The paper focuses on the development of a mesh moving method based on non-conformal topologically changing grids applied to the simulation of IC engines, where the prescribed motion of piston and valves is accomplished by rigidly translating the sub-domain representing the moving component. With respect to authors previous work, a more robust and efficient algorithm to handle the connectivity of non-conformal interfaces and a mesh-motion solver supporting multiple layer addition/removal of cells, to decouple the time-step constraints of the mesh motion and of the fluid dynamics, has been implemented as a C++ library to extend the already existing classes for dynamic mesh handling of the finite-volume, open-source CFD code OpenFOAM ® . Other new features include automatic decomposition of large multiple region domains to preserve processors load balance with topological changes for parallel computations and additional tools for automatic preprocessing and case setup. Finally, a transient solver for compressible viscous flows based on the transient SIMPLE algorithm has been implemented in order to enhance conservation of mass and energy for domains sliding over dynamically attached/detached boundaries. The advantages are significant: mesh changes in terms of topology and deformation are fully managed by the mesh motion solver without remeshing, with a consequent reduction of the overall simulation time. Most important, the method allows to preserve the quality of the mesh initially defined by the user (skewness, non-orthogonality and aspect ratio) during the whole engine cycle, favoring a faster convergence of the solver and a very accurate fluid-dynamic solution. Used in conjunction with LES turbulence modeling, the method allows to decouple mesh motion by LES filter operation, since the filter width is kept constant during the entire cycle. Validation tests have been performed on the full-cycle simulation of a Transparent Combustion Chamber (TCC) engine, whose experimental data are available through the Engine Combustion Network database (ECN). The implementation of the described methodology is absolutely general, it works on any number of processors and it can be applied to any application where moving parts and non-conformal interfaces are involved.

Swirling flows are very dominant in applied technical problems, especially in IC engines, and their prediction requires rather sophisticated modeling. An adaptive low-pass filtering procedure for the modeled turbulent length and time scales is derived and applied to Menter' original k − ω SST turbulence model. The modeled length and time scales are compared to what can potentially be resolved by the computational grid and time step. If the modeled scales are larger than the resolvable scales, the resolvable scales will replace the modeled scales in the formulation of the eddy viscosity; therefore, the filtering technique helps the turbulence model to adapt in accordance with the mesh resolution and the scales to capture. The novel turbulence model presented in this work will be called Dynamic Length Scale Resolution Model (DLRM), because of its capability to dynamically adapt its behavior according to the grid resolution and to consequently switch from modeling to resolving the turbulent length scales. Validation has been carried out both on a strongly swirling flow through a sudden expansion and on a simple IC engine geometry with one axial central valve; the model seems able to capture unsteady effects and to produce accurate time-averaged results (especially if compared to its standard RANS formulation) and looks particularly suitable when used with grids where turbulence would not be sufficiently resolved for an accurate LES.

No abstract

An International Journal published in association with ERCOFTAC• Topics cover fluid dynamics, scalar transport, multi-physics interactions and flow/turbulence control, analytical, computational and experimental methods• Emphasis on novelty, originality, timeliness, quality and thematic relevance to engineering and environmental applications

No abstract