February 29, 2016

Will automation benefit electric vehicle efficiency?

Lots of people have been asking how the results of our analysis of automated vehicles and energy demand might be different if cars are largely electric instead of gasoline-powered in the future. Last week I posted about our new article in which we tried to place some bounds on the potential energy and carbon implications of automated vehicles. The article got lots of great press, but one question we weren’t expecting was how our results might change if we are all driving electric vehicles (EVs) in the future.

Before addressing this question, the first thing to note, which we did address in the article, is that full automation could well make EVs more attractive. First, cars that can drive themselves to refuel without the driver could get by with a much sparser (i.e. cheaper) alternative fuel infrastructure than would be necessary if a human is going out of their way. Second, shared robo-taxi fleets would rack up miles faster than privately owned vehicles, improving the payback period on investments in vehicles that use lower-cost alternative fuels. Third, whereas many private car buyers expect investments in efficiency to pay off within just a couple of years, fleet operators would presumably take more of a “green eyeshades” view of investments in lower operating costs. All of these factors seem to point towards automation making EVs more attractive.

But back to the question at hand: how would EVs affect the relative impacts of automation? I don’t have a dispositive answer just yet, but here are some initial thoughts. One important difference between EVs and internal combustion engine vehicles (ICEVs) is the relationship between speed (how fast the engine is spinning), load (how hard the engine is working), and efficiency. In an ICE, efficiency is strongly dependent on speed and load:

Typical internal combustion engine maphttp://www.mdpi.com/1996-1073/6/1/117/htm

As you can see from the above “engine map”, the amount of fuel burned per unit of work done (shown by the contour lines) by the engine can pretty easily vary by a factor of 3-4, depending on speed and load. This means that engine efficiency is strongly dependent on vehicle speed:

Berry, I. M. (2010).

In contrast, electric vehicle motors are much less sensitive to speed and load. As shown in the figure below, the Leaf EV’s motor efficiency remains above 80% for a wide range of speed and load conditions. (Moreover, a drop in efficiency from 95% to 80% only increases energy use by a factor of 1.2, whereas a drop from 25% to 10% efficiency increases energy use by a factor of 2.5).

EV motor efficiency is much less sensitive speed and load than are internal combustion engines.Tim Burress, Oak Ridge National Lab

What does this all mean? For one, the benefits of congestion mitigation and eco-driving practices are likely to be less pronounced for EVs than for ICEVs. There is simply less opportunity for EVs to reduce energy consumption by avoiding low-efficiency operating conditions. Second, EVs are likely to realize larger benefits from platooning and to be more sensitive to higher-speed highway travel than are ICEVs. This is because as aerodynamic loads increase on an ICEV, there is a countervailing increase in engine efficiency that mitigates the overall change in per-mile energy consumption. Since EV efficiency is less sensitive to speed and load, its overall energy consumption will be more sensitive to aerodynamic loads. Other factors, such as lightweighting enabled by improved crash avoidance and vehicle right-sizing, should be about the same for ICEVs and EVs. Same goes for energy penalties of additional comfort & convenience features.

One wildcard here is the tradeoff between acceleration performance and energy consumption for EVs – I’ve looked at this for ICEVs, but am not sure how it would play out for EVs. Feel free to send an email or post a comment if you have any thoughts.