Plugin hybrid electric vehicles (PHEVs) have several attractive features in terms of reduction of greenhouse gas (GHG) emissions. Compared to conventional vehicles (CVs) that only have an internal combustion engine (ICE), PHEVs have better energy efficiency like regular hybrids (HEVs), allow for electrifying an appreciable portion of traveled miles, and have no range anxiety issues like battery-only electric vehicles (BEVs). However, in terms of criteria emissions (e.g., NOx, NMOG, HC), it is unclear if PHEVs are any better than HEVs or CVs. Unlike GHG emissions, criteria emissions are not continuously emitted in proportional quantities to fossil fuel consumption. Rather, the amount and type of criteria emissions is a rather complex function of many factors, including type of fuel, ICE temperature, speed and torque, catalyst temperature, as well as the ICE controls (e.g., fuel-to-air ratio, valve and ignition timing). For the most part, however, criteria emissions are mostly eliminated once the catalyst reaches a certain temperature. As such, most of the criteria pollutant emissions by a vehicle happen during the time between turning on the ICE and until the catalyst warms up. In this work, a counter for cold starts is used as a proxy indicator of criteria emissions. In case of CVs and HEVs, the catalyst is kept warm (after first ICE ignition) for the remainder of a trip. On the other hand, the frequency of ICE cold starts in PHEVs for real-world driving can be complex to model. This work presents a simplified methodology for analyzing large datasets of real-world driving in order to infer estimates for the frequency of PHEV cold starts as function of i) all electric range (AER) and ii) electric-only peak power capability of the powertrain. Results of the study are then summarized as guideline curves for sizing of PHEV powertrains.
