CMU study explores optimizing PHEV design and allocation to minimize life cycle cost, petroleum consumption, and GHG emissions
|Optimal three-segment vehicle design and allocation for various scenarios. Source: Shiau et al. Click to enlarge.|
Researchers at Carnegie Mellon University (CMU) constructed an optimization model to determine the optimal vehicle design and allocation of conventional, hybrid, and plug-in hybrid vehicles to drivers in order to minimize life cycle cost, petroleum consumption, and GHG emissions.
The study by Dr. Jeremy Michalek and colleagues found that while a fleet with plug-in hybrid electric vehicles (PHEVs) with universal high electric range (i.e., 87 miles all electric) minimizes petroleum consumption, minimum lifecycle greenhouse gas (GHG) emissions are achieved with a mix of low-range (~25 miles) and mid-range (40-50 miles) PHEVs. Minimum life cycle cost is achieved by assigning low-range (~15–25 miles) PHEVs to the ~75% of drivers who travel less than ~50 miles/day and hybrid electric vehicles (HEVs) to drivers who travel further.
The study found that optimal allocation of vehicles to drivers appears to be of second-order importance for net life cycle cost and GHG emissions compared with an overall shift from conventional vehicles (CVs) to HEVs or PHEVs.
Under our base case assumptions, life cycle costs and GHGs of HEVs and PHEVs are comparable, particularly for drivers who charge frequently, and the least-cost solution is sensitive to the discount rate and the price of gasoline, electricity, and batteries. Relative to our base case of $3.30/gal gasoline, $0.11/kW h electricity, $400/kW h Li-ion batteries, $600/kWh NiMH batteries, and 5% discount rate, PHEVs are part of the least-cost solution for gas prices above $2.6/gal, electricity prices below $0.16/kW h, Li-ion battery prices below $590/kW h, or nominal discount rates below 11%.
At a 10% discount rate, Li-ion pack cost must fall below $410/kWh for PHEVs to be part of the least-cost solution. Consumers are often observed to use discount rates above 10% in practice, so battery pack costs significantly below $400/ kWh may be needed to drive mass consumer adoption unless gasoline prices rise.—Shiau et al.
The study found that life cycle cost and GHG emissions are minimized using high battery swing (above 60%) and replacing batteries as needed, rather than designing underutilized capacity into the vehicle with corresponding production, weight, and cost implications. This operating strategy contrasts with the current practice of restricting swing to values near 50% to improve battery life, the team points out.
Allowing optimized swing rather than restricting swing to 50% reduces life cycle cost and GHGs of PHEVs by about 1–2% in our model—small enough that other factors such as logistics, customer satisfaction, regulation, and incentives may play a significant role in determining battery swing in PHEV design. Current incentives for PHEVs, such as those outlined in the ARRA, provide subsidies based on battery size, rather than usable battery capacity, all-electric range, or effective GHG reduction. This encourages more PHEVs with larger battery packs but results in increased social costs and could produce unintended incentives for battery swing selection.
PHEV battery subsidies are likely only economically justified as a temporary stimulus if battery and energy costs are expected to quickly reach levels that make PHEVs cost competitive with HEVs over the life cycle.—Shiau et al.
Carbon allowance prices offer little leverage for improving cost competitiveness of PHEVs, according to the study. PHEV life cycle costs must fall to within a few percent of HEVs in order to offer a cost-effective approach to GHG reduction. A paper on their work was published in the ASME Journal of Mechanical Design
A 2009 CMU study found that when charged frequently (every 20 miles or less), using average US electricity, small capacity (i.e., lower all-electric range) PHEVs are less expensive operationally and release fewer greenhouse gases (GHGs) than hybrid-electric (HEVs) or conventional vehicles. (Earlier post.)
For this latest study, the CMU team developed a “benevolent dictator” optimization model integrating vehicle physics simulation, battery degradation data, and US driving data. The model identifies optimal vehicle designs and allocation of vehicles to drivers for minimum net life cycle cost, GHG emissions, and petroleum consumption under a range of scenarios.
The study compared conventional and hybrid electric vehicles (HEVs) to PHEVs with equivalent size and performance (similar to a Toyota Prius) under urban driving conditions. The study focused on a split configuration because of its flexibility to operate similarly to a parallel or series drivetrain, and adopted an all-electric control strategy, which disables engine operation in charge-depleting (CD) mode, drawing propulsion energy entirely from the battery until it reaches a target state of charge (SOC).
Ching-Shin Norman Shiau, Nikhil Kaushal, Chris T. Hendrickson, Scott B. Peterson, Jay F. Whitacre, and Jeremy J. Michalek (2010) Optimal Plug-In Hybrid Electric Vehicle Design and Allocation for Minimum Life Cycle Cost, Petroleum Consumption, and Greenhouse Gas Emissions. J. Mech. Des. 132, 091013 doi: 10.1115/1.4002194