A reduced PM index for evaluating the effect of fuel properties on the particulate matter emissions from gasoline vehicles. (1st October 2019)

• Record Type: Journal Article
• Authors: Wu, Taoyang; Yao, Anren; Feng, Jun; Wang, Hui; Li, Zhuangzhuang; Liu, Mingkuan; Yao, Chunde
• Is Part Of: Fuel. Volume 253(2019)
• Page Start: 691
• Page End: 702
• Publication Date: 2019-10-01
• Subjects: PM particulate matter -- GDI gasoline direct injection -- PMI PM index -- RPMI reduced PM index -- MLR multivariate linear regression -- PN particle number -- NEDC New European Driving Cycle -- OEMs original equipment manufacturers -- SI spark ignition -- PFI port fuel injection -- DPF diesel particulate filter -- DBE double bond equivalent -- ANOVA analysis of variance -- ECU electronic control unit -- TWC three way catalysts -- RH relative humidity -- VIF variance inflation factor -- CVS constant volume sampler -- the US the United States of America -- MTBE methyl tert-butyl ether; Particulate matter -- Gasoline vehicles -- Distillation characteristic -- Reduced PM index; Fuel -- Periodicals; Coal -- Periodicals; Coal; Fuel; Periodicals; 662.6
• DOI: 10.1016/j.fuel.2019.05.059
• Languages: English
• ISSNs: 0016-2361

Abstract

Highlights: A reduced model of PM index linking gasoline properties to PM emissions from was developed. Distillating parameters of T90 and T70 are crucial fuel properties affecting PM emissions. The reduced PM index was verified by mathematical analysis and comprehensive tests. Abstract: Particulate matter (PM) emissions have become an increasingly noticeable concern for gasoline vehicles, especially for gasoline direct injection (GDI) one. The PM index (PMI) model developed by Aikiwa et al. provides an effective model linking the fuel properties to the PM emissions from gasoline vehicles. In this study, a more practical reduced PM index (RPMI) was developed based on the detailed analysis of 22 fuels with various physical properties and chemical compositions. Two approaches of statistical mathematics of correlation analysis and multivariate linear regression (MLR) were adopted in the modeling process. The RPMI involving two global fuel properties of T90 (distillation temperature of 90% by volume) and T70 (distillation temperature of 70% by volume) was tested to be statistically valid. In addition, the model was verified through the engine bench tests and the vehicle emissions tests. The results reveal that the RPMI showed good correlations to the engine-out particle number (PN) emissions under all of the typical test conditions, including PFI (port fuel injection) mode, GDI (gasoline direct injection) mode and compound injection (PFI + GDI) mode. In addition, the RPMI demonstrated a significantly high correlation to vehicle-out PN emissions over the New European Driving Cycle (NEDC) with the determination coefficient R 2 = 0.947. Moreover, a moderate correlation (R2 = 0.801) of the index to the filter PM mass emissions was observed. In the light of no additional components analysis is needed except for the legitimate tests of fuels, the RPMI has the potential to be a useful tool for original equipment manufacturers (OEMs) and environmental certification departments to expediently evaluate the fuel effect on PM emissions.