The Effect of Non-Ideal Vapour-Liquid Equilibrium and Non-Ideal Liquid Diffusion on Multi-Component Droplet Evaporation for Gasoline Direct Injection Engines University of Oxford

Format:

Conference/Event

Author/Creator:

Camm, Camm, author.

Contributor:

Davy, Martin

Richardson, David

Stone, Richard

Conference Name:

SAE 2015 World Congress & Exhibition (2015-04-21 : Detroit, Michigan, United States)

Language:

English

Physical Description:

1 online resource

Place of Publication:

Warrendale, PA SAE International 2015

Summary:

AbstractA model for the evaporation of a multi-component fuel droplet is presented that takes account of temperature dependent fuel and vapour properties, evolving droplet internal temperature distribution and composition, and enhancement to heat and mass transfer due to droplet motion. The effect on the internal droplet mixing of non-ideal fluid diffusion is accounted for. Activity coefficients for vapour-liquid equilibrium and diffusion coefficients are determined using the UNIFAC method.Both well-mixed droplet evaporation (assuming infinite liquid mass diffusivity) and liquid diffusion-controlled droplet evaporation (iteratively solving the multi-component diffusion equation) have been considered. Well-mixed droplet evaporation may be applicable with slow evaporation, for example early gasoline direct injection; diffusion-controlled droplet evaporation must be considered when faster evaporation is encountered, for example when injection is later, or when the fuel mixture is non-ideal.A bi-component iso-octane/ethanol fuel and a multi-component model gasoline fuel have been simulated with a range of initial conditions representative of gasoline direct injection. The model gasoline fuel has also been tested with a range of ethanol contents between 0% and 20%, to reflect current practice and future trends. It is shown that by ignoring the non-ideal nature of the liquid diffusion of ethanol through hydrocarbons, the predicted droplet lifetime can be under-estimated or over-estimated depending on the ambient conditions and depending on the type of fuel tested. For the bi-component fuel, the maximum error found was 57% and for the model gasoline the maximum error found was 23%. This has implications for future fuel spray modelling for accurate predictions of spray lifetime and mixture preparation, and subsequently pollutant formation

Notes:

Vendor supplied data

Publisher Number:

2015-01-0924

Access Restriction:

Restricted for use by site license