LAUSR

ABSTRACT  To understand the spatial and temporal distribution characteristics of the NOx emissions of urban buses, actual road NOx emissions of the buses in Kunming City were obtained by an onboard monitoring platform. A method combining Bayesians network and probabilistic inference was used to fill in the missing data so as to form a complete data set. NOx emission heat map was generated on the basis of complete data set, and the spatial autocorrelation analysis method was used to study the spatial and temporal distribution characteristics of NOx in the test process. The results showed that Bayesian networks and probabilistic reasoning methods have high accuracy in filling in missing data. The spatial autocorrelation analysis found that the spatial autocorrelation indices for morning, noon, afternoon, and evening were respectively 0.648, 0.836, 0.935, and 0.798. NOx emissions were spatially correlated at all four time periods, and pollution emissions were spatially aggregated. The heat map showed that the highest-concentration times for NOx emissions were noon and afternoon. At each time fraction, highest emissions accumulated at road sections 1–3 and 6–9, and high emission-intensity growth rates were found at road sections 5–9.