Driving Behavior of Hybrid Bus

摘要

Transportation consumed 71% of oil consumption and produced 27% of greenhouse gas in the United States in 2012. In addition, transportation also accounts for 50 %, 31.9%, 21.5%, and 1% in carbon monoxide, nitrogen oxides, volatile organic compounds, and sulfur dioxide respectively in the U.S. Due to soaring fuel prices and environmental concerns, hybrid vehicle technology attracts more and more attentions in recent years. The electric motor on hybrid bus converts braking power into electricity during deceleration, which could be used later during acceleration. Therefore, the research hypothesis is that the driver\u27 driving behavior are closely related to the amount of electricity generated, which is directly related to fuel economy. Therefore, this thesis designed the study to test the variability in driving behavior parameters. However, the impact of those driving behavior parameters on fuel economy is recommended to be investigated in future research.In order to measure the bus driving activities, six GPS data loggers were installed on three hybrid buses and three control buses. The data was collected on ten weekdays from November 29 to December 12, 2011. Two routes were chosen in this study, which are arterial route and campus route. Several variables were created to characterize driving behaviors, including acceleration, deceleration, and vehicle specific power (VSP), etc. Nonparametric analysis of variance method was used to test the variability in driving behavior parameters. The results showed that the driver had the dominant impacts on most driving behavior parameters. The comparison test also found the hybrid buses accelerated slower than regular diesel bus. In addition, the regression model was also built to fit the same dataset. The model results from both nonparametric method and regression method did not agree with each other for some driving behavior variables since they used different model estimation techniques. It is recommended to draw conclusion based on nonparametric model because it requires fewer assumptions with more statistical power. In conclusion, this study found the driving behavior was statistically different among drivers, and it is recommended to evaluate how those differences in driving behavior affect the fuel economy and emissions of hybrid buses in future research.