Now look at how much more expensive it is to run the grid on gas compared to offshore wind: https://www.energyflux.news/p/less-offshore-wind-more-gas-more