Arkiv / Fordon

Analysis of advanced battery-electric long haul trucks

Creative commons license

Skrivet av Lew Fulton och Andy Burke

Battery-electric (American class 8) long-haul tractor-trailer trucks are being considered for commercialization by at least several companies, such as Tesla.  These trucks can have many attractive features, but there are questions concerning whether the trucks can meet range requirements for long-haul trucking with heavy payloads.  The Tesla truck, for example, is claimed to be able to cover 800 kilometers on a single charge (a typical criteria for fleets).

There are major questions around whether such a long-haul truck can be designed to be battery electric and also maintain its functionality, such as its payload. The concept has been roundly criticized on the basis that the needed batteries would be extremely heavy and recharging times too long to work for most long-haul trucking applications.  UC Davis has recently finished a report (here) on this topic where we have simulated this truck, in terms of what it would look like, weigh, cost, and – very importantly – how much energy per mile it would likely use in service and how this would affect its viability. This report sheds important light on the question.

What we did:

We analyzed the performance and costs of an 800 km heavy duty (such as US Class 8 tractor trailer) battery-electric truck, operating on level roads but also on uneven terrain, using test cycles for California.  We used specifications for this truck that are close to what Tesla has announced for their planned truck of this type.

We used our vehicle travel simulation model, ADVISOR, with associated cost calculations, to undertake this exercise. While the particular comparison is made to the Tesla truck, the results are more broadly applicable to any class 8 electric truck.  For this truck we looked at how its weight and other attributes affect its on-road energy use and thus energy cost. We also considered a 500 km variant, which has the shorter range but also much less battery weight than the base truck, and somewhat better efficiency. We considered this truck with the assumption that it would have to be recharged once during the day to become an 800 km range truck.

What we found:

In short, we found that this BEV long-haul truck might be viable but there would certainly be compromises to payload capacity (due to battery weight), while it is likely to be more expensive than a comparable diesel truck on a “total cost of ownership” basis. Though there are scenarios where it may be cost competitive.

We found that the performance of fully loaded long haul BEV trucks (energy consumption, range and acceleration times) are generally consistent with claims made by Tesla for their truck if the truck is operating on level roads. However, operating on roads with varying grades, the performance, including efficiency (km/kWh) and range, will be significantly reduced.

Specifically:

  • We estimate that the required battery pack in the level road 800 km electric truck will store 1418 kWh of electrical energy (1134 kWh useable). The battery will weigh about 7 tonnes, reducing a 25 tonne payload by 28%. Further, we estimate the battery will last 5 years, sustaining 1500 charge/discharge cycles in that time.
  • The economics of the truck operation are dependent primarily on the unit cost ($/kWh) of the batteries and their cycle life, the energy efficiency of the truck in real world operation, and the cost of electricity.
  • The cost calculations indicate that for battery pack costs close to $100/kWh and $0.10/kWh electricity, the higher purchase price of the 800 km electric truck compared to a diesel truck (diesel price $0.92/liter) can be recovered from fuel/electricity cost savings in about 6 years if the trucks are operated 200,000 km per year.
  • The average per-mile cost of the electric truck ($.32/km) over 5 years is close to that of the diesel truck for level road operation. This assumes a relatively low electricity and battery costs. At higher electricity costs or battery costs, the results are less favorable for the electric truck.

Our exploration of real-world driving cycles shows that a significant reduction of efficiency, and thus range, can occur even with modest grades, such as with slightly hilly terrain. With level conditions, a range of 800 kilometers may be possible with an energy use of the truck is about 2 kWh/mile (1.25 kWh/km), but in any conditions with greater energy use, range of course drops. In our variant driving scenarios the energy use per km can rise by 50%, cutting the driving range by a third. This is also true for diesel trucks, but since they can refuel in just a few minutes, this loss in range is not a significant concern.

We consider a range of cost scenario, as shown in the figure and accompanying table. This figure assumes the higher energy use case for both the diesel and BEV long haul trucks, and shows cases with variations in battery cost, electricity cost, and includes a variant with lower range and lower battery cost.

The operating cost of the electric truck using high cost batteries ($200/kWh) and high cost electricity ($0.15/kWh) is well above that of diesel; with lower cost batteries ($100/kWh) it gets close to that of diesel, and with both low cost batteries and low-cost electricity it pretty much matches diesel cost per mile.

As shows in the last two bars, fast charging to accommodate high energy use seems like a good approach. This would require fast charge stations for heavy-duty battery-electric trucks along the roadway system. The economics of lower battery capacity with daily recharging can be reasonable if both the battery and electricity costs are low.

Table 1. Operating costs for electric and diesel trucks for various battery and electricity costs for using 1.86 kWh/km, diesel truck 0.4 L/km

Figure 1. Sensitivity cases for lower energy efficiency trucks using 3 kWh/mi (1.86 kWh/km), 6 mpg diesel (0.4 L/km)

Conclusions

The primary limitations of the battery- electric long-haul truck are loss in payload (estimated at around 30% in the US context) and driving range that can be expected in real world operation using a reasonable weight and volume of batteries. A range of 800 kms may be possible if the energy use of the truck is about 1.25 kWh/km or less, but if the energy use is 1.8 kWh/km or greater, which appears likely in many real-world trips, an 800 km range is not realistic even using batteries with cell energy densities of 400-500 Wh/kg. Fast charging (in 1 hr. or less) would allow the use of smaller, less costly batteries and permit the accommodation of the large differences in the energy efficiency (km/kWh) of the electric trucks in different regions and terrains.

References

  1. Palmer, D., “What is the Tesla Semi? Everything you want to know about Tesla’s semi-autonomous electric truck”, March 22, 2018, https://www.zdnet.com/article/what-is-the-tesla-semi-everything-you-need-to-know-about-teslas-semi-autonomous-electric-truck/
  2. Ritter, K., “The Tesla Semi Costs-Part 4”, Feb. 2, 2018, https://insideevs.com/news/336275/the-tesla-semi-costs-part-4/
  3. Ferris, D., “Tesla Semi-competitor Nikola set to unveil battery only Semi-truck at Launch Event”, April 2019, https://www.teslarati.com/tesla-semi-competitor-nikola-motor-battery-only-trucks/
  4. Burke, A.F., Miller, M., and Zhao, H., Fuel Cells and Batteries in Buses and MD/HD Trucks, STEPS 2018 Report, December 2018
  5. Public hearing to consider the proposed innovative clean transit regulation – a replacement of the fleet rule for transit agencies, Staff Report: Initial statement of reasons, Appendix E: Battery Cost for Heavy-Duty Electric Vehicles, August 7, 2018
  6. Collantes, G., Burke, A.F., Miller, M., and Zhao, H., Analytical tool to support the implementation of electric vehicle programs, Research Report UCD-ITS-RR-15-08, 2015, PIER Grant