Coupled level set volume of fluid simulations of prefilming airblast atomization with adaptive meshing

The fuel atomisation process and the resultant spray affects nearly all aspects of combustion system performance, and must be well understood to enable the design of future combustion systems. The design of a fuel injector makes both numerical and experimental testing difficult, so simplified test pieces are often used, however, this does not accurately capture atomisation mechanisms and fuel distributions. This paper presents a computational method combining a Coupled Level Set Volume of Fluid model with Adaptive Mesh Refinement. A simple prefilmer has been used to validate the method. Comparisons of the flow field and ligament length distributions show good agreement with published DNS data. The use of AMR allows a lower total cell count, and so a reduction in computational cost of over 60% compared to previously reported results for the same case has been achieved. Further work will look to apply this method to more realistic injector geometry.