A study to quantitatively analyze cold start emissions for a gasoline direct injection engine




Hu, Jinghu

Journal Title

Journal ISSN

Volume Title



The gasoline direct injection (GDI) technology is a technology with which the gasoline is directly injected in the cylinder. GDI technology has been gaining popularity among vehicle manufacturers due to multiple advantages it presents compared with the port fuel injection technology, and has been widely implemented in the light-duty passenger vehicles on the US market. One weakness of the GDI engine is the excessive hydrocarbon (HC) emission during the cold start, where the engine speed, cylinder and piston top temperature and engine fuel rail pressure are all far from optimal. Given the more stringent Tier 3 HC emissions regulations enforced by United States Environmental Protection Agency and California Air Resources Board, a detailed research on the GDI engine cold start HC emissions was essential to facilitate the compliance with HC emission standards from the modern GDI engines. A novel experimental system was designed, prototyped and installed. The in-house instrumentation and control system was designed based on the National Instruments hardware and aimed to control the Ford 2.0 L GDI engine and realize the engine cold start using custom engine powertrain parameters. The novel gas collection and analysis system was designed and prototyped to allow a cycle-based emission analysis. The entire study was carried out using three steps. First, the validation experiment was conducted to validate whether the designed system hardware and software operated as desired, and to provide some basic qualitative understanding of the GDI engine cold start profiles. Second, the preliminary quantitative analysis was carried out using both gasoline and iso-pentane as fuel to further understand the contributing factors of the cold start HC emissions for GDI engines. In the final step, a parametric study, multiple parametric sweeps were carried out for various powertrain parameters to identify the quantitative effect of each parameter on the engine power output and emission performances respectively. The initial validation experiment results showed that the designed novel experimental system performed as expected, and that HC emissions actually decreased monotonically among the first five firing cycles of the cold start. The preliminary quantitative analysis revealed that for gasoline-fueled cold starts not all the injected fuel was collected in the exhaust gas. The non-collected fuel was potentially due to fuel wall wetting and piston top impingement, which could be the main reason for the HC emissions. The parametric study found that the main contributing factor of the HC emissions for the very first firing cycle was the injected fuel that did not evaporate in time for combustion but still in time for the emissions. The parametric study also found that the HC emissions increased with injected equivalence ratio. The change in fuel rail pressure had a complicated effect on the HC emissions at the first firing cycle. The increase in injection times, from 2 to 4 injections given the same amount of total injected fuel, did improve the fuel evaporation and combustion status, and led to higher power output and lower HC emissions given the same injected fuel mass. The study showed that the key to mitigate the HC emissions during the GDI engine cold start was improving the fuel evaporation and air-fuel profile, so as to minimize the fuel wall wetting and piston top impingement effect.


LCSH Subject Headings