Urban Bus Emission Characteristics And Models

Controlling motor vehicle pollution emissions and improving the environmental quality of cities has become a consensus.Urban public transport has the characteristics of large passenger capacity,high efficiency and ease of traffic congestion.However,its exhaust emissions cannot be ignored due to long-term idle operation and frequent acceleration and deceleration.At present,there is a lack of comprehensive systematic research on the current bus emissions for various fuel types in China,and there is still much research space for accurately estimating the real-time emission models for different types buses.This paper aims to establish a neural network model for estimating bus emissions considering time memory by analyzing the emission characteristics and influencing factors of the four types including gas-electric hybrid electric buses?GEHE buses?,compressed natural gas buses?CNG buses?,EURO 4 heavy-duty diesel engine buses?EURO 4 buses?,and EURO 5 heavy-duty diesel engine buses?EURO 5 buses?.The forecasting model lays the foundation for effective control of bus emissions and emission simulation.Based on this purpose,the experimental scheme was first designed.The GPS data of four types of buses were collected in Zhenjiang City,Jiangsu Province,and the data was pre-processed,integrated into the second-by-second emission data.The sample sizes and statistical indicators of speed,acceleration,and CO,HC,NOx,and CO₂ emission rates for four fuel types of buses was calculated.Secondly,combined with the emission mechanism and the operating characteristics of buses,this paper analyzed the spatio-temporal characteristics of different types bus emissions,the emission differences under different driving conditions,the vehicle specific power?VSP?characteristics,the difference in emission characteristics at road sections,bus stops,and intersections.Next,this paper qualitatively analyzes the effects of speed,acceleration,and passenger mass on the emissions of CO,CO2,HC,and NOX for four fuel types of buses.Then,the Pearson coefficient and the Spearman coefficient were used to quantitatively analyzed the linear and non-linear relationship between the three factors and the bus emissions.The results showed that speed,acceleration and passenger quality are the key factors affecting the bus emissions.Finally,a long short-term memory network?LSTM?method for estimating CO,CO2,HC,and NOx emission rates for GEHE,CNG,EURO4,EURO5 buses was proposed.And VSP based polynomial regression models were developed for comparison.The prediction accuracy of the proposed models was verified in microscopic and aggregate levels.