Allen Michael Danis

A Probabilistic Model for the Ignition of a Monodisperse Fuel Spray

Abstract Recent experimental work in the area of spark ignition of hydrocarbon fuel sprays has demonstrated that it is impossible to define a unique minimum ignition energy. Instead there exists a range of energies over which ignition becomes more or less probable. In connection with recent experimental and modelling work studying the spark ignition of a monodisperse fuel spray, a model has been developed to predict the ignition frequency based on the variation of the spark energy and the instantaneous spark gap equivalence ratio. Random normal distributions of these two parameters were generated with a Monte Carlo routine and used in conjunction with a Characteristic Time Model for ignition. The resulting ignition frequency simulations were nearly identical to the experimentally observed values, supporting the hypothesis that the probabilistic nature of spray ignition is the result of variations in the energy levels of individual sparks and the random distribution of droplets in the fuel spray.