An investigation into the effects of variable valve actuation on combustion and emissions in an SI engine
The work reported in this thesis was conducted to study the effects of variable valve actuation on combustion, emissions, and fuel economy in a modern design of 4-valve per cylinder SI engine. The use of statistically-based procedures for the design of experiments allowed a limited number of tests to be used to explore a wide region of each of the experimental variables. A series of steady-flow tests was conducted to assess the effects of valve lift on flow past the valves and the nature of any in-cylinder motion generated. Results from the former were incorporated into a filling and emptying model that allowed levels of trapped residuals and pumping work to be estimated for different valve strategies. The in-cylinder motion tests explored asymmetric valve lifts, that is to say where the two valves were opened by a different amount. These results allowed a pair of response surfaces to be established to model the intensity of both axial and barrel swirl within the cylinder over the range of valve lifts. Engine tests were conducted in two parts. The first explored the effects of changes in exhaust event phasing, intake event phasing, intake event duration, and peak intake valve lift. The design of the experiment allowed linear, quadratic, and interactions between the variables to be modelled using regression analysis. Statistical analysis allowed the most influential factors (both main effects and interactions) to be identified. Contour plots of the modelled response were used to draw conclusions about the nature of the response surface and to isolate the effects of valve opening and closure angles as well as overlap. The results were correlated with those from the steady-flow tests and from the computer model. The strategy for the second phase of tests was chosen after considering the previous results. The steady-flow tests indicated that there was considerable potential for enhancing in-cylinder motion by adopting a valve deactivation strategy and combining it with a low lift of the active intake valve. The second phase investigated the use of such a technique in conjunction with large overlaps over a range of duration of the intake valve event. The results from both phases of engine tests indicated possible strategies to reduce emissions from future engines.