Use of fluent for the development of a di-si engine

Abstract

The recent surge of electric vehicles has put pressure on the development and manufacture of batteries. However, batteries are still expensive, bulky and heavy, creating the need for inboard electricity generation using an internal combustion engine, usually referred as “range extender”. This paper presents the initial development of a DI-SI engine to work as range extender, focusing on the interaction between fuel spray and airflow inside the combustion chamber. To enable efficient combustion of lean and extra lean mixtures, a technique called stratified charge, is used. With direct injection spark ignition (DI-SI) engines it is important, under part load, to direct the fuel spray to the vicinities of the spark plug, enabling a fast and stable combustion of a lean mixture. A rich mixture region is created near the spark plug allowing an easy kernel formation and development. There are three types of systems for “directing” the fuel spray towards the spark plug: wall guided, air guided and spray guided. The developed design is a mixture of wall and air guided systems and the idea is to inject the spray towards the piston crown and to divert it to the spark plug location by the barrel swirl existent within the combustion chamber at this time. The system development was carried out using CFD FLUENT code. The study comprises three parts, the design of the components and its location (combustion chamber, piston crown, intake passage and injector location and aim), the air flow modeling and finally, the two phase modelling. A simple engine geometry and mesh were created in the Ansys CFD software. The air flow was considered to be transient, incompressible, Newtonian and viscous turbulent. The turbulence model used was the standard k-ε model, since it is the most common, simple and well-known model of turbulence. The spray has been simulated using the Discrete Phase Model. The Lagrangian discrete phase model in Fluent™ follows the Euler-Lagrange approach, where the fluid phase is treated as a continuum by solving the time-averaged Navier-Stokes equations, while the dispersed phase is solved by tracking a large number of particles through the calculated flow field. Preliminary results are now being obtained.