Gasoline Direct Injection (GDI) engines work with stratified charge at part load and burn with lean mixtures in order to save fuel, whilst at full load, the fuel and air mix homogeneously for maximum power output. The higher compression ratio and the absence of throttling are two of the most significant benefits of GDI engines. The key issues facing GDI combustion include in-cylinder mixture preparation and post-combustion soot formation. This work was intended to investigate these aspects and was undertaken on a dedicated Jaguar single-cylinder optical GDI engine with a spray-guided combustion system. The spray-guided concept does not rely as much on charge motion or piston design, and can avoid wall-wetting effects so as to reduce engine emissions. Relevant engine control hardware and data acquisition equipment were commissioned. Data/image processing software was also developed to suit the measurements. A data-processing case study with data from a small two-stroke glow ignition engine has been conducted to develop a method to combine the burn rate and heat release analyses in the study of engines with premixed charge but compression ignition. Difficulties such as unknown ignition timing and polytropic index have been addressed. Results for all operating conditions have shown good correlations between the two methods. The technique of quantitative planar laser-induced fluorescence is useful for measuring 2-D fuel distribution in GDI engines. The relevant physics and literature were reviewed in depth. A multi-component fuel was designed to give reasonable co-evaporation characteristics with tracers matching different fuel fractions. The absorption and fluorescence features of each fuel component and tracer were characterised. Optimisation of hardware and signal-to-noise ratio was performed. A recirculating loop was set up for the calibration of the technique. The technique of colour-ratio pyrometry (CRP) for estimating the temperature and loading of soot was applied on the GDI engine. Critical features of the candidate CCD colour camera including its spectral response and noise behaviours were fully studied. Validation tests with reference sources together with an error analysis suggested an accuracy of ±50K within the combustion temperature range. Engine combustion images were then taken under various operating conditions. Temperature estimates were shown to be insensitive to the concentration of soot. Simulation with a thermodynamic modelling package, ISIS, was introduced for comparison with the experimental data. With careful tuning, ISIS gave outputs comparable to the CRP and proved to be a cost-effective tool to study GDI engines. High-speed combustion imaging was carried out using a CMOS camera, allowing the study of flame properties as well as crank-angle resolved CRP. By using a lens in the piston crown to give full bore optical access and appropriate image processing, the flame front could be detected reliably throughout the main combustion process.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:442828 |
| Date | January 2006 |
| Creators | Ma, Hongrui |
| Contributors | Stone, Richard |
| Publisher | University of Oxford |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://ora.ox.ac.uk/objects/uuid:46364c15-08d3-4984-af3b-819ef0bdc847 |
Page generated in 0.0019 seconds