For 41 gasoline LDPVs, the average fuel consumption and total
CO2 emission factors are 11.1 2.7 l/100 km and 257 63 g/km,
respectively, under the average on-road driving pattern for LDPVs.
Some older gasoline LDPVs are significantly higher in indirect CO2
emissions, due to poor emission control for CO and THC emissions.
For 16 diesel LDPVs, the average fuel consumption and total CO2
emission factor are 6.6 1.7 l/100 km and 178 46 g/km, which
could save 22% fuel use by energy value compared to their gasoline
counterparts with similar model years. For 3 LPG taxis, their
average total CO2 emission factor is 297 10 g/km under the
average driving pattern. The on-road fuel consumption under the
average on-road driving pattern is approximately 10% higher than
those normalized to the NEDC. Furthermore, the NEDC-normalized
on-road fuel consumption is significantly higher by 30 12%
compared the type-approval fuel consumption for some tested
gasoline LDPVs. As a result, a test cycle with more real-world
driving features is of great necessity to narrow the gap between
on-road and type-approval fuel consumption values of LDPVs. It
should be pointed out that average fuel consumption of LDPVs can
vary significantly by vehicle models. Therefore, on-road data
covering statistically sufficient vehicle models are required to
provide a more precise image of real-world fuel consumption for
LDPV fleet in China.
We identified very strong correlations (R2 ¼ w0.9) between
relative fuel consumption of LDPVs and average speed of generated
micro-trips, which are best fitted with power functions. Statistical
analysis indicates those functions are able to precisely and quantitatively
estimate the effects of the change of average speed within
10 km/he60 km/h from the link-level to the net-level. As a case
study of the UAB illustrates, estimated impacts on relative fuel
consumption can reach approximately 30% to þ40% due to daily
fluctuations in hourly average speed of LDPVs within the UAB.
Meanwhile, estimated impacts from annual variations in average
speed on relative fuel consumption would reach w5%. Therefore,
the traffic conditions cannot be ignored in the discussion regarding
real-world fuel consumption. The effort to improve fuel economy
by encouraging manufacturers to improve vehicle energy-saving
technology could almost be offset by the deterioration of driving
conditions for some traffic-populated areas. Therefore, it is