Combustion analysis and cycle-by-cycle variations in spark ignition enginecombustion - Part 2: A new parameter for completeness of combustion and its use in modelling cycle-by-cycle variations in combustion
Jk. Ball et al., Combustion analysis and cycle-by-cycle variations in spark ignition enginecombustion - Part 2: A new parameter for completeness of combustion and its use in modelling cycle-by-cycle variations in combustion, P I MEC E D, 212(D6), 1998, pp. 507-523
Citations number
19
Categorie Soggetti
Mechanical Engineering
Journal title
PROCEEDINGS OF THE INSTITUTION OF MECHANICAL ENGINEERS PART D-JOURNAL OF AUTOMOBILE ENGINEERING
This paper investigates a technique of calculating the completeness of comb
ustion on a cycle-by-cycle basis. The technique introduces the normalized p
ressure rise due to the combustion parameter, Psi, to describe the complete
ness of combustion. This parameter is based on the Rassweiler-Withrow metho
d of calculating the mass fraction burned and is derived from the pressure-
crank angle record of the engine.
Experimental data were obtained from a Rover K4 optical access engine and a
nalysed with a combustion analysis package. A computer simulation was then
used to model the data on a cycle-by-cycle basis, both with and without the
completeness of combustion parameter. The paper discusses the conditions u
nder which it is suitable to model mean engine cycles, compared with the ne
ed to model cycle-by-cycle variability, and comments on the situations in w
hich each type of modelling would be most appropriate. The engine simulatio
n model is also used to investigate cycle-by-cycle variability of NO emissi
ons that have recently been obtained experimentally.
The successful aspects of this investigation are that the cycle-by-cycle va
riability in the completeness of combustion can be determined by use of the
parameter Psi, that the inclusion of the completeness of combustion parame
ter improves the simulation's ability to model the experimental data both i
n a statistical sense (the coefficient of variation of the indicated mean e
ffective pressure) and on a cycle-by-cycle basis and that cycle-by-cycle NO
modelling results are found to compare well with experiment.