Real world driving emissions from hybrid electric vehicles and the impact of biofuels

In order to reduce CO2 emissions it is important to consider hybrid electric vehicles (HEVs) and the increased blending of biofuels with conventional fuels, whilst ensuring that type approval test procedures are robust enough to protect local air quality. This thesis investigates the behaviour and emissions of an HEV. Modal regulated and unregulated pollutant emissions including particle number (PN) and size distribution (PNSD) were captured. Chassis dynamometer testing, including those following type approval New European Drive Cycle (NEDC) and World-harmonised Light Duty Test Cycle (WLTC) procedures and cycles, are compared with on-road Real Driving Emissions (RDE) and non-compliance, real-world on-road testing. The variations in drive cycle properties are investigated, followed by an investigation of HEV behaviour and the resultant emissions from these different tests. Various characteristics of HEVs that make them different to conventional internal combustion engine vehicles are then studied in unprecedented detail. Finally, the effect of biofuel blends of 5% ethanol (E5), 10% ethanol (E10), and 10% n-butanol (B10) with gasoline (E0) on the HEV emissions is investigated. This work found that the frequency of vehicle stops (and hence engine re-start events) had the greatest impact on HEV CO and PN emission factors, while test cycle distance (and hence cold start penalty) had the greatest impact on HEV CO2 and NOx emission factors, across test cycles. CO2, CO, NOx and PN were 9%, 66%, 58% and 71% lower from the RDE than the WLTC. The urban section of the RDE was found to be unrepresentative of a real world city-centre drive cycle in this regard. The HEV PNSD was dependent on drive style, with engine re-ignition events having a unimodal maximum at 50nm, and constant engine operation having a bimodal PNSD with maxima at 15nm and 110nm. The three-way catalyst (TWC) was found to light-off very fast – within approximately 50s – for the HEV, despite slow warm-up times due to engine inactivity, and displayed good resistance to de-lighting during periods of engine-off. Emissions of CO, PN, ammonia and methane were associated with periods of fuel enrichment on engine re-ignition, while NOx, N2O and other unregulated pollutants were unaffected. Tests fuelled with E10 and B10 had 28% and 22% lower respective CO emissions over the WLTC than E0, with lower vehicle speeds exhibiting the greatest decrease. E10 and B10 led to 24% and 18% respective decreases of NOx over the RDE, while E10 and B10 led to 41% and 21% decreases in PN over the WLTC, compared to E0.